Context 9781591028055, 1591028051

Citation preview

Context

This page intentionally left blank

Context The Effects of Environment on Product Design and Evaluation

Edited by

Herbert L. Meiselman

An imprint of Elsevier

Woodhead Publishing is an imprint of Elsevier The Officers’ Mess Business Centre, Royston Road, Duxford, CB22 4QH, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, OX5 1GB, United Kingdom Copyright © 2019 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/ permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-12-814495-4 For information on all Woodhead publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Charlotte Cockle Acquisition Editor: Megan R. Ball Editorial Project Manager: Amy M M. Clark / Michael Lutz Production Project Manager: Joy Christel Neumarin Honest Thangiah Cover Designer: Miles Hitchen Typeset by SPi Global, India

Dedication

Dedicated to our families, friends, and colleagues who provide the social context of our lives.

This page intentionally left blank

Contents

Contributors

Section A The basics 1

2

3

The language of context research Herbert L. Meiselman 1.1 Definitions and synonyms 1.2 Frequency of context words References Further reading

xvii

1 3 3 9 12 18

People in context—The social perspective Suzanne Higgs, Helen Ruddock, Nicolas Darcel 2.1 Introduction 2.2 Studying the social context of consumer behavior 2.3 How does social context influence consumer behavior? 2.4 Competing social influences 2.5 Implications for consumer research and product development 2.6 Conclusions Acknowledgments Conflict of Interest References Further reading

19

Context effects at the level of the sip and bite Armand V. Cardello 3.1 Introduction 3.2 Objectives of the chapter 3.3 Simultaneous contextual effects in taste and smell 3.4 Contextual interactions of trigeminal (irritation) and somatosensory (texture/thermal) sensations with taste and smell 3.5 Contextual interactions of color and chemosensory perception 3.6 Contextual interactions of sound and chemosensory perception 3.7 Temporal contextual effects in chemosensory perception

39

19 20 21 29 29 31 31 31 31 38

39 41 41 46 48 49 50

viii

4

5

6

Contents

3.8 Contextual effects on hedonics 3.9 Conclusions References

51 53 54

In-home testing Elizabeth H. Zandstra, Ren
e Lion 4.1 In-home testing—Bringing lab results to the real world 4.2 Bring the world into the lab—creating standardized, real-life contexts 4.3 Conclusion References Further reading

67

Useful observational research Brian Wansink 5.1 Introduction 5.2 Generating insights for coding categories 5.3 Coding, collecting, and analyzing observational data 5.4 Using observational insights to improve the consumer experience 5.5 Practice example: Reimagineering a Disney World attraction 5.6 Conclusion References Further reading

87

Situational appropriateness in food-oriented consumer research: Concept, method, and applications Davide Giacalone 6.1 Appropriateness as a basic context construct 6.2 The “Item-by-use” (IBU) approach to measuring product appropriateness 6.3 Current directions in appropriateness research 6.4 Conclusions References

Section B Meals in context 7

Food choices in context Maartje P. Poelman, Ingrid H.M. Steenhuis 7.1 Why is the food choice context important?—A theoretical perspective 7.2 Macro context

67 79 80 81 85

87 89 97 102 105 107 107 110

111 111 114 130 135 136

141 143 143 146

Contents

8

9

10

11

ix

7.3 Local context 7.4 Social context 7.5 Setting context 7.6 Closing paragraph and future directions Acknowledgments References Further reading

148 151 153 159 159 160 168

Meal and snack: Two different contexts for foods and drinks Uyen Thuy Xuan Phan 8.1 Defining meals and snacks 8.2 The dynamics of meals and snacks 8.3 Methods for studying food choice in the context of meals and snacks 8.4 Implication for future research References

169

The meal as the proper context for food and drinks € a, € Mari Niva Johanna Makel 9.1 Defining proper meals 9.2 What elements are regarded as essential in a proper meal? 9.3 The proper context of a meal 9.4 Changing and challenging the properness of eating 9.5 Moving contexts: What will be proper in the future? References

191

The value of studying laboratory meals France Bellisle 10.1 Introduction 10.2 A brief historical perspective 10.3 The laboratory context 10.4 The demands of laboratory testing of human food consumption 10.5 Limitations of the laboratory approach 10.6 Conclusions and consideration for future developments References Further reading

209

Are family meals declining? The example of Denmark Lotte Holm, Thomas Bøker Lund 11.1 Introduction 11.2 The significance of family meals 11.3 Family meals in decline?

227

169 175 180 184 185

191 193 198 201 203 205

209 210 212 217 219 221 222 225

227 228 229

x

Contents

11.4 Social organization of family meals 11.5 Aim of analysis 11.6 The data 11.7 Results 11.8 Discussion Acknowledgments References 12

13

14

Studying natural meals: What are the benefits of the living lab approach? Anestis Dougkas, Laure Saulais, Agne`s Giboreau 12.1 Introduction 12.2 Why is it important to take context into account in food research? 12.3 Evaluating eating behaviors within a meal context 12.4 Living lab case studies 12.5 Future trends References The effects of environment on product design and evaluation: Meals in context, institutional foodservice John S.A. Edwards, Heather J. Hartwell, Sarah Price 13.1 Introduction 13.2 Purpose 13.3 What is foodservice? 13.4 What is institutional foodservice? 13.5 A brief history and development of institutional foodservice 13.6 Types and categories of institutional foodservice 13.7 Consumer attitudes and expectations of institutional foodservice, issues, and challenges 13.8 Summary and conclusions References Further reading The effect of context on children’s eating behavior Monica Laureati, Ella Pagliarini 14.1 Introduction 14.2 The meal 14.3 The physical environment 14.4 The social environment 14.5 Conclusions and future perspectives of study References

230 231 232 233 236 238 238

241 241 242 245 250 253 254

259 259 259 260 262 263 265 268 281 282 285 287 287 288 294 296 301 301

Contents

15

16

Food combinations and food and beverage combinations in meals Jacob Lahne 15.1 Introduction 15.2 Challenges in food pairing research 15.3 What is a food pairing? 15.4 Balance, harmony, complexity… pairing? 15.5 When chemistry dictates pairing 15.6 Flavor pairing theory 15.7 New and old ideas: Alternative conceptions of food pairing 15.8 Conclusion: A “manifesto” for researching food pairing References Virtual reality and immersive approaches to contextual food testing Christina Hartmann, Michael Siegrist 16.1 Introduction 16.2 A new research paradigm 16.3 Food selection behavior 16.4 Food-evoked emotions in VR 16.5 Appropriateness of contextual cues 16.6 A word on presence 16.7 Summary and future directions References

Section C Testing products in context 17

Healthcare supplements in context Carla Lynn Kuesten 17.1 Introduction 17.2 Oral nutritional supplements 17.3 Initiation, adherence/compliance, and persistence 17.4 Product factors affecting consumer usage of nutritional supplements 17.5 Choice of research design, methodology, and protocol 17.6 Industry case study applications of health care supplements product research 17.7 Product story telling—Making sense of it all 17.8 Practical considerations and future needs 17.9 Summary Acknowledgments References Further reading

xi

307 307 309 310 313 314 314 316 317 318

323 323 324 325 330 332 332 334 336

339 341 341 342 342 343 345 346 359 359 360 361 361 364

xii

18

19

20

21

Contents

Personal and home care products in context Porcherot C., Vignon-Mares M.C., Goisbault I. 18.1 Main considerations while testing personal and home care products in context 18.2 What are the implications while testing personal and home care products? 18.3 Intermediate alternatives such as evoking or mimicking/simulating contexts in CLT 18.4 Recommendations and perspectives when evoking or mimicking/simulating context in CLTs 18.5 Conclusion References Further reading

365 365 367 368 377 380 381 385

Beverages in context Sara Spinelli 19.1 Context has many meanings 19.2 Context contributes to shape expectations and responses to beverages 19.3 Beverages and situational appropriateness 19.4 Individual differences in preferred context to consume a product: A case study on coffee 19.5 Conclusions References

387

Automobiles in context Nathalie Herbeth, David Blumenthal 20.1 Introduction 20.2 Characterizing the context to design relevant cars’ evaluations: “Between-products” context 20.3 Characterizing the context to design relevant automotive systems’ evaluations: “Inside-product” context 20.4 Research perspectives References Further reading

409

The office architecture: A contextual experience with influences at the individual and group level Christina Bodin Danielsson 21.1 Introduction 21.2 The spatial context of the office at an individual level 21.3 Discussion and conclusion of the three studies 21.4 The spatial context of the office from a group and organizational perspective

387 390 396 397 402 403

409 411 420 425 426 430

431 431 433 446 448

Contents

21.5 Concluding remarks Acknowledgments References 22

23

24

25

Conducting contextualized and real-life product tests: Benefits and experimental challenges Julien Delarue, Thierry Lageat 22.1 Testing contexts and product experience 22.2 Measuring consumer responses in contextualized and real-life environments 22.3 Conclusions References Inducing context with immersive technologies in sensory consumer testing Patrick Hehn, Dariah Lutsch, Frank Pessel 23.1 Fundamentals of immersive technologies 23.2 Immersive applications 23.3 Benefits and restrictions of testing in context induced by immersive media—Learnings from different case studies 23.4 Conclusion and outlook Appendix References Contextual product testing for small to medium sized enterprises (SMEs) Rebecca N. Bleibaum, Martin J. Kern, Heather Thomas 24.1 Introduction 24.2 Body 24.3 The test environment References Further reading Assessment of the comfort of workwear for the food industry Edith Classen 25.1 Introduction 25.2 Workwear for the food industry 25.3 Antimicrobial functionalization of textile materials 25.4 Comfort of clothing 25.5 Testing of wear comfort 25.6 Quality of workwear

xiii

450 451 451

457 457 460 469 471

475 475 479 480 496 497 498

501 501 502 503 519 519 521 521 522 524 527 529 538

xiv

Contents

25.7 Conclusion References Further reading

Section D Other contextual variables 26

27

28

Evoked consumption context matters in food-related consumer affective research Betina Piqueras-Fiszman, Sara R. Jaeger 26.1 Introduction 26.2 The role of consumption context in emotion research in sensory and consumer science 26.3 Using sensory imagery to evoke a context and exploring its impact in potential satisfaction, attitudes, and meal choices 26.4 General discussion 26.5 Conclusion Acknowledgments References

540 540 542

543 545 545 546 550 558 560 560 560

Packaging in context € Boya € O. Lawrence L. Garber, Jr., Eva M. Hyatt, Unal 27.1 Packaging in context 27.2 The relativity of visual phenomena 27.3 Packaging and brand performance 27.4 Creating a new package 27.5 Means of testing packages in context 27.6 In-store testing 27.7 A/B testing 27.8 Asking questions 27.9 Consumer neuroscience/neuromarketing 27.10 Conclusion References Further reading

565

From photographs to real stores: Context squared Katelijn Quartier, Jan Vanrie 28.1 The design of retail spaces 28.2 Research on lighting in retail 28.3 Lighting supermarkets: From photographs to real stores 28.4 General discussion References

585

565 565 568 569 576 576 578 579 580 581 581 584

585 586 587 600 602

Contents

29

30

Alcoholic beverages in context Susan E.P. Bastian, Lukas Danner, Jun Niimi, Renata Ristic, Trent E. Johnson 29.1 Introduction 29.2 Contextual effects on alcoholic beverages 29.3 Research and industry practical perspectives 29.4 Conclusions and future directions References

605

Learning from the real world—Creating relevant research designs Jacqueline H. Beckley 30.1 Observe the real world 30.2 Hypothesis setting to guide the design process 30.3 Putting the pieces together Acknowledgments References Further reading

631

Section E Summary 31

xv

Summary Herbert L. Meiselman References

Index

605 608 621 624 624

634 640 647 649 649 651

653 655 668 669

This page intentionally left blank

Contributors

Susan E.P. Bastian School of Agriculture, Food and Wine, Waite Research Institute, The University of Adelaide (UA), Glen Osmond, SA, Australia Jacqueline H. Beckley The Understanding & Insight Group LLC, Denville, NJ, United States France Bellisle Nutritional Epidemiology Research Team (EREN), Paris 13 University, INSERM (U1153), INRA (U1125), CNAM, Bobigny, France Rebecca N. Bleibaum Dragonfly SCI, Inc., Santa Rosa, CA, United States David Blumenthal UMR Ing
enierie Proc
ed
es Aliments, AgroParisTech, Inra, Universit
e Paris-Saclay, Massy, France Christina Bodin Danielsson The School of Architecture, The Royal Institute of Technology (KTH), Stockholm, Sweden € Boya University of North Carolina at Chapel Hill, Appalachian State € Unal O. University, Boone, NC, United States Armand V. Cardello U.S. Army Natick RD&E Center, Natick, MA, United States Edith Classen Hohenstein, Boennigheim, Germany Lukas Danner School of Agriculture, Food and Wine, Waite Research Institute, The University of Adelaide (UA), Glen Osmond, SA, Australia Nicolas Darcel AgroParisTech, Paris, France Julien Delarue AgroParisTech, INRA, Universit
e Paris-Saclay, Massy, France Anestis Dougkas Institut Paul Bocuse Research Centre, Ecully, France John S.A. Edwards Faculty of Management, Bournemouth University, Dorset, United Kingdom Lawrence L. Garber, Jr. University of North Carolina at Chapel Hill, Elon University, Elon, NC, United States

xviii

Contributors

Davide Giacalone SDU Innovation & Design Engineering, Department of Technology and Innovation, Faculty of Engineering, University of Southern Denmark, Odense, Denmark Agne`s Giboreau Institut Paul Bocuse Research Centre, Ecully, France I. Goisbault Strategir, Bordeaux, France Christina Hartmann Department of Health Science and Technology, Consumer Behavior, ETH Zurich, Z€ urich, Switzerland Heather J. Hartwell Faculty of Management, Bournemouth University, Dorset, United Kingdom Patrick Hehn Marketing and Consumer Psychology, Harz University of Applied Sciences, Wernigerode, Germany Nathalie Herbeth Groupe Renault, Guyancourt, France Suzanne Higgs School of Psychology, University of Birmingham, Birmingham, United Kingdom; AgroParisTech, Paris, France Lotte Holm Department of Food and Resource Economics (IFRO), University of Copenhagen, Frederiksberg, Denmark Eva M. Hyatt University of South Carolina, Appalachian State University, Boone, NC, United States Sara R. Jaeger The New Zealand Institute for Plant and Food Research Ltd., Auckland, New Zealand Trent E. Johnson School of Agriculture, Food and Wine, Waite Research Institute, The University of Adelaide (UA), Glen Osmond, SA, Australia Martin J. Kern SAM, Sensory and Marketing International, Munich, Germany Carla Lynn Kuesten Consumer Product Research, Amway, Ada, MI, United States Thierry Lageat EUROSYN, Villebon-sur-Yvette, France Jacob Lahne Department of Food Science & Technology, Virginia Tech, Blacksburg, VA, United States Monica Laureati Department of Food, Environmental and Nutritional Sciences (DeFENS), University of Milan, Milan, Italy

Contributors

xix

Ren
e Lion Consumer Science, Unilever R&D Vlaardingen, Vlaardingen, The Netherlands Thomas Bøker Lund Department of Food and Resource Economics (IFRO), University of Copenhagen, Frederiksberg, Denmark Dariah Lutsch Sensory & Consumer Insights, Symrise AG, Holzminden, Germany Johanna M€ akel€ a Faculty of Educational Sciences, University of Helsinki, Helsinki, Finland Herbert L. Meiselman Herb Meiselman Training and Consulting, Rockport, MA, United States Jun Niimi School of Agriculture, Food and Wine, Waite Research Institute, The University of Adelaide (UA), Glen Osmond, SA, Australia Mari Niva Faculty of Educational Sciences, University of Helsinki, Helsinki, Finland Ella Pagliarini Department of Food, Environmental and Nutritional Sciences (DeFENS), University of Milan, Milan, Italy Frank Pessel Fragrance Development, Henkel AG & Co. KGaA, D€usseldorf, Germany Uyen Thuy Xuan Phan Institute of Biotechnology and Food Technology, Industrial University of Ho Chi Minh City, Ho Chi Minh City, Vietnam Betina Piqueras-Fiszman Wageningen University, Wageningen, The Netherlands Maartje P. Poelman Utrecht University, Utrecht, The Netherlands C. Porcherot Firmenich SA, Geneva, Switzerland Sarah Price Faculty of Management, Bournemouth University, Dorset, United Kingdom Katelijn Quartier Hasselt University, Faculty of Architecture and Arts, Diepenbeek, Belgium Renata Ristic School of Agriculture, Food and Wine, Waite Research Institute, The University of Adelaide (UA), Glen Osmond, SA, Australia Helen Ruddock School of Psychology, University of Birmingham, Birmingham, United Kingdom

xx

Contributors

Laure Saulais Institut Paul Bocuse Research Centre, Ecully, France; Department of Agri-food Economics and Consumer Science, Laval University, Quebec, QC, Canada Michael Siegrist Department of Health Science and Technology, Consumer Behavior, ETH Zurich, Z€ urich, Switzerland Sara Spinelli DAGRI, University of Florence, Florence, Italy Ingrid H.M. Steenhuis VU University Amsterdam, Amsterdam, The Netherlands Heather Thomas Dragonfly SCI, Inc., Santa Rosa, CA, United States Jan Vanrie Hasselt University, Faculty of Architecture and Arts, Diepenbeek, Belgium M.C. Vignon-Mares Firmenich SA, Geneva, Switzerland Brian Wansink Cornell University, Ithaca, NY, United States Elizabeth H. Zandstra Consumer Science, Unilever R&D Vlaardingen, Vlaardingen; Division of Human Nutrition and Health, Wageningen University, Wageningen, The Netherlands

Section A The basics

This page intentionally left blank

The language of context research Herbert L. Meiselman Herb Meiselman Training and Consulting, Rockport, MA, United States

1

In order to introduce the topic of context, this chapter will review words that are used in context research and writing, with the goal of summarizing how terms have been used. Words relating to context have been used inconsistently.

1.1

Definitions and synonyms

This review of terms begins with online dictionary definitions, followed by synonyms (i.e., words with the same meaning) if available, and then any additional considerations required in the discussion of context. Words appearing in the titles of papers and synonyms that are found in this list of words to be defined are underlined, to ease the task of seeing similarities. When no synonyms are listed, that means that none were listed on the specific website. The definitions and synonyms that follow are given in English (see Table 1.1); please note that a table translating these words (context, situation, etc.) into other languages is presented in Table 1.2. In some cases, a word has more than one definition or meaning; in those cases the definition closest to the meaning of setting is used, and sometimes, several meanings related to setting (e.g., see “natural,” as follows) are included. When multiple definitions are given, they are usually given in the order of most common usage to least common; for example, definition #1 is the most common definition, followed by definition #2. The preceding words are translated into a number of languages in Table 1.2. From the preceding tables, the reader can see that these words can be useful in contextual research and discussion, but also that these words might be used inappropriately when referring to context. Thus, the words context, situation, environment, setting, and location all refer to “circumstances, conditions, surroundings, factors, state of affairs, situation, environment, milieu, setting, background, backdrop, scene, climate, atmosphere, ambience,” (from Oxford dictionaries.com). This is unfortunate, because it means that we are not limited to one or two words for context—we have at least these 5 words, and from the lists of synonyms presented herein, we probably have more words. When one author refers to context, does he/ she mean the same thing as another author referring to situation or location? The situation is even more complicated when one considers the words natural, naturalistic, real, and real life. Natural means that something exists in nature. Naturalistic means something “derived from or closely imitating nature” or “looking like what appears in nature.” Thus, natural means something in nature, but naturalistic means Context. https://doi.org/10.1016/B978-0-12-814495-4.00001-5 Copyright © 2019 Elsevier Inc. All rights reserved.

4

Table 1.1 Online dictionary definitions of context words, followed by synonyms of context words (words with the same meaning) (A)

Context

Situation

Environment

Setting

Location

Oxford dictionaries.com

Merriam-Webster.com

The circumstances that form the setting for an event, statement, or idea, and in terms of which it can be fully understood. SYNONYMS: circumstances, conditions, surroundings, factors, state of affairs, situation, environment, milieu, setting, background, backdrop, scene, climate, atmosphere, ambience, mood, feel 1. A set of circumstances in which one finds oneself; a state of affairs. SYNONYMS: circumstances, set of circumstances, state of affairs, affairs, state, condition, case

2. the interrelated conditions in which something exists or occurs SYNONYMS: environment, setting

2. The location and surroundings of a place. SYNONYMS: location, place, position, spot, site, locality, locale 1. The surroundings or conditions in which a person, animal, or plant lives or operates. SYNONYMS: habitat, territory, domain, home, abode, surroundings, conditions, environs, circumstance 1. The place or type of surroundings where something is positioned or where an event takes place. SYNONYMS: surroundings, position, situation, environment, background, backdrop, milieu, environs

1a: the way in which something is placed in relation to its surroundings

1: the manner, position, or direction in which something is set 3a: the time, place, and circumstances in which something occurs or develops SYNONYMS: ambient, atmosphere, climate, clime, context, contexture, environs, medium, milieu, mise-en-sce`ne, environment, surround, surroundings, terrain 1a: a position or site occupied or available for occupancy or marked by some distinguishing feature SYNONYMS: emplacement, locale, locality, place, locus, point, position, site, spot, venue, where.

Context

1. A particular place or position. SYNONYMS: position, place, situation, site, locality, locale, spot, whereabouts, point, placement An actual place or natural setting in which a film or broadcast is made, as distinct from a simulation in a studio

1a: the way in which something is placed in relation to its surroundings

(B) 1. Existing in or derived from nature; not made or caused by humankind

Naturalistic

1. Derived from or closely imitating real life or nature. 2. Based on the theory of naturalism in art or literature

Real

1. Actually existing as a thing or occurring in fact; not imagined or supposed. SYNONYMS: actual, existent, non-fictional, non-fictitious, factual 2. (of a thing) not imitation or artificial; genuine. SYNONYMS: genuine, authentic, bona fide, pukka

Real life

13a: closely resembling an original SYNONYMS: born, congenital of, characterized by, or according with naturalism looking like what appears in nature: not looking artificial or man-made. SYNONYMS: lifelike, living, natural (also naturalist), near, photo-realistic, realistic, three-dimensional 1. of or relating to fixed, permanent, or immovable things 2a: not artificial, fraudulent, or illusory: genuine; also: being precisely what the name implies b (1): occurring or existing in actuality (2): of or relating to practical or everyday concerns or activities (3): existing as a physical entity and having properties that deviate from an ideal, law, or standard SYNONYMS: bona fide, certifiable, certified, dinkum [Australian & New Zealand], echt, genuine, honest, pukka (also pucka), authentic, right, sure-enough, true Real-life. Merriam-Webster.com (note spelling) existing or occurring in reality: drawn from or drawing on actual events or situations

5

Real life. Oxford dictionaries.com (note spelling) Life as it is lived in reality, involving unwelcome as well as welcome experiences, as distinct from a fictional world. SYNONYMS: true to life, lifelike, true, truthful, faithful, reallife, close, naturalistic, authentic, genuine, representational, graphic, convincing

2a: being in accordance with or determined by nature Also: existing in nature and not made or caused by people: coming from nature b: having or constituting a classification based on features existing in nature 8a: occurring in conformity with the ordinary course of nature 12a: having a physical or real existence as contrasted with one that is spiritual, intellectual, or fictitious

The language of context research

Natural

Continued

6

Table 1.1 Continued (C) Ecological validity Appropriateness

Not listed

Not listed

The quality of being suitable or proper in the circumstances.

Immersive reality or immersive context Virtual reality

Not listed

Merriam Webster.com for appropriate: especially suitable or compatible SYNONYMS for Appropriateness: appositeness, aptness, felicitousness, felicity, fitness, fittingness, happiness, properness, propriety, rightness, seemliness, suitability, suitableness Not listed

Augmented Reality

A technology that superimposes a computer-generated image on a user’s view of the real world, thus providing a composite view

The computer-generated simulation of a three-dimensional image or environment that can be interacted with in a seemingly real or physical way by a person using special electronic equipment, such as a helmet with a screen inside or gloves fitted with sensors.

Context

an artificial environment that is experienced through sensory stimuli (such as sights and sounds) provided by a computer and in which one’s actions partially determine what happens in the environment; also: the technology used to create or access a virtual reality First Known Use: 1987 Some suggest that virtual reality is the umbrella term for all immersive methods; thus, virtual reality might be broader than immersive reality. an enhanced version of reality created by the use of technology to overlay digital information on an image of something being viewed through a device (such as a smartphone camera); also: the technology used to create augmented reality

Table 1.2 Context words translated into languages; in some cases more than one translation is given English

Spanish

Portuguese

German

French

Italian

Korean

Japanese

Mandarin

context

contexto

contexto

Kontext / Zusammenhang

contexte

contesto

文脈

场景

situation environment

situacio´n ambiente

situac¸a˜o ambiente

situation Umgebung as well: Umwelt

situation environnement

状況 環境

状况 环境

setting

ambientacio´n

cena´rio

setting

cadre / milieu

situazione ambiente; condizione ambientale ambientazione

정황 (circumstance); 맥락 상황 경

設定

设定

location

local

Ort

lieu

luogo

位置、場 所

场所

natural naturalistic

localizacio´n/ lugar/ locacio´n natural naturalista

경 (environment) This Korean word is the same for the ‘environment’ 장소

natural naturalista

naturel naturaliste

naturale naturale

자연스러운 자연의

real real life

real vida real

real vida real

reale vita reale

실 실

validez ecolo´gica

validade ecolo´gica

reel Vraie vie/ situation reelle validite ecologique

validita` ecologica

 당성

自然な 自然主義 的な 現実の 現実の生 活 生態学的 妥当性

自然的 自然主义 的 现实 现实生活

ecological validity

nat€urlich lebensnah / naturalistisch real / echt Wirkliches Leben / Realit€at €okologische Validit€at / €okologische G€ ultigkeit / €okologische Aussagekraft

生态效度

Continued

Table 1.2 Continued English

Spanish

Portuguese

German

French

Italian

Korean

Japanese

Mandarin

appropriateness

adecuacio´n

adequac¸a˜o

성

適切さ

适当性

realidad virtual inmersiva

realidade virtual imersiva

caracte`re approprie/ pertinence realite virtuelle immersive

appropriatezza

immersive virtual reality

Angemessenheit / Zweckm€assigkeit / Eignung immersive virtuelle Realit€at / lebensechte virtuelle Realit€at

realta` virtuale immersiva

몰입 가상실

沉浸式虚 拟现实

virtual reality

realidad virtual

realidade virtual

Virtuelle Realit€at

realite virtuelle

realta` virtuale

가상실

augmented reality

realidad aumentada

realidade aumentada

erweiterte Realit€at

realite augmentee

realta` aumentata

증강 실

没入型バ ーチャル リアリ ティ バーチャ ルリアリ ティ 張現実

虚拟现实 增强现实

The language of context research

9

resembling nature. A home kitchen and a restaurant are both natural contexts for food and eating; a university cafeteria is a natural context for eating, but a university laboratory is not a natural context. One could ask whether a laboratory could be decorated to be naturalistic, that is, to resemble nature. Would a laboratory ever resemble a natural context for eating?

1.2

Frequency of context words

In order to determine the words most frequently used for context and environment, the following lists the use of these terms in the titles and in the key words in the publications listed in the References at the end of this chapter (key words are required in many journals). This list of journal articles is not meant to be exhaustive of all papers on context; conference papers and book chapters are not included. The terms followed by the references (in alphabetical order) that used them as key words are as follows, with the defined words in italics (Table 1.3A): In addition to the words that were defined above, the following contextual terms were also identified in the references: meal, home, restaurant (Table 1.3B). Inspection of the preceding references shows a preference to use the words context and environment rather than the words situation, setting, or location. But it is important to note that all of these words are used somewhat interchangeably. Some words are used infrequently in the context literature, including natural, naturalistic, real, and real life. It is probably best to be very careful in using these words in writing about context. Interestingly, a number of references deal with context, even emphasize context, but do not use the word context (or the words situation, setting, location, or environment) in their titles or in their keywords (for examples, papers by Wansink, PiquerasFiszman and Spence, Blundell, and others). More recent reviews ( Jaeger, Hort, et al., 2017; Jaeger & Porcherot, 2017) use the word context with greater frequency, but older papers do not (for examples, Bell et al., 1994; Meiselman, 1992a, 1992b—use situation). Giuliani and Scopelliti (2009) have reviewed the field of environmental psychology research using a number of key variables: (1) the type of setting, (2) the type of people in the setting, and (3) methodological variables. They noted a preponderance of laboratory research, but also noted the observational study of actual behavior in the environment. The reader should note that the Journal of Environmental Psychology emphasizes the influence of the environment on people, and has a strong environmental/ecological perspective. The paper by Giuliani and Scopelliti (2009) includes an interesting presentation of the historical development of environmental psychology from ecological psychology in the 1960s, to environmental psychology beginning in the 1970s and adapting to a changing political and research environment from the 1990s to the present time. In their review of trends in sensory and consumer research, Jaeger, Hort, et al. (2017) included context as one of the key trends in the field, and concluded the following:

Context

References using the word Context as a key word or in the title deAndrade et al., 2017; Bangcuyo et al., 2015; Barbopoulos & Johansson, 2017; Blake, Bisogni, Sobal, Devine, & Jastran, 2007; Boutrolle, Delarue, Arranz, Rogeaux, & Koster, 2007; Cardello, 1995; Cliceri, Petit, Garrel, Monteleone, & Giboreau, 2018; Cohen & Babey, 2012; Danner et al., 2016; DiMonaco, Giacalone, Pepe, Masi, & Cavella, 2014; Divert, Laghmaoui, Crema, Issanchou, & Sulmont-Rosse, 2015; Edwards, Meiselman, Edwards, & Lesher, 2003; Ellison, Duff, Wang, & White, 2016; Gimenez, Gagliardi, & Ares, 2015; Go´mez-Corona, Chollet, Escalona-Buendı´a, & Valentin, 2017; Hein, Hamid, Jaeger, & Delahunty, 2012; Hersleth, Mevik, Næs, & Guinard, 2003; Hersleth, Monteleone, Segtnan, & Næs, 2015; Hersleth, Ueland, Allain, & Næs, 2005; Holm, Lauridsen, Lund, Gronow, & M€akel€a, 2016; Holthuysen, Vrijhof, de Wijk, & Kremer, 2017; Jaeger, 2006; Jaeger & Rose, 2008; Jaeger, Danaher, & Brodie, 2009; Jaeger, Marshall, & Dawson, 2009; Jaeger, Hort, et al., 2017; Jimenez et al., 2015; Kantono et al., 2018; Kim, Dessirier, van Hout, & Lee, 2015; King, Meiselman, Hottenstein, Work, & Cronk, 2007; King, Weber, Meiselman, & Lv, 2004; Liu, Petit, Brit, & Giboreau, 2019; Lock, Brindal, Hendrie, & Cox, 2016; Lusk, Hamid, Delahunty, & Jaeger, 2015; Meiselman, 2013; M€ orlein et al., 2015; Piqueras-Fiszman & Jaeger, 2014a, 2014b, 2014c, 2015; Pla-Sanjuanelo, Ferrer-Garcı´a, Gutierrez-Maldonado, Riva, & SanchezPlanell, 2015; Sester et al., 2013; Spinelli et al., 2017; Stelick & Dando, 2018; Stelick, Penano, Riak, & Dando, 2018; Weber, King, & Meiselman, 2004; Zellner, 2007 References using the word Situation as a key word or in the title Cardello, Schutz, Snow, & Lesher, 2000; Edwards et al., 2003; Go´mez-Corona et al., 2017; Hetherington, Anderson, Norton, & Newson, 2006; Jaeger & Rose, 2008; Kim, Lee, & Kim, 2016; King et al., 2007, 2004; K€ oster, 2009; Liu et al., 2019; Lock et al., 2016; Meiselman, 1992b; Petit & Sieffermann, 2007; Wansink, 2004; Wansink, Payne, & Shimizu, 2010. (Note: lots of references on situation-awareness) References using Environment as a key word or in the title Ding et al., 2012; Divert et al., 2015; Edelman, Engell, Bronstein, & Hirsch, 1986; Ellison et al., 2016; Ferrer-Garcia et al., 2017; Gardner et al., 2014; Go, Kim, & Chung, 2017; Giuliani & Scopelliti, 2009; Gorini, Griez, Petrova, & Riva, 2010; Hirsch, Kramer, & Meiselman, 2005; Kanjanakorn & Lee, 2017; Kim et al., 2016, King et al., 2007, 2004; Klesges, Bartsch, Norwood, Kautzrnan, & Haugrud, 1984; Kontukoski, Paakki, Thureson, Uimonen, & Hopia, 2016; Lock et al., 2016; Meiselman, Johnson, Reeve, & Crouch, 2000; Peneau et al., 2009; Petit & Sieffermann, 2007; PlaSanjuanelo et al., 2015; Saarela, Lapvetel€ainen, Mykk€anen, Kantanen, & Rissanen, 2013; Stelick & Dando, 2018; Wansink, 2004; Weber et al., 2004 References using the word Setting as a key word or in the title Blake et al., 2007; Fernandez, Bensafi, Rouby, & Giboreau, 2013; Garcı´a-Segovia, Harrington, & Seo, 2015; Go et al., 2017; Porcherot, Petit, Giboreau, Gaudreau, & Cayeux, 2015; Saarela et al., 2013; Thomas et al., 2017; van der Meji, Wijnhoven, Finlayson, Oosten, & Visser, 2015 References using the word Location as a key word or in the title Boutrolle, Arranz, Rogeaux, & Delarue, 2005; Edwards et al., 2003, Edwards & Hartwell, 2004; Garcı´a-Segovia et al., 2015; Jaeger, Danaher, & Brodie, 2009, 2010

10

Table 1.3A References using context words as key words in publications

The language of context research

References using the term natural as a key word or in the title Kanjanakorn & Lee, 2017; Meiselman, 1992a, 1992b References using the term naturalistic as a key word or in the title Geller, Russ, & Altomari, 1986 References using the term real as a key word or in the title Go´mez-Corona et al., 2017; Meiselman, 1992b; Quartier, Vanrie, & van Cleempoel, 2014 References using the term real life as key words or in the title Cliceri et al., 2018; Liu et al., 2019; Saarela et al., 2013; Sinesio et al., 2018 References using the term ecological validity as key words or in the title Bangcuyo et al., 2015; Porcherot et al., 2015 (ecological only); Stelick & Dando, 2018 References using the word appropriateness as a key word or in the title Cardello & Schutz, 1996; Piqueras-Fiszman & Jaeger, 2014a, 2014b, 2014c References using the term augmented reality, immersive reality, virtual reality, or immersive virtual reality as key words or in the title Bangcuyo et al., 2015; Ferrer-Garcia et al., 2017; Gorini et al., 2010; Hathaway & Simons, 2017; Liu, Hooker, Parasidis, & Simons, 2017; PlaSanjuanelo et al., 2015; Schnack, Wright, & Holdershaw, 2018; Sester et al., 2013 (immersive only); Siegrist et al., 2018; Sinesio et al., 2018; Stelick et al., 2018; Ung, Menozzi, Hartmann, & Siegrist, 2018; van Herpen, van den Broek, van Trijp, & Yu, 2016

Table 1.3B References using additional context words for location (meal, home, restaurant) as key words in publications

11

References using the word Meal as a key word or in the title Allirot et al., 2012; Brindal, Wilson, Mohr, & Wittert, 2011; DiMonaco et al., 2014; Divert et al., 2015; Hersleth et al., 2015; Hoek et al., 2013; Holm et al., 2016; Holthuysen et al., 2017; Jimenez et al., 2015; King et al., 2007; King et al., 2004; Libotte, Siegrist, & Bucher, 2014; Meiselman, 1992b; Phan & Chambers, 2018; Piqueras-Fiszman & Jaeger, 2014a, 2014b, 2014c; Piqueras-Fiszman & Jaeger, 2015; Wansink et al., 2010; Weber et al., 2004 References using the word home as a key word or in the title Boutrolle et al., 2005, 2007; Hoek et al., 2013; Kasparian, Mann, Serrano, & Farris, 2017; M€ orlein et al., 2015; M€ uller, Libuda, Diethelm, Huybrechts, & Kersting, 2013 References using the word restaurant as a key word or in the title Allirot et al., 2012; Bell, Meiselman, Pierson, & Reeve, 1994; Brindal, Wilson, Mohr, & Wittert, 2015; Cavazza, Graziani, & Guidetti, 2011; Cohen & Babey, 2012; Jaeger et al., 2010; Kasparian et al., 2017; King et al., 2007; King et al., 2004; Kontukoski et al., 2016; Thomas et al., 2017; Wansink, Van Ittersum, & Painter, 2005

12

Context

…With regard to the role of context and situation in future research and the top ranked statement for this topic was: “We must increase the number of real-life consumer studies for more ecological validity”… “. …on the topic of future sensory and consumer research in industry, the top ranked statement was: “We must better capture consumer responses in context (at the moment, in the environment, etc.).” Three broad themes emerged: increasing the ecological validity of sensory and consumer research, doing inter-disciplinary research and accounting for individual differences (perception and decision making). These coincide with the trend from past decades toward a broadening of this science domain.

Thus, this book shows the development of the field of context in sensory and consumer research, and the diversity of language used to describe different contexts.

References Allirot, X., Saulais, L., Disse, E., Roth, H., Cazal, C., & Laville, M. (2012). Validation of a buffet meal design in an experimental restaurant. Appetite, 58(2012), 889–897. Bangcuyo, R. G., Smith, K. J., Zumach, J. L., Pierce, A. M., Guttman, G. A., & Simons, C. T. (2015). The use of immersive technologies to improve consumer testing: The role of ecological validity, context and engagement in evaluating coffee. Food Quality and Preference, 41(2015), 84–95. Barbopoulos, I., & Johansson, L.-O. (2017). The consumer motivation scale: Development of a multi-dimensional and context-sensitive measure of consumption goals. Journal of Business Research, 76, 118–126. Bell, R., Meiselman, H. L., Pierson, B. J., & Reeve, W. G. (1994). Effects of adding an Italian theme to a restaurant on the perceived ethnicity, acceptability, and selection of foods. Appetite, 22, 11–24. Blake, C. E., Bisogni, C. A., Sobal, J., Devine, C. M., & Jastran, M. (2007). Classifying foods in contexts: How adults categorize foods for different eating settings. Appetite, 49(2007), 500–510. Boutrolle, I., Arranz, D., Rogeaux, M., & Delarue, J. (2005). Comparing central location test and home use test results: Application of a new criterion. Food Quality and Preference, 16(8), 704–713. Boutrolle, I., Delarue, J., Arranz, D., Rogeaux, M., & Koster, E. P. (2007). Central location test vs. home use test: Contrasting results depending on product type. Food Quality and Preference, 18, 490–499. Brindal, E., Wilson, C., Mohr, P., & Wittert, G. (2011). Short communication: Does meal duration predict amount consumed in lone diners? An evaluation of the time-extension hypothesis. Appetite, 57, 77–79. Brindal, E., Wilson, C., Mohr, P., & Wittert, G. (2015). Eating in groups: Do multiple social influences affect intake in a fat-food restaurant? Journal of Health Psychology, 20(5), 483–489. Cardello, A. V. (1995). Food quality: Relativity, context and consumer expectations. Food Quality and Preference, 6(3), 163–170. Cardello, A. V., Schutz, H., Snow, C., & Lesher, L. (2000). Predictors of food acceptance, consumption and satisfaction in specific eating situations. Food Quality and Preference, 11(3), 201–216.

The language of context research

13

Cardello, A. V., & Schutz, H. G. (1996). Food appropriateness measures as an adjunct to consumer preference/acceptability evaluation. Food Quality and Preference, 7(3–4), 239–249. Cavazza, N., Graziani, A. R., & Guidetti, M. (2011). Looking for the “right” amount to eat at the restaurant: Social influence effects when ordering. Social Influence, 6(4), 274–290. Cliceri, D., Petit, E., Garrel, C., Monteleone, E., & Giboreau, A. (2018). Short communication effect of glass shape on subjective and behavioral consumer responses in a real-life context of drinking consumption. Food Quality and Preference, 64, 187–191. Cohen, D. A., & Babey, S. H. (2012). Contextual influences on eating behaviours: Heuristic processing and dietary choices. Obesity Reviews, 13(9), 766–779. Danner, L., Ristic, R., Johnson, T. E., Meiselman, H. L., Hoek, A. C., Jeffery, D. W., et al. (2016). Context and wine quality effects on consumers’ mood, emotions, liking and willingness to pay for Australian Shiraz wines. Food Research International, 89, 254–265. deAndrade, J. C., Nalerio, E. S., Giongo, C., deBarcellos, M. D., Ares, G., & Delize, R. (2017). Consumer perception of dry-cured sheep meat products: Influence of process parameters under different evoked contexts. Meat Science, 130, 30–37. DiMonaco, R., Giacalone, D., Pepe, O., Masi, P., & Cavella, S. (2014). Effect of social interaction and meal accompaniments on acceptability of sourdough prepared croissants: An exploratory study. Food Research International, 66, 325–331. Ding, D., Sallis, J. F., Norman, G. J., Saelens, B. E., Harris, S. K., Kerr, J., et al. (2012). Community food environment, home food environment, and fruit and vegetable intake of children and adolescents. Journal of Nutrition Education and Behavior, 44(6), 634–638. Divert, C., Laghmaoui, R., Crema, C., Issanchou, S., & Sulmont-Rosse, C. (2015). Improving meal context in nursing homes. Impact of four strategies on food intake and meal pleasure. Appetite, 84, 139–147. Edelman, B., Engell, D., Bronstein, P., & Hirsch, E. (1986). Environmental effects on the intake of overweight and normal-weight men. Appetite, 7(1), 71–83. Edwards, J. S. A., & Hartwell, H. J. (2004). Brief communication: A comparison of energy intake between eating positions in a NHS hospital—A pilot study. Appetite, 43, 323–325. Edwards, J. S. A., Meiselman, H. L., Edwards, A., & Lesher, L. (2003). The influence of eating location on the acceptability of identically prepared foods. Food Quality and Preference, 14(2003), 647–652. Ellison, B., Duff, B. R. L., Wang, Z., & White, T. B. (2016). Putting the organic label in context: Examining the interactions between the organic label, product type, and retail outlet. Food Quality and Preference, 49, 140–150. Fernandez, P., Bensafi, M., Rouby, C., & Giboreau, A. (2013). Does olfactory specific satiety take place in a natural setting? Appetite, 60, 1–4. Ferrer-Garcia, M., Joana Pla-Sanjuanelo, J., Dakanalis, A., Vilalta-Abella, F., Riva, G., Fernandez-Aranda, F., et al. (2017). Eating behavior style predicts craving and anxiety experienced in food-related virtual environments by patients with eating disorders and healthy controls. Appetite, 117, 284–293. Garcı´a-Segovia, P., Harrington, R. J., & Seo, H.-S. (2015). Influences of table setting and eating location on food acceptance and intake. Food Quality and Preference, 39, 1–7. Gardner, C. D., Whitesel, L. P., Thorndike, A. N., Marrow, M. W., Otten, J. J., Foster, G. D., et al. (2014). Food-and-beverage environment and procurement policies for healthier work environments. Nutrition Reviews, 72(6), 390–410. Geller, E. S., Russ, N. W., & Altomari, M. G. (1986). Naturalistic observations of beer drinking among college students. Journal of Applied Behavior Analysis, 19(4), 391–396.

14

Context

Gimenez, A., Gagliardi, A., & Ares, G. (2015). Influence of evoked contexts on consumers’ rejection of two products: Implications for shelf life estimation. Food Research International, 76, 527–531. Giuliani, M. V., & Scopelliti, M. (2009). Empirical research in environmental psychology: Past, present, and future. Journal of Environmental Psychology, 29, 375–386. Go, J.-E., Kim, M.-R., & Chung, S.-J. (2017). Acquired (dis)liking of natural cheese in different repeated exposure environment. Food Research International, 99, 403–412 Volume. Go´mez-Corona, C., Chollet, S., Escalona-Buendı´a, H. B., & Valentin, D. (2017). Measuring the drinking experience of beer in real context situations. The impact of affects, senses, and cognition. Food Quality and Preference, 60, 113–122. Gorini, A., Griez, E., Petrova, A., & Riva, G. (2010). Assessment of the emotional responses produced by exposure to real food, virtual food and photographs of food in patients affected by eating disorders. Annals of General Psychiatry, 9, 30. Hathaway, D., & Simons, C. T. (2017). The impact of multiple immersion levels on data quality and panelist engagement for the evaluation of cookies under a preparation-based scenario. Food Quality and Preference, 57, 114–125. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2012). Effects of evoked consumption contexts on hedonic ratings: A case study with two fruit beverages. Food Quality and Preference, 26, 35–44. Hersleth, M., Mevik, B.-H., Næs, T., & Guinard, J.-X. (2003). Effect of contextual factors on liking for wine—Use of robust design methodology. Food Quality and Preference, 14, 615–622. Hersleth, M., Monteleone, E., Segtnan, A., & Næs, T. (2015). Effects of evoked meal contexts on consumers’ responses to intrinsic and extrinsic product attributes in dry-cured ham. Food Quality and Preference, 40, 191–198. Hersleth, M., Ueland, O., Allain, H., & Næs, T. (2005). Consumer acceptance of cheese, influence of different testing conditions. Food Quality and Preference, 16, 103–110. Hetherington, M. M., Anderson, A. S., Norton, G. N., & Newson, L. (2006). Situational effects on meal intake. A comparison of eating alone and eating with others. Physiology & Behavior, 88, 498–505. Hirsch, E. S., Kramer, F. M., & Meiselman, H. L. (2005). Research review effects of food attributes and feeding environment on acceptance, consumption and body weight: Lessons learned in a twenty-year program of military ration research US Army Research (Part 2). Appetite, 44, 33–45. Hoek, A. C., Elzerman, J. E., Hageman, R., Kok, F. J., Luning, P. A., & de Graaf, C. (2013). Are meat substitutes liked better over time? A repeated in-home use test with meat substitutes or meat in meals. Food Quality and Preference, 28, 253–263. Holm, L., Lauridsen, D., Lund, T. B., Gronow, J., & M€akel€a, J. (2016). Changes in the social context and conduct of eating in four Nordic countries between 1997 and 2012. Appetite, 103, 358–368. Holthuysen, N. T. E., Vrijhof, M. N., de Wijk, R. A., & Kremer, S. (2017). “Welcome on board”: Overall liking and just-about-right ratings of airplane meals in three different consumption contexts—Laboratory, re-created airplane, and actual airplane. Journal of Sensory Studies, 32(2)e12254. Jaeger, S., Piqueras-Fiszman, S., Reis, F., Cheang, S. L., Kam, K., Pineau, B., et al. (2017). Influence of evoked contexts on hedonic product discrimination and sensory characterizations using CATA questions. Food Quality and Preference, 56, 138–148. Jaeger, S., & Porcherot, C. (2017). Consumption context in consumer research: Methodological perspectives. Current Opinion in Food Science, 15, 30–37.

The language of context research

15

Jaeger, S. R. (2006). Non-sensory factors in sensory science research. Food Quality and Preference, 17(2006), 132–144. Jaeger, S. R., Danaher, P. J., & Brodie, R. J. (2009). Wine purchase decisions and consumption behaviours: Insights from a probability sample drawn in Auckland, New Zealand. Food Quality and Preference, 20, 312–319. Jaeger, S. R., Danaher, P. J., & Brodie, R. J. (2010). Short communication: Consumption decisions made in restaurants: The case of wine selection. Food Quality and Preference, 21, 439–442. Jaeger, S. R., Hort, J., Porcherot, C., Ares, G., Pecore, S., & MacFie, H. J. H. (2017). Future directions in sensory and consumer science: Four perspectives and audience voting. Food Quality and Preference, 56, 301–309. Jaeger, S. R., Marshall, D. W., & Dawson, J. (2009). A quantitative characterisation of meals and their contexts in a sample of 25 to 49-year-old Spanish people. Appetite, 52, 318–327. Jaeger, S. R., & Rose, J. M. (2008). Stated choice experimentation, contextual influences and food choice: A case study. Food Quality and Preference, 19(6), 539–564. Jimenez, M., Rodriguez, D., Greene, N., Zellner, D. A., Cardello, A. V., & Nestrud, M. (2015). Seeing a meal is not eating it: Hedonic context effects differ for visually presented and actually eaten foods. Food Quality and Preference, 41, 96–102. Kanjanakorn, A., & Lee, J. (2017). Examining emotions and comparing the EsSense Profile® and the coffee drinking experience in coffee drinkers in the natural environment. Food Quality and Preference, 56, 69–79. Kantono, K., Hamid, N., Shepherd, D., Lin, Y. H. T., Brard, C., Grazioli, G., et al. (2018). The effect of music on gelato perception in different eating contexts. Food Research International, 113, 43–56. Kasparian, M., Mann, G., Serrano, E. L., & Farris, A. R. (2017). Parenting practices toward food and children’s behavior: Eating away from home versus at home. Appetite, 114, 194–199. Kim, M. A., Dessirier, J. M., van Hout, D., & Lee, H. S. (2015). Consumer context- specific sensory acceptance tests: Effects of a cognitive warm-up on affective product discrimination. Food Quality and Preference, 41, 163–171. Kim, S.-E., Lee, S. M., & Kim, K.-O. (2016). Consumer acceptability of coffee as affected by situational conditions and involvement. Food Quality and Preference, 52, 124–132. King, S. C., Meiselman, H. L., Hottenstein, A. W., Work, T. M., & Cronk, V. (2007). The effects of contextual variables on food acceptability: A confirmatory study. Food Quality and Preference, 18, 58–65. King, S. C., Weber, A. J., Meiselman, H. L., & Lv, N. (2004). The effect of meal situation, social interaction, physical environment and choice on food acceptability. Food Quality and Preference, 15, 645–653. Klesges, R. C., Bartsch, D., Norwood, J. D., Kautzrnan, D., & Haugrud, S. (1984). The effects of selected social and environmental variables on the eating behavior of adults in the natural environment. International Journal of Eating Disorders, 3(4), 35–41. Kontukoski, M., Paakki, M., Thureson, J., Uimonen, H., & Hopia, A. (2016). Imagined salad and steak restaurants: Consumers’ colour, music and emotion associations with different dishes. International Journal of Gastronomy and Food Science, 4, 1–11. K€ oster, E. P. (2009). Diversity in the determinants of food choice. A psychological perspective. Food Quality and Preference, 20(2), 70–82. Libotte, E., Siegrist, M., & Bucher, T. (2014). The influence of plate size on meal composition. Literature review and experiment. Appetite, 82, 91–96.

16

Context

Liu, J., Petit, E., Brit, A.-C., & Giboreau, A. (2019). The impact of tablecloth on consumers’ food perception in real-life eating situation. Food Quality and Preference, 71, 168–171. Liu, R., Hooker, N. H., Parasidis, E., & Simons, C. T. (2017). A natural experiment: Using immersive technologies to study the impact of “all-natural” labeling on perceived food quality, nutritional content, and liking. Journal of Food Science, 82(3), 825–833. Lock, C., Brindal, E., Hendrie, G. A., & Cox, D. N. (2016). Contextual and environmental influences on reported dietary energy intake at evening eating occasions. Eating Behaviors, 21, 155–160. Lusk, K. A., Hamid, N., Delahunty, C. M., & Jaeger, S. (2015). Effects of an evoked refreshing consumption context on hedonic responses to apple juice measured using best-worst scaling and the 9-pt hedonic category scale. Food Quality and Preference, 43, 21–25. Meiselman, H. L. (1992a). Methodology and theory in human eating research. Appetite, 19, 49–55. Meiselman, H. L. (1992b). Obstacles to studying real people eating real meals in real situations. Appetite, 19, 84–86. Meiselman, H. L. (2013). The future in sensory/consumer research: ……….....Evolving to a better science. Food Quality and Preference, 27, 208–214. Meiselman, H. L., Johnson, J. L., Reeve, W., & Crouch, J. E. (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35(3), 231–237. M€ orlein, D., Schiermann, C., Meier-Dinkel, L., Trautmann, J., Wigger, R., Buttinger, G., et al. (2015). Effects of context and repeated exposure on food liking: The case of boar taint. Food Research International, 67, 390–399. M€ uller, K., Libuda, L., Diethelm, K., Huybrechts, I., & Kersting, M. (2013). Lunch at school, at home or elsewhere. Where do adolescents usually get it and what do they eat? Results of the HELENA study. Appetite, 71, 332–339. Peneau, S., Mekhmoukh, A., Chapelot, D., Dalix, A.-M., Airinei, G., Hercberg, S., et al. (2009). Influence of environmental factors on food intake and choice of beverage during meals in teenagers. A laboratory study. British Journal of Nutrition, 102, 1854–1859. Petit, C., & Sieffermann, J. M. (2007). Testing consumer preferences for iced-coffee: Does the drinking environment have any influence? Food Quality and Preference, 18, 161–172. Phan, U. T. X., & Chambers, E. (2018). Motivations for meal and snack times: Three approaches reveal similar constructs. Food Quality and Preference, 68, 267–275. Piqueras-Fiszman, B., & Jaeger, S. R. (2014a). The impact of evoked consumption contexts and appropriateness on emotion responses. Food Quality and Preference, 32, 277–288. Piqueras-Fiszman, B., & Jaeger, S. R. (2014b). Emotion responses under evoked consumption contexts: A focus on the consumers’ frequency of product consumption and the stability of responses. Food Quality and Preference, 35, 24–31. Piqueras-Fiszman, B., & Jaeger, S. R. (2014c). The impact of the means of context evocation on consumers’ emotion associations towards eating occasions. Food Quality and Preference, 37, 61–70. Piqueras-Fiszman, B., & Jaeger, S. R. (2015). Emotions associated to mealtimes: Memorable meals and typical evening meals. Food Research International, 76, 243–252. Pla-Sanjuanelo, J., Ferrer-Garcı´a, M., Gutierrez-Maldonado, J., Riva, G., & Sanchez-Planell, L. (2015). Identifying specific cues and contexts related to bingeing behavior for the development of effective virtual environments. Appetite, 87, 81–89. Porcherot, C., Petit, E., Giboreau, A., Gaudreau, N., & Cayeux, I. (2015). Measurement of selfreported affective feelings when an aperitif is consumed in an ecological setting. Food Quality and Preference, 39, 277–284.

The language of context research

17

Quartier, K., Vanrie, J., & van Cleempoel, K. (2014). As real as it gets: What role does lighting have on consumer’s perception of atmosphere, emotions and behaviour? Journal of Environmental Psychology, 39, 32–39. Saarela, A.-M., Lapvetel€ainen, A. T., Mykk€anen, H. M., Kantanen, T. T., & Rissanen, R. L. (2013). Real-life setting in data collection. The role of nutrition knowledge whilst selecting food products for weight management purposes in a supermarket environment. Appetite, 71, 196–208. Schnack, A., Wright, M. J., & Holdershaw, J. L. (2018). Immersive virtual reality technology in a three-dimensional virtual simulated store: Investigating telepresence and usability. Food Research International, (in press). Sester, C., Deroy, O., Sutan, A., Galia, F., Desmarchelier, J. F., Valentin, D., et al. (2013). “Having a drink in a bar”: An immersive approach to explore the effects of context on drink choice. Food Quality and Preference, 28(1), 23–31. Siegrist, M., Ung, C.-Y., Zank, M., Marinello, M., Kunz, A., Hartmann, C., et al. (2018). Consumers’ food selection behaviors in three-dimensional (3D) virtual reality. Food Research International. https://doi.org/10.1016/j.foodres.2018.02.033 (in press). Sinesio, F., Saba, A., Peparaio, M., Saggia Civitelli, E., Paoletti, F., & Moneta, E. (2018). Capturing consumer perception of vegetable freshness in a simulated real-life taste situation. Food Research International, 105, 764–771. Spinelli, S., Dinnella, C., Masi, C., Zoboli, G. P., Prescott, J., & Monteleone, E. (2017). Investigating preferred coffee consumption contexts using open-ended questions. Food Quality and Preference, 61, 63–73. Stelick, A., & Dando, R. (2018). Thinking outside the booth—The eating environment, context and ecological validity in sensory and consumer research. Current Opinion in Food Science, 21, 26–31. Stelick, A., Penano, A. G., Riak, A. C., & Dando, R. (2018). New horizons in food research: Dynamic context sensory testing – A proof of concept study bringing virtual reality to the sensory booth. Journal of Food Science, 83, 2047–2051. Thomas, J. M., Ursell, A., Robinson, E. L., Aveyard, P., Jebb, S. A., Herman, C. P., et al. (2017). Using a descriptive social norm to increase vegetable selection in workplace restaurant settings. Health Psychology, 36(11), 1026. Ung, C.-Y., Menozzi, M., Hartmann, C., & Siegrist, M. (2018). Innovations in consumer research: The virtual food buffet. Food Quality and Preference, 63, 12–17. van der Meji, B. S., Wijnhoven, H. A. H., Finlayson, G. S., Oosten, B. S., & Visser, M. (2015). Specific food preferences of older adults with a poor appetite. A forced-choice test conducted in various care settings. Appetite, 90, 168–175. van Herpen, E., van den Broek, E., van Trijp, H. C. M., & Yu, T. (2016). Can a virtual supermarket bring realism into the lab? Comparing shopping behavior using virtual and pictorial store representations to behavior in a physical store. Appetite, 107, 196–207. Wansink, B. (2004). Environmental factors that increase the food intake and consumption volume of unknowing consumers. Annual Review of Nutrition, 24, 455–479. Wansink, B., Payne, C. R., & Shimizu, M. (2010). Short communication “is this a meal or snack?” situational cues that drive perceptions. Appetite, 54, 214–216. Wansink, B., Van Ittersum, K., & Painter, J. E. (2005). How descriptive food names bias sensory perceptions in restaurants. Food Quality and Preference, 16, 393–400. Weber, A. J., King, S. C., & Meiselman, H. L. (2004). Effects of social interaction, physical environment and food choice freedom on consumption in a meal-testing environment. Appetite, 42, 115–118. Zellner, D. A. (2007). Contextual influences on liking and preference. Appetite, 49, 679–682.

18

Context

Further reading Blundell, J., de Graaf, C., Hulshof, T., Jebb, S., Livingstone, B., Lluch, A., et al. (2010). ILSI supplement: Appetite control: Methodological aspects of the evaluation of foods. Obesity Reviews, 11, 251–270. Higgs, S., & Thomas, J. (2016). Social influences on eating. Current Opinion in Behavioral Sciences, 9, 1–6. K€ oster, E. P. (2003). The psychology of food choice: Some often encountered fallacies. Food Quality and Preference, 14, 359–373. Lund, T. B., Kjærnes, U., & Holm, L. (2017). Eating out in four Nordic countries: National patterns and social stratification. Appetite, 119, 23–33. Onwezen, M. C., Reinders, M. J., van der Lans, I. A., Sijtsema, S. J., Jasiulewicz, A., Dolors Guardia, M., et al. (2012). A cross-national consumer segmentation based on food benefits: The link with consumption situations and food perceptions. Food Quality and Preference, 24, 276–286. Piqueras-Fiszman, B., Alcaide, J., Roura, E., & Spence, C. (2012). Is it the plate or is it the food? Assessing the influence of the color (black or white) and shape of the plate on the perception of the food placed on it. Food Quality and Preference, 24(1), 205–208. Piqueras-Fiszman, B., Harrar, V., Alcaide, J., & Spence, C. (2011). Does the weight of the dish influence our perception of food? Food Quality and Preference, 22(8), 753–756. Polivy, J., & Pliner, P. (2015). “She got more than me”. Social comparison and the social context of eating. Appetite, 86, 88–95. Piqueras-Fiszman, B., & Spence, C. (2015). Sensory expectations based on product- extrinsic food cues: An interdisciplinary review of the empirical evidence and theoretical accounts. Food Quality and Preference, 40, 165–179. Robinson, E., & Higgs, S. (2013). Making food choices in the presence of ‘healthy’ and ‘unhealthy’ companions. British Journal of Nutrition, 109(4), 765–771. Salmon, S. J., De Vet, E., Adriaanse, M. A., Fennis, B. M., Veltkamp, M., & De Ridder, D. T. (2015). Social proof in the supermarket: Promoting healthy choices under low self-control conditions. Food Quality and Preference, 45, 113–120. Silva, A. P., Voss, H.-P., van Zyl, H., Hogg, T., de Graaf, C., Pintado, M., et al. (2018). Temporal dominance of sensations, emotions, and temporal liking measured in a bar for two similar wines using a multi-sip approach. Journal of Sensory Studies, 33, e12459. Sommer, R., & Sommer, B. A. (1989). Social facilitation effects in coffeehouses. Environment and Behavior, 21(6), 651–666. Tuorila, H., & Monteleone, E. (2009). Sensory food science in the changing society: Opportunities, needs, and challenges. Trends in Food Science & Technology, 20(2), 54–62. Vyth, E. L., et al. (2011). Influence of placement of a nutrition logo on cafeteria menu items on lunchtime food choices at Dutch Work sites. Journal of the American Dietetic Association, 111(1), 131–136. Wansink, B., Van Ittersum, K., & Painter, J. E. (2006). Ice cream illusions: Bowls, spoons, and self-served portion sizes. American Journal of Preventive Medicine, 31, 240–243.

People in context—The social perspective

2

Suzanne Higgs*,†, Helen Ruddock*, Nicolas Darcel† *School of Psychology, University of Birmingham, Birmingham, United Kingdom, † AgroParisTech, Paris, France

2.1

Introduction

Consumer choice and product evaluation are affected by the social context in which those decisions are made. For example, social context has a profound influence on food choices and amounts eaten (Herman, Roth, & Polivy, 2003; Rozin, 1996). People make different choices when they eat in company compared with when they eat alone (Higgs & Thomas, 2016). In addition, the same food may be evaluated very differently if it is consumed with others, compared with when it is consumed on a solo dining occasion: a shared consumption experience is rated as more pleasant than a non-shared experience (Boothby, Clark, & Bargh, 2014). Here, we define social context as the people who may be present when a decision is made, as well as knowledge of the behaviors and evaluations of others. There is evidence that mere knowledge of the food selections made by others in a similar context affects consumer behavior, even if those other people are not present at the time of choosing (e.g., Robinson, Benwell, & Higgs, 2013). In addition, many choices are made jointly by romantic partners or family members, or influenced indirectly by knowledge about the preferences of close others (Cavanaugh, 2016). Hence, the social context of a consumer includes the people who may be present when choices are made, but also our understanding of the choices and preferences of socially connected others, social norms, and aspects of the situation that infer such norms. Despite the wealth of evidence that has accumulated over many years on social influences on consumer behaviors, social context remains a neglected factor in product development and evaluation, particularly in relation to the development of new food products (K€ oster & Mojet, 2018; Meiselman, 2013). The aim of this chapter is to provide a brief overview of the methods that have been used to study consumer behavior in social contexts, and the ways in which social context influences consumer behavior. We will also discuss how social context might be better incorporated into consumer research. The emphasis will be on research that has examined social context and food-related decision making, but where appropriate, reference will be made to other consumer experiences.

Context. https://doi.org/10.1016/B978-0-12-814495-4.00002-7 Copyright © 2019 Elsevier Inc. All rights reserved.

20

2.2

Context

Studying the social context of consumer behavior

Research examining how social context affects consumer behavior has been conducted in laboratory settings, as well as home and field settings, and has used both quantitative and qualitative methods. The literature examining social influences on eating therefore provides a good example of how data collected using a range of techniques has converged to provide strong support for the validity of the findings. Much of the research into social context and eating has examined the extent to which food intake varies in accordance with the number, or type, of people present at a meal. To do so, diary studies (e.g., de Castro & Brewer, 1992) and observational techniques (e.g., Klesges, Bartsch, Norwood, Kautzrnan, & Haugrud, 1984) have been used to identify associations between eating behavior and social context in usual eating situations. Subsequent experimental, laboratory-based studies have examined the causal relationship between social context and eating behavior by systematically manipulating the number/type of people present at that eating occasion (e.g., Clendenen, Herman, & Polivy, 1994). Other lab-based studies have focused on how people adapt their eating behavior in response to the behavior of others (e.g., Goldman, Herman, & Polivy, 1991; Prinsen, de Ridder, & de Vet, 2013). Observational investigations of dyadic eating interactions assess the degree to which eating partners match their intake. However, such investigations are theoretically problematic, as modeling of intake between non-randomly assigned dyads may occur because of pre-existing similarities between partners. It is also important to note that data from this type of study should be analyzed using multilevel modeling, due to the non-independence of the data (e.g., Salvy, Vartanian, Coelho, Jarrin, & Pliner, 2008). An alternative approach is to fix the intake of one eating partner by employing a confederate (usually another researcher) who pretends to be a participant. In these studies, the confederate always eats a set amount of food, so that the extent to which the real participant matches this amount can be assessed. A variant on this design is to use a ‘remote confederate,’ in which the participant is exposed to fictional accounts of the amount of food consumed by previous study participants, rather than a person playing the role of a participant (Feeney, Polivy, Pliner, & Sullivan, 2011; Pliner & Mann, 2004; Roth, Herman, Polivy, & Pliner, 2001). Such remote confederate designs allow researchers to examine the extent to which matching of intake occurs in the absence of other social influences (e.g., attempts to bond with the confederate). Studies using the “live” and “remote” confederate design have generally yielded similar results (Feeney et al., 2011). It is important to bear in mind that participants’ behavior may be affected by their knowledge of the fact that they are taking part in an experiment. It is possible that some participants eat differently in company, not because there is a real influence of the dining partner on consumption, but because they think that is what is expected of them in the experiment. To minimize the influence of such demand characteristics, experimenters should provide a convincing cover story to reduce awareness of the aims of the study, and to distract the participants from the experimenter’s interest

People in context—The social perspective

21

in social eating behavior. Participants may be either provided with a non-food related cover story for the experiment (e.g., Kaisari & Higgs, 2015), or told that they are participating in a taste-test study, in which the interest is in their evaluations of the food, rather than amounts consumed (Goldman et al., 1991). The extent to which participants are aware of the experiment aims can be assessed during the post-experiment debrief, by asking participants to write down their thoughts on the aims of the study. Laboratory assessment of social influences on eating has yielded similar findings to those obtained within field settings, including work in lunchroom settings, canteens, restaurants, and supermarkets (e.g., Mollen, Rimal, Ruiter, & Kok, 2013; Gierl & Huettl, 2010; Salmon et al., 2015; Thomas et al., 2017). Evidence for the importance of the social context of eating is also derived from large-scale surveys of self-reported food intake (e.g., Pedersen, Grønhøj, & Thøgersen, 2015; Pelletier, Graham, & Laska, 2014), and social network analysis (e.g., Haye, Robins, Mohr, & Wilson, 2013; Pachucki, Jacques, & Christakis, 2011). In the latter technique, participants are asked to nominate others with whom they are socially connected, and associations between these networks and dietary patterns are analyzed. Data from these studies suggest that people’s eating choices are influenced by the eating choices of those with whom they are socially connected. This pattern of results reflects findings obtained from qualitative research based on interviews exploring people’s eating experiences (Kristensen, Holm, Raben, & Astrup, 2002) and meal time interactions (Laurier & Wiggins, 2011).

2.3

How does social context influence consumer behavior?

The literature documenting the effects of social context on eating behaviors has been divided into three broad areas: (1) social facilitation, (2) modeling, and (3) impression management. These phenomena have been well described for eating behaviors, but similar effects are observed for social contextual effects on other consumer behaviors, including shopping and responses to artwork, ads, and TV programs (e.g., Geller, Russ, & Altomari, 1986; Sommer, Wynes, & Brinkley, 1992).

2.3.1 Social facilitation Social facilitation is the term used to describe the finding that the mere presence of other people enhances the predominant behavioral responses in that situation. Social facilitation of eating was first described in detail by John de Castro, who conducted a series of diary studies in which participants were asked to record what and how much they ate over 7 days, alongside information about where and with whom they ate. Data from these studies revealed that people ate much more food when they ate in company than when they ate alone (de Castro & Brewer, 1992; de Castro & de Castro, 1989). These findings were observed for meals consumed during weekends and weekdays, thus ruling out the possibility that the social facilitation of eating reflects an artifact

22

Context

that arises because people eat more, and are more likely to eat with others, during weekends (de Castro, 1991). Social facilitation has been observed consistently across different meal types, including breakfast, snacks, meals eaten at home, and meals eaten without alcohol (de Castro, 1991). Further analyses by de Castro also suggested that the amounts eaten increases as the number of diners increases, such that groups of twelve consumed, on average, 60% more than did groups of two. Indeed, de Castro concluded that social facilitation was the single most powerful influence on eating, and that “the number of people eating with the subject …is the best predictor of how much food an individual will consume” (Redd & de Castro, 1992). The conclusions based on these diary studies have been corroborated by results obtained from studies examining social facilitation within laboratory and field settings. For example, Berry, Beatty, and Klesges (1985) found that participants ate much more ice cream in 3- or 4-person groups than when alone. Similarly, Klesges et al. (1984) found that people dining out in a restaurant ate more in groups than when eating alone. The weight of evidence from numerous studies employing different methodological approaches supports the suggestion that social facilitation of eating is a real phenomenon (see Herman, 2015 for a review). There are some factors that moderate the extent to which social facilitation of eating is observed. Social facilitation of eating is more likely to occur when friends eat together than when strangers dine in a group (de Castro, 1994). In fact, when eating with strangers, people may eat less than they would if they were eating alone, possibly because they feel self-conscious about their choices (e.g., Hetherington, Anderson, Norton, & Newson, 2006; Peneau et al., 2009). In this situation, impression management concerns may override any effect of social facilitation: intake may be supressed to avoid appearing “greedy.” Similarly, people with obesity have been observed to eat less in a group than when dining alone, and it has been proposed that this is due to concerns about the stigma associated with appearing to eat excessively (Krantz, 1979). Recently, it has also been observed that people with higher BMI were more likely to consume high-energy snacks when alone, and were more likely to consume low-energy snacks in the presence of others eating (Sch€uz, Revell, Hills, Sch€uz, & Ferguson, 2017). The gender composition of a group can also moderate social facilitation effects. Specifically, Brindal, Wilson, Mohr, and Wittert (2015) reported that males eating in mixed-sex larger groups ate more than those eating in mixed- or same-sex pairs (reflective of social facilitation). Conversely, females eating in mixed-sex larger groups did not eat more than those eating in pairs, and ate significantly less than those eating in same-sex larger groups. These findings have been attributed to concerns about the image portrayed to others, such that when in mixed-sexed groups, women may eat less in order to convey a feminine impression (Brindal et al., 2015; Pliner & Chaiken, 1990). Whether people eat more in very large groups (e.g., in a crowd) has not been thoroughly investigated. The results of a recent series of studies suggest that eating in a crowded environment is associated with increased intake (Hock & Bagchi, 2017), although other work suggests that eating in a very large group of more than 50 people does not facilitate intake (Hirsch & Kramer, 1993). Further work is required to assess

People in context—The social perspective

23

the influence of eating in a crowd on intake, and to determine the limits of social facilitation in terms of group size. Several explanations have been forwarded to explain social facilitation of eating (Herman, 2015). One theory is that social meals last longer than do solo meals, due to social interaction, thus extending the opportunity for eating (de Castro, 1990). However, exactly why lingering over a meal causes people to eat more is unclear. It has been suggested that social interaction may distract people from monitoring how much they are eating, or their awareness of internal cues that might inhibit eating (e.g., fullness). Additionally, seeing others eating may automatically trigger eating based on the conditioning of appetite to the social context (Sch€uz, Bower, & Ferguson, 2015). In this manner, social facilitation of eating may become habitual, or part of the ritual of commensal meals. Another possibility is that meals eaten alone are smaller than social meals because eating alone is not as enjoyable as eating with company. However, there is only indirect evidence in support of this assumption. de Castro (1990) found that people were generally happier when eating with others than when eating alone, but his analysis found that mood and the number of people present contributed independently to variance in intake. In a more recent study, Boothby et al. (2014) reported that participants’ ‘liking’ evaluations of a good-tasting chocolate were higher when in the presence of a co-eater. However, the researchers did not assess participants’ food intake. Overall, there is some evidence that social meals may be larger because eating with others is more enjoyable, and the presence of others may disrupt usual processes associated with satiety. However, there has yet to be a systematic investigation of the effects of social context on these aspects of appetite while controlling for the amount consumed. Importantly, explanations for social facilitation of eating fail to address the fact that, in order to eat more during a social meal, more food must be available. It is possible that both social and lone eaters serve the same amount of food, but that lone eaters do not finish all of their portion. This seems unlikely because recent evidence suggests that people tend to serve themselves the amount of food they believe will make them feel comfortably full, and then eat all of that portion (Brunstrom & Shakeshaft, 2009). In other words, decisions about portion size are made before eating in the pre-meal planning stage (e.g., Fay et al., 2011). Therefore, one explanation of social facilitation of eating is that people plan to provide more food when they know they will be eating socially (Herman, 2015). For example, people might cook or order larger portion sizes (or a greater number of dishes), per person, for meals with others versus meals eaten alone. In support of this idea, Cavazza, Graziani, and Guidetti (2011) reported that the number of dishes ordered per person, within a restaurant setting, increased as a direct function of group size. Although we know that social factors influence people’s food intake, very little is known about the relationship between social eating and obesity. There is evidence that obesity spreads via social networks (Christakis & Fowler, 2007), and one plausible underlying mechanism is that eating in social groups promotes food intake. However, as most direct evidence for social facilitation of eating comes from laboratory studies, in which intake is measured at one eating occasion, it remains unclear whether social

24

Context

eating leads to cumulative increases in energy intake, and ultimately weight gain. There is evidence to suggest that the effect of merely providing large portion sizes to individuals, which increases food intake, is not compensated for by eating less at subsequent meals (Rolls, Roe, & Meengs, 2007). Hence, it is possible that socially-induced increases in food intake might not be offset by subsequent reductions in intake. In support of this suggestion, Hirsch and Kramer (1993) found that total daily caloric intake of soldiers increased as a function of the number of meals eaten socially. The phenomenon of social facilitation has been observed for other consumer behaviors, including alcohol and coffee drinking (Geller et al., 1986; Sommer & Sommer, 1989) and shopping (Sommer et al., 1992). People have also been shown to attend more to visual marketing stimuli when they are viewed in the presence of others, versus when they are viewed alone (Pozharliev, Verbeke, Van Strien, & Bagozzi, 2015). Taken together, these data suggest that social facilitation of consumer behaviors is a robust phenomenon, and that lone consumption experiences are not the same as consumption experiences shared with others.

2.3.2 Modeling Modeling refers to the tendency to adapt one’s behavior to conform to that of other people, or to what is thought to be ‘normal’ in that environment. In the case of eating behavior, this means that decisions about how much, and what, to eat are influenced by people’s perceptions of the choices of others. The eating behavior of others provides a norm of appropriate intake in that context (Vartanian, Sokol, Herman, & Polivy, 2013). Modeling occurs when the appropriate behavior is set by another present person (i.e. another diner), but appropriate behavior may also be communicated by environmental cues (e.g., portion sizes), or by the transmission of information about how other people behave. People may also model culturally agreed upon norms, such as cuisine rules. Indeed, research suggests that people tend to eat more when in the presence of someone who is eating a large amount, and less when with someone eating a small amount, compared with when they are eating alone. This is true for both adults and children (e.g., Bevelander, Ansch€ utz, & Engels, 2012; Robinson et al., 2013; Salvy et al., 2008). There is also evidence from lab-based studies that people model the food choices of others (Prinsen et al., 2013; Robinson & Higgs, 2013). Being accompanied in a cafeteria by others who choose dessert has been reported to increase the likelihood of choosing a dessert (Guarino, Fridrich, & Sitton, 1995), and providing information about the most popular dish choices in a restaurant considerably increases demand for those dishes (Cai, Chen, & Fang, 2009). Two systematic reviews of a large collection of literature have provided evidence that modeling of eating behavior is an extremely robust phenomenon that occurs regardless of current hunger state, dieting status, current health goals, age, or familiarity with the model (Cruwys, Bevelander, & Hermans, 2015; Vartanian, Spanos, Herman, & Polivy, 2015). A few factors have been shown to moderate the extent to which modeling is observed. In particular, people are more likely to model the behavior of others who

People in context—The social perspective

25

are perceived as similar in some way, and are less likely to model the behavior of a social “out-group” (Cruwys et al., 2012; Stok, Ridder, Vet, & Wit, 2014). McFerren and colleagues reported that non-overweight participants were less influenced by the choices of an overweight confederate at a buffet than a lean model (McFerran, Dahl, Fitzsimons, & Morales, 2010). Similarly, lean (but not underweight) participants in another study did not model an obviously underweight person, possibly because they did not identify with the model (Hermans, Larsen, Herman, & Engels, 2008). These findings suggest that people are more likely to model others of a similar body size. In addition, the tendency to eat more when served a large portion size (the portion size effect) is reduced when participants are told that the size of the portion served is based on the behavior of a less relevant social group (Versluis & Papies, 2016). On the other hand, eating norms may exert a stronger influence on behavior if there is a strong motivation to associate oneself with the referent group. Indeed, Guendelman and colleagues reported that a desire to prove they belong in America motivated U.S. immigrants to consume more typically American food (e.g., fast food) (Guendelman, Cheryan, & Monin, 2011). Similar motivations underlie, at least in part, the persuasiveness of celebrity endorsements of products (Hoffman & Tan, 2015). There is some evidence that the tendency to model is reduced if we are already quite certain about how to behave in a particular context (Leone, Pliner, & Herman, 2007). For example, modeling is attenuated for meals such as breakfast and lunch, for which people may have clear expectations about how much one should eat (Hermans, Herman, Larsen, & Engels, 2010). On the other hand, people are more likely to model others’ choices under conditions in which they are uncertain about their own preferences (Huh, Vosgerau, & Morewedge, 2014). Interestingly, the choices that participants make in a group setting in a restaurant appear to be dynamically influenced by competing influences. People may be less likely to conform to the norm when the unanimity of others’ choices is low, and more likely to conform if a dominant preference of the group emerges (Quester & Steyer, 2009). However, reactance may be observed when preference is very strong around a given option leading to divergence from the norm (Quester & Steyer, 2009). One reason why people model the eating behavior of others is because doing so enhances feelings of social connectedness. Humans are social creatures with a strong desire to be liked (Baumeister & Leary, 1995), and this goal of affiliation may be achieved through modeling. When we imitate the behavior of another person, this has the effect of increasing the sense of rapport we have with that person (Chartrand & Bargh, 1999). The idea that people model eating behavior to affiliate is supported by findings that modeling is reduced in circumstances in which participants feel socially accepted (Hermans, Engels, Larsen, & Herman, 2009; Robinson, Tobias, Shaw, Freeman, & Higgs, 2011), and is enhanced for individuals who are low in self-esteem (Robinson et al., 2011). Interestingly, there is evidence that the feelings of rapport and liking generated by mimicking others’ behavior can have spill-over effect on consumer behaviors more generally. Being mimicked can enhance both product preferences and memory for the consumption experience (Kulesza et al., 2017; Ramanathan & McGill, 2007; Tanner, Ferraro, Chartrand, Bettman, & Van Baaren, 2008). Furthermore, expressing

26

Context

views that are in agreement with a person sharing that experience enhances enjoyment of that experience, and triggers a more holistic (less analytic) processing, which affects how the experience is evaluated retrospectively (Raghunathan & Corfman, 2006). Furthermore, Bhargave and Montgomery (2013) found that people were more likely to be influenced by first impressions of an experience when remembering a shared, compared with an unshared, experience. Modeling also occurs because the information provided by others offers information about what is the most appropriate or “correct” choice in that context (Deutsch & Gerard, 1955). This has been demonstrated in studies that have used a remote confederate, in which the norm set by the confederate does not serve to promote affiliation (i.e. because there is no other person present). Instead, the intake of the fictitious participants indicates the “right” way to behave in terms of how much to eat or what foods to choose, and so that norm is adopted (e.g., Roth et al., 2001). Similarly, social norms for other behaviors, including recycling behavior, avoidance of littering, and energysaving behaviors, operate when others are not physically present (e.g., Cialdini, Reno, & Kallgren, 1990; Kallgren, Reno, & Cialdini, 2000; Larimer, Turner, Mallett, & Geisner, 2004). These data are consistent with the suggestion that people look to others to provide information about the most appropriate way to act. The mechanisms underlying modeling are unclear, but some evidence suggests that modeling of consumer behaviors is underpinned by a fundamental propensity to imitate the behavior of others (Chartrand & Bargh, 1999). In line with this idea, people mimic the consumption behaviors of others directly by taking a sip or reaching for food immediately after an observed person performs the same behavior (Hermans et al., 2012; Koordeman, Kuntsche, Anschutz, van Baaren, & Engels, 2011; Larsen, Engels, Granic, & Overbeek, 2009; Sharps et al., 2015). This behavior may be underpinned by basic neural processes that link perception with action, known as the “mirror neuron system” (Rizzolatti & Craighero, 2004). Indeed, it has been reported recently that modeling of eating is associated with activity in the mirror neuron system (Mcgeown & Davis, 2018). Modeling is also likely underpinned by changes in preferences for modeled items. If we learn that peers have a preference for a product, then we expect to like it too, and will place a higher value on the item (Nook & Zaki, 2015; Robinson & Higgs, 2012). Such expectancies can also produce placebo effects. Participants who consumed water falsely-labeled as containing caffeine experienced more alertness and demonstrated stronger product endorsement when a confederate reported a similar response, relative to when the confederate reported no response (Crum, Phillips, Goyer, Akinola, & Higgins, 2016). Another study found that labelling a product with information that more and more people are reducing their salt intake, increased product choice (Zandstra, Carvalho, & Van Herpen, 2017). Interestingly, products that connect with a person’s social identity are also evaluated more positively. Hackel and colleagues found that participants whose Canadian social identity was made salient demonstrated greater preference for maple syrup (versus honey), compared with those whose Canadian identity was not made salient (Hackel, Coppin, Wohl, & Van Bavel, 2018). These data suggest that modeling affects food choice and intake by altering the sensory/ hedonic evaluation of foods.

People in context—The social perspective

27

Strong modeling effects have also been observed for alcohol consumption (Caudill & Kong, 2001; Larsen et al., 2009; Robinson et al., 2016), and perceptions of normative behavior have been linked to various health-related behaviors, including stair climbing, drunk driving, smoking, and unsafe sex (Burger & Shelton, 2011; Chernoff & Davison, 2005; DeJong, Larimer, Wood, & Hartman, 2009; Linkenbach & Perkins, 2003; Perkins, Linkenbach, Lewis, & Neighbors, 2010). There is also evidence that people follow norms provided by others for a wide range of consumer behaviors. For example, social norm messages have been reported to increase towel reuse behaviors (Goldstein, Cialdini, & Griskevicius, 2008), energy conservation behaviors (Schultz, Nolan, Cialdini, Goldstein, & Griskevicius, 2007) littering (Cialdini et al., 1990), and recycling (Schultz, 1999). In summary, there is strong evidence that behavior of consumers is influenced by the preferences of others. This is because other people provide information about “good” choices, and aligning our choices with valued others enhances our personal relationships and sense of self (Wood & Hayes, 2012).

2.3.3 Impression management Adapting one’s behavior to create a particular impression of oneself to others is known as impression management (Leary, 1995). We are usually motivated to present ourselves in a positive light to others, especially to people who do not know us well (Baumeister & Leary, 1995). We are also aware that our behavior reflects the kind of person we are, and so we may make specific consumption choices to express something about ourselves to others (Vartanian, Herman, & Polivy, 2007). For example, when having lunch with colleagues for the first time, we may choose to eat a salad rather than a burger, because this may fit with the image of ourselves we might want to convey to colleagues at that moment. Impression management is based on shared assumptions about the personal characteristics of people who engage in particular behaviors (also known as stereotypes). In the case of eating behavior, people readily make judgements about others based on what and how much they eat, and there is some evidence that people may use these stereotypes to manage how they are perceived, especially in the context of unfamiliar others (Vartanian et al., 2015). Research on consumption stereotypes has identified that certain foods and food choices are associated with impressions of masculinity or femininity. Meat eating is associated with masculinity (Rothgerber, 2013; Rozin, Hormes, Faith, & Wansink, 2012), whereas meat avoidance and consumption of vegetables, salad, fish, and sweet foods is associated with femininity (Cavazza, Guidetti, & Butera, 2015; Jensen & Holm, 1999; Rothgerber, 2013; Ruby & Heine, 2011). In a study of Japanese students, Kimura and colleagues found that sweet foods and salad were more likely to be associated with feminine names, whereas meat dishes were associated with masculine names (Kimura et al., 2009). These findings suggest that people implicitly categorize foods as either feminine or masculine. Furthermore, there is evidence that eating “good” foods is seen as feminine, and eating “bad” foods is seen as masculine. Stein and Nemeroff (1995) reported that men who

28

Context

ate “bad” foods (i.e. high-calorie foods thought to be bad for health) were rated as more masculine (and less feminine) than were men who ate “good” foods (i.e. low-calorie foods thought to be good for health). Both women and men were judged as being more feminine and less masculine when they were depicted eating low-fat foods, compared with when they were depicted eating high-fat foods (Barker, Tandy, & Stookey, 1999). Other evidence suggests that food choices reflect the character of the consumer more generally. Overall, people who eat “good” foods are perceived as being “better” people than are those who eat “bad” foods. For example, people who eat “good” foods were judged to be more attractive, healthier, more moral, and more intelligent than were consumers of “bad” foods (Stein & Nemeroff, 1995). However, they were also judged as more serious and less likable (Barker et al., 1999). Conversely, consumers of high-fat diets were perceived to be unattractive, unintelligent, and working class, but also fun-loving, happy, and sociable (Barker et al., 1999). Consumption stereotypes also extend to judgments about amounts eaten. Bock and Kanarek (1995) provided participants with descriptions of a target person eating a small, moderate, or large amount. As meal size increased, participants rated both female and male targets as more masculine and less feminine. Eating smaller meals has also been associated with perceptions of neatness and attractiveness, especially for women (Bock & Kanarek, 1995; Chaiken & Pliner, 1987), and small portions of foods are rated as more feminine than are larger portions of the same food (Cavazza et al., 2015). There is indirect evidence that people use consumption stereotypes to express something about themselves within social interactions. This evidence is derived from studies examining eating behavior in situations where people may be particularly motivated to manage impressions. For example, women tend to eat lightly in the company of men, and this has been suggested to be because eating a small amount is an affirmation of gender identity (Pliner & Chaiken, 1990). White and Dahl (2006) conducted a study in which participants were asked to imagine that they had been sent on a training course by their employer. They read a scenario that described that they had been in workshops all day, and were planning on ordering something from the room service menu for dinner. To encourage the choice of steak, and in particular, a small steak, participants were told: “You aren’t feeling too hungry because you had a late lunch; however, you are tempted to select steak for dinner.” They were then asked to select from a hypothetical menu, and to evaluate each menu option. Men were less likely to choose a small steak (versus a large steak) when it was described as a ladies’ cut than when it was described as a chef’s cut. However, this was only observed when men thought they would be consuming the steak in public, and not when they thought they would be consuming the steak in private. The men also evaluated the ladies’ cut steak less favorably than the chef’s cut steak. The authors suggested that the menu choices were motivated by impression management concerns, such that the men avoided choosing a food associated with femininity to maintain their masculine identity. A similar conclusion was drawn by Gal and Wilkie (2010), who found that men were less likely to choose stereotypically feminine foods, after their masculine identity was challenged. However, this effect was observed only when they were not under time pressure, and thus had the cognitive recourses to regulate their choice to maintain their gender identity. More recently, Cavazza and colleagues (Cavazza et al., 2015)

People in context—The social perspective

29

found that certain aspects of a dish (e.g., food type, portion size and presentation) affected the perception of the dish as feminine or masculine. These perceptions influenced women’s intentions to consume the food such that women were more likely to say they intended to consume a small portion of salad when it was elegantly presented, because they perceived it to be more feminine. Drawing upon the preceding information, there is substantial evidence to suggest that foods and dishes can be perceived as “good” or “bad” and “feminine” or “masculine,” and that people are judged on their food choices on the basis of these stereotypes. There is also some evidence that people adjust their eating behavior in light of these stereotypes to create and/or maintain a particular self-image. There are clear implications of these findings for the marketing of foods, and for the design of healthy eating campaigns.

2.4

Competing social influences

We have reviewed the ways in which social context influences consumer choices, using eating behaviors as an example. In doing so, we have highlighted (1) that the mere presence of others facilitates eating behaviors, (2) that people tend to follow social norms when making decisions about what and how much to eat, and (3) that people also adapt these choices based on impression management concerns. What has been less well studied is what happens in real-world consumption contexts when there are multiple social influences at play. In real-world contexts, there will be potentially conflicting social motivations that influence behavior. For example, in the case of food choices, a woman might be motivated to model the hearty appetite of a male dining companion to ingratiate herself, but this might conflict with attempts to convey a feminine gender identity. Interestingly, when faced with such competing motives in an experimental study, women used portion size to signal their gender identity (choosing a smaller portion of a food), but modeled the food choices of their partner by choosing a similar dish (Cavazza, Guidetti, & Butera, 2017). Further research is required to examine how competing motives play out in different scenarios, particularly in the case of joint decision making in couples and families (Wood & Hayes, 2012). For example, how would someone reconcile a desire to align their choices with a romantic partner with potentially conflicting motives regarding their desire to create a good impression on their wider social group? Understanding of consumer behavior will be advanced by a clearer account of the salient motives that govern choice in specific social contexts.

2.5

Implications for consumer research and product development

It is clear from the evidence reviewed thus far that the utilization and evaluation of a product varies according to the social context. Some choices will be specific to certain contexts, and that context will influence the evaluation of the consumption experience via processes such as social facilitation and modeling. Moreover, the same choices

30

Context

may be made in different social contexts, but the underlying motives may differ. For example, I may choose French cheese to signal national identity in one context, and because my dining partners are choosing it in another context. These motives may also change the nature of the consumption experience. Therefore, if the social context of consumption is not taken into account when developing a product, important information will be missed. It has been argued that new product failures can often be attributed to not studying consumers in the contexts in which they actually interact with the products (K€ oster & Mojet, 2018), and this includes the social context of consumption. Other people are often part of the context in which consumption occurs, but social factors shape the meaning of consumption, interactions with products, and product decision making. The social context of eating and the meaning attached to such contexts will also vary depending on the prevailing cultural context, which highlights that social effects on consumption will also vary cross-culturally (Danesi, 2017). In the case of food products, it may be particularly important to examine the interaction between consumer and social context, because many consumption experiences are social events. Commensality is often seen as an inherent core of meals (Fischler, 2011), and food and eating play integral parts in our social lives, as they reinforce social connections and help us communicate and express ourselves (M€akel€a, 2009; Murcott, 1983). Many people make time to eat together, especially at the evening meal. A 2014 YouGov survey found that 77% of UK adults eat as a household at least once a week, and 48% eat together daily. A survey from the USA (NPD Group, 2014) found that 68% of those surveyed reported eating an evening meal socially. According to figures from the Credoc Research Institute, 80% of meals are taken with other people in France. It is also important to acknowledge that choices today are often made in the context of social media and signifiers of social popularity (e.g., “likes”) that are also likely to have a large influence on preferences. There have been extensive discussions around how best to incorporate contextual influences in consumer-based research generally (e.g., see Meiselman, 1992). Various methods/approaches have been proposed, including item-by-use methods, home use testing, experimental restaurants, and evoked contexts (for reviews see Jaeger & Porcherot, 2017). However, the social context of eating has perhaps been a neglected factor in some of these methods, possibly because appropriate social eating environments are seen as difficult to recreate in controlled settings (Delarue & Boutrolle, 2010). There is scope for future research to provide a more detailed assessment of the specific social contexts in which consumption occurs, and how this relates to product evaluation and decision making. Insights can be gained from the exploration of social contextual effects on consumer behavior using ethnographic and qualitative methods, as well as context evocation using written scenarios. Developments in the use of virtual reality techniques also offer opportunities to create immersive environments that combine contextual influences with product interaction (Bangcuyo et al., 2015). Finally, there is the potential to harness recent advances in artificial intelligence research to analyze information from large sets of observations of eating behaviors, and uncover the complex relationships between social context and eaters’ decisions (Akkoyunlu, Manfredotti, Cornuejols, Darcel, & Delaere, 2017). Non-supervised algorithms such as Bayesian network approaches (Getoor & Taskar, 2007;

People in context—The social perspective

31

Heckerman, 2013) or machine learning techniques to measure causality links (Scholkopf & Janzing, 2018) can be developed and employed on very large numbers of behavioral observations to disentangle relationships between social contexts and food decisions or food preferences. Although very promising, this emerging field of research still requires major scientific and methodological improvements to fully benefit consumer and sensory sciences. Methodological research in this domain will be needed to facilitate the acquisition and the thorough pre-processing of a sufficient number of observations. Also, development of algorithms will be needed to adapt to the specificity of food decisions because they are mostly expressed as positive choices (foods that are chosen), and information on negative decisions (foods that are not chosen) are therefore often scarce or missing.

2.6

Conclusions

The social context of consumption has powerful and pervasive effects on consumer choices and experiences. Social influences on consumption are many and varied, and considerable work has been done recently, particularly in the area of social influences on eating behaviors, which has expanded the evidence base and our appreciation of the factors that moderate social influences on consumption. Many unanswered questions remain, but one area ripe for further work is how best to incorporate social context into consumer research. Greater understanding of people and products in their social context is necessary to provide a more complete understanding of consumer behaviors.

Acknowledgments Funding: This work and the work cited herein were supported by a grant from the Economic and Social Research Council (ESRC), grant number: ES/P01027X/1. The funder had no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report, and in the decision to submit the article for publication.

Conflict of Interest Suzanne Higgs and Helen Ruddock and Nicolas Darcel declare they have no conflict of interest.

References Akkoyunlu, S., Manfredotti, C., Cornuejols, A., Darcel, N., & Delaere, F. (2017). Investigating substitutability of food items in consumption data. In Second international workshop on health recommender systems (p. 27). Bangcuyo, R. G., Smith, K. J., Zumach, J. L., Pierce, A. M., Guttman, G. A., & Simons, C. T. (2015). The use of immersive technologies to improve consumer testing: The role of ecological validity, context and engagement in evaluating coffee. Food Quality and Preference, 41, 84–95.

32

Context

Barker, M. E., Tandy, M., & Stookey, J. D. (1999). How are consumers of low-fat and high-fat diets perceived by those with lower and higher fat intake? Appetite, 33(3), 309–317. Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497–529. Berry, S. L., Beatty, W. W., & Klesges, R. C. (1985). Sensory and social influences on ice-cream consumption by males and females in a laboratory setting. Appetite, 6(1), 41–45. Bevelander, K. E., Ansch€utz, D. J., & Engels, R. C. (2012). Social norms in food intake among normal weight and overweight children. Appetite, 58(3), 864–872. Bhargave, R., & Montgomery, N. V. (2013). The social context of temporal sequences: Why first impressions shape shared experiences. Journal of Consumer Research, 40(3), 501–517. Bock, B. C., & Kanarek, R. B. (1995). Women and men are what they eat: The effects of gender and reported meal size on perceived characteristics. Sex Roles, 33(1–2), 109–119. Boothby, E. J., Clark, M. S., & Bargh, J. A. (2014). Shared experiences are amplified. Psychological Science, 25, 2209–2216. Brindal, E., Wilson, C., Mohr, P., & Wittert, G. (2015). Eating in groups: Do multiple social influences affect intake in a fast-food restaurant? Journal of Health Psychology, 20(5), 483–489. Brunstrom, J. M., & Shakeshaft, N. G. (2009). Measuring affective (liking) and non- affective (expected satiety) determinants of portion size and food reward. Appetite, 52, 108–114. Burger, J. M., & Shelton, M. (2011). Changing everyday health behaviors through descriptive norm manipulations. Social Influence, 6(2), 69–77. Cai, H., Chen, Y., & Fang, H. (2009). Observational learning: Evidence from a randomized natural field experiment. American Economic Review, 99(3), 864–882. Caudill, B. D., & Kong, F. H. (2001). Social approval and facilitation in predicting modeling effects in alcohol consumption. Journal of Substance Abuse, 13(4), 425–441. Cavanaugh, L. A. (2016). Consumer behavior in close relationships. Current Opinion in Psychology, 10, 101–106. Cavazza, N., Graziani, A. R., & Guidetti, M. (2011). Looking for the “right” amount to eat at the restaurant: Social influence effects when ordering. Social Influence, 6(4), 274–290. Cavazza, N., Guidetti, M., & Butera, F. (2015). Ingredients of gender-based stereotypes about food. Indirect influence of food type, portion size and presentation on gendered intentions to eat. Appetite, 91, 266–272. Cavazza, N., Guidetti, M., & Butera, F. (2017). Portion size tells who I am, food type tells who you are: Specific functions of amount and type of food in same-and opposite-sex dyadic eating contexts. Appetite, 112, 96–101. Chaiken, S., & Pliner, P. (1987). Women, but not men, are what they eat: The effect of meal size and gender on perceived femininity and masculinity. Personality and Social Psychology Bulletin, 13(2), 166–176. Chartrand, T. L., & Bargh, J. A. (1999). The chameleon effect: The perception-behavior link and social interaction. Journal of Personality and Social Psychology, 76, 893–910. Chernoff, R. A., & Davison, G. C. (2005). An evaluation of a brief HIV/AIDS prevention intervention for college students using normative feedback and goal setting. AIDS Education & Prevention, 17(2), 91–104. Christakis, N. A., & Fowler, J. H. (2007). The spread of obesity in a large social network over 32 years. New England Journal of Medicine, 357(4), 370–379. Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58, 1015–1026.

People in context—The social perspective

33

Clendenen, V. I., Herman, C. P., & Polivy, J. (1994). Social facilitation of eating among friends and strangers. Appetite, 23, 1–13. Crum, A. J., Phillips, D. J., Goyer, J. P., Akinola, M., & Higgins, E. T. (2016). Transforming water: Social influence moderates psychological, physiological, and functional response to a placebo product. PLoS One, 11(11), e0167121. Cruwys, T., Bevelander, K. E., & Hermans, R. C. (2015). Social modeling of eating: A review of when and why social influence affects food intake and choice. Appetite, 86, 3–18. Cruwys, T., Platow, M. J., Angullia, S. A., Chang, J. M., Diler, S. E., Kirchner, J. L., et al. (2012). Modeling of food intake is moderated by salient psychological group membership. Appetite, 58(2), 754–757. Danesi, G. (2017). A cross-cultural approach to eating together: Practices of commensality among French, German and Spanish young adults. Social Science Information, 57(1), 99–120. de Castro, J. M. (1990). Social facilitation of duration and size but not rate of the spontaneous meal intake of humans. Physiology and Behavior, 47, 1129–1135. de Castro, J. M. (1991). Social facilitation of the spontaneous meal size humans occurs on both weekdays and weekends. Physiology and Behavior, 49, 1289–1291. de Castro, J. M. (1994). Family and friends produce greater social facilitation of food intake than other companions. Physiology & Behaviour, 56(3), 445–455. de Castro, J. M., & Brewer, E. M. (1992). The amount eaten in meals by humans is a power function of the number of people present. Physiology & Behaviour, 51(1), 121–125. de Castro, J. M., & de Castro, E. S. (1989). Spontaneous meal patterns in humans: Influence of the presence of other people. American Journal of Clinical Nutrition, 50, 237–247. DeJong, W., Larimer, M. E., Wood, M. D., & Hartman, R. (2009). NIAAA’s rapid response to college drinking problems initiative: Reinforcing the use of evidence-based approaches in college alcohol prevention. Journal of Studies on Alcohol and Drugs, 16, 5–11. Delarue, J., & Boutrolle, I. (2010). The effects of context on liking: Implications for hedonic measurements in new product development. In Consumer-driven innovation in food and personal care products (pp. 175–218). Deutsch, M., & Gerard, H. B. (1955). A study of normative and informational social influences upon individual judgment. The Journal of Abnormal and Social Psychology, 51(3), 629. Fay, S. H., Ferriday, D., Hinton, E. C., Shakeshaft, N. G., Rogers, P. J., & Brunstrom, J. M. (2011). What determines real-world meal size? Evidence for pre-meal planning. Appetite, 56, 284–289. Feeney, J. R., Polivy, J., Pliner, P., & Sullivan, M. D. (2011). Comparing live and remote models in eating conformity research. Eating Behaviors, 12(1), 75–77. Fischler, C. (2011). Commensality, society and culture. Social Science Information, 50(3–4), 528–548. Gal, D., & Wilkie, J. (2010). Real men don’t eat quiche: Regulation of gender-expressive choices by men. Social Psychological and Personality Science, 1(4), 291–301. Geller, E. S., Russ, N. W., & Altomari, M. G. (1986). Naturalistic observations of beer drinking among college students. Journal of Applied Behavior Analysis, 19(4), 391–396. Getoor, L., & Taskar, B. (2007). Introduction to statistical relational learning. MIT Press. Gierl, H., & Huettl, V. (2010). Are scarce products always more attractive? The interaction of different types of scarcity signals with products’ suitability for conspicuous consumption. International Journal of Research in Marketing, 27, 225–235. Goldman, S. J., Herman, C. P., & Polivy, J. (1991). Is the effect of a social model attenuated by hunger? Appetite, 17, 129–140.

34

Context

Goldstein, N. J., Cialdini, R. B., & Griskevicius, V. (2008). A room with a viewpoint: Using social norms to motivate environmental conservation in hotels. Journal of Consumer Research, 35(3), 472–482. Guarino, M., Fridrich, P., & Sitton, S. (1995). Male and female conformity in eating behavior. Psychological Reports, 75, 603–609. Guendelman, M. D., Cheryan, S., & Monin, B. (2011). Fitting in but getting fat: Identity threat and dietary choice among U.S. immigrant groups. Psychological Science, 22, 959–967. Hackel, L. M., Coppin, G., Wohl, M. J., & Van Bavel, J. J. (2018). From groups to grits: Social identity shapes evaluations of food pleasantness. Journal of Experimental Social Psychology, 74, 270–280. Haye, K., Robins, G., Mohr, P., & Wilson, C. (2013). Adolescents’ intake of junk food: Processes and mechanisms driving consumption similarities among friends. Journal of Research on Adolescence, 23, 524–536. Heckerman, D. (2013). A Bayesian approach to learning causal networks. ArXiv:1302.4958 [Cs]. Herman, C. P. (2015). The social facilitation of eating. A review. Appetite, 86, 61–73. Herman, C. P., Roth, D. A., & Polivy, J. (2003). Effects of the presence of others on food intake: A normative interpretation. Psychological Bulletin, 129(6), 873–886. Hermans, R. C., Engels, R. C., Larsen, J. K., & Herman, P. C. (2009). Modeling of palatable food intake. The influence of quality of social interaction. Appetite, 52, 801–804. Hermans, R. C., Herman, P. C., Larsen, J., & Engels, R. C. (2010). Social modeling effects on young women’s breakfast intake. Journal of the American Dietetic Association, 110, 1901–1905. Hermans, R. C., Larsen, J. K., Herman, C. P., & Engels, R. C. (2008). Modeling of palatable food intake in female young adults. Effects of perceived body size. Appetite, 51(3), 512–518. Hermans, R. C., Lichtwarck-Aschoff, A., Bevelander, K. E., Herman, C. P., Larsen, J. K., & Engels, R. C. (2012). Mimicry of food intake: The dynamic interplay between eating companions. PLoS One, 7(2), e31027. Hetherington, M. M., Anderson, A. S., Norton, G. N., & Newson, L. (2006). Situational effects on meal intake. A comparison of eating alone and eating with others. Physiology & Behavior, 88, 498–505. Higgs, S., & Thomas, J. (2016). Social influences on eating. Current Opinion in Behavioral Sciences, 9, 1–6. Hirsch, E. S., & Kramer, F. M. (1993). Situational influences on food intake. In B. M. Marriott (Ed.), Nutrional needs in hot environments: Applications for military personnel in field operations (pp. 215–243). Washington, DC: National Academies Press. Hock, S. J., & Bagchi, R. (2017). The impact of crowding on calorie consumption. Journal of Consumer Research, 44(5), 1123–1140. Hoffman, S. J., & Tan, C. (2015). Biological, psychological and social processes that explain celebrities’ influence on patients’ health-related behaviors. Archives of Public Health, 73(1), 3. Huh, Y. E., Vosgerau, J., & Morewedge, C. K. (2014). Social defaults: Observed choices become choice defaults. Journal of Consumer Research, 41(3), 746–760. Jaeger, S. R., & Porcherot, C. (2017). Consumption context in consumer research: Methodological perspectives. Current Opinion in Food Science, 15, 30–37. Jensen, K. O. D., & Holm, L. (1999). Preferences, quantities and concerns: Socio-cultural perspectives on the gendered consumption of foods. European Journal of Clinical Nutrition, 53(5), 351. Kaisari, P., & Higgs, S. (2015). Social modelling of food intake. The role of familiarity of the dining partners and food type. Appetite, 86, 19–24.

People in context—The social perspective

35

Kallgren, C. A., Reno, R. R., & Cialdini, R. B. (2000). A focus theory of normative conduct. When norms do and do not affect behavior. Personality and Social Psychology Bulletin, 26 (8), 1002–1012. Kimura, A., Wada, Y., Goto, S. I., Tsuzuki, D., Cai, D., Oka, T., et al. (2009). Implicit gender-based food stereotypes. Semantic priming experiments on young Japanese. Appetite, 52(2), 521–524. Klesges, R. C., Bartsch, D., Norwood, J. D., Kautzrnan, D., & Haugrud, S. (1984). The effects of selected social and environmental variables on the eating behavior of adults in the natural environment. International Journal of Eating Disorders, 3(4), 35–41. Koordeman, R., Kuntsche, E., Anschutz, D. J., van Baaren, R. B., & Engels, R. C. (2011). Do we act upon what we see? Direct effects of alcohol cues in movies on young adults’ alcohol drinking. Alcohol and Alcoholism, 46(4), 393–398. K€ oster, E., & Mojet, J. (2018). Complexity of consumer perception: Thoughts on pre–product launch research. Methods in Consumer Research, 1, 23–45. Krantz, D. S. (1979). Naturalistic study of social influences on meal size among moderately obese and non-obese subjects. Psychosomatic Medicine, 41, 19–27. Kristensen, S. T., Holm, L., Raben, A., & Astrup, A. (2002). Achieving “proper” satiety in different social contexts–qualitative interpretations from a cross-disciplinary project, sociomæt. Appetite, 39(3), 207–215. Kulesza, W., Dolinski, D., Migon, M., Rizulla, A., Gamian-Wilk, M., & Grzyb, T. (2017). The use of mimicry to improve evaluation of unsought beverages. Food Quality and Preference, 62, 137–143. Larimer, M. E., Turner, A. P., Mallett, K. A., & Geisner, I. M. (2004). Predicting drinking behavior and alcohol-related problems among fraternity and sorority members. Examining the role of descriptive and injunctive norms. Psychology of Addictive Behaviors, 18(3), 203–212. Larsen, H., Engels, R. C., Granic, I., & Overbeek, G. (2009). An experimental study on imitation of alcohol consumption in same-sex dyads. Alcohol & Alcoholism, 44(3), 250–255. Laurier, E., & Wiggins, S. (2011). Finishing the family meal. The interactional organisation of satiety. Appetite, 56(1), 53–64. Leary, M. R. (1995). Self-presentation: Impression management and interpersonal behavior. Brown & Benchmark Publishers. Leone, T., Pliner, P., & Herman, C. P. (2007). Influence of clear versus ambiguous normative information on food intake. Appetite, 49(1), 58–65. Linkenbach, J. W., & Perkins, H. W. (2003). MOST of us are tobacco free: An eight-month social norms campaign reducing youth initiation of smoking in Montana. In H. W. Perkins (Ed.), The social norms approach to preventing school and college age substance abuse: A handbook for educators, counselors, and clinicians (pp. 224–234). San Francisco, CA, US: Jossey-Bass. M€akel€a, J. (2009). Meals: The social perspective. In H. L. Meiselman (Ed.), Meals in science and practice (pp. 37–49). Cambridge: Woodhead Publishing. McFerran, B., Dahl, D., Fitzsimons, G., & Morales, A. (2010). I’ll have what she’s having. Effects of social influence and body type on the food choices of others. Journal of Consumer Research, 36, 915–929. Mcgeown, L., & Davis, R. (2018). Social modeling of eating mediated by mirror neuron activity: A causal model moderated by frontal asymmetry and BMI. Behavioural Brain Research, 338, 66–75. Meiselman, H. L. (1992). Methodology and theory in human eating research. Appetite, 19(1), 49–55.

36

Context

Meiselman, H. L. (2013). The future in sensory/consumer research:….evolving to a better science. Food Quality and Preference, 27, 208–214. Mollen, S., Rimal, R. N., Ruiter, R. A., & Kok, G. (2013). Healthy and unhealthy social norms and food selection. Findings from a field experiment. Appetite, 65, 83–89. Murcott, A. (1983). The sociology of food and eating. Aldershot, UK: Gower. Nook, E. C., & Zaki, J. (2015). Social norms shift behavioral and neural responses to foods. Journal of Cognitive Neuroscience, 27(7), 1412–1426. NPD Group (2014). Consumers are alone over half of eating occasions as a result of changing lifestyles and more single-person households. Available from: https://www.npd.com/wps/ portal/npd/us/news/press-releases/consumers-are-alone-over-half-of-eating-occasions-as-aresult-of-changing-lifestyles-and-more-single-person-households-reports-npd/. (Accessed 28 March 2018). Pachucki, M. A., Jacques, P. F., & Christakis, N. A. (2011). Social network concordance in food choice among spouses, friends, and siblings. American Journal of Public Health, 101(11), 2170–2177. Pedersen, S., Grønhøj, A., & Thøgersen, J. (2015). Following family or friends. Social norms in adolescent healthy eating. Appetite, 86, 54–60. Pelletier, J. E., Graham, D. J., & Laska, M. N. (2014). Social norms and dietary behaviors among young adults. American Journal of Health Behavior, 38, 144–152. Peneau, S., Mekhmoukh, A., Chapelot, D., Dalix, A.-M., Airinei, G., Hercberg, S., et al. (2009). Influence of environmental factors on food intake and choice of beverage during meals in teenagers. A laboratory study. British Journal of Nutrition, 102, 1854–1859. Perkins, H. W., Linkenbach, J. W., Lewis, M. A., & Neighbors, C. (2010). Effectiveness of social norms media marketing in reducing drinking and driving: A statewide campaign. Addictive Behaviors, 35(10), 866–874. Pliner, P., & Chaiken, S. (1990). Eating, social motives, and self-presentation in women and men. Journal of Experimental Social Psychology, 26(3), 240–254. Pliner, P., & Mann, N. (2004). Influence of social norms and palatability on amount consumed and food choice. Appetite, 42(2), 227–237. Pozharliev, R., Verbeke, W. J., Van Strien, J. W., & Bagozzi, R. P. (2015). Merely being with you increases my attention to luxury products: Using EEG to understand consumers’ emotional experience with luxury branded products. Journal of Marketing Research, 52 (4), 546–558. Prinsen, S., de Ridder, D. T., & de Vet, E. (2013). Eating by example. Effects of environmental cues on dietary decisions. Appetite, 70, 1–5. Quester, P., & Steyer, A. (2009). Revisiting individual choices in group settings: The long and winding (less traveled) road? Journal of Consumer Research, 36(6), 1050–1057. Raghunathan, R., & Corfman, K. (2006). Is happiness shared doubled and sadness shared halved? Social influence on enjoyment of hedonic experiences. Journal of Marketing Research, 43(3), 386–394. Ramanathan, S., & McGill, A. L. (2007). Consuming with others: Social influences on momentto-moment and retrospective evaluations of an experience. Journal of Consumer Research, 34(4), 506–524. Redd, E. M., & de Castro, J. M. (1992). Social facilitation of eating: Effects of instructions to eat alone or with others. Physiology & Behavior, 52, 749–754. Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of Neuroscience, 27, 169–192. Robinson, E., Benwell, H., & Higgs, S. (2013). Food intake norms increase and decrease snack food intake in a remote confederate study. Appetite, 65(1), 20–24.

People in context—The social perspective

37

Robinson, E., & Higgs, S. (2012). Liking food less: The impact of social influence on food liking evaluations in female students. PLoS One, 7(11)e48858. Robinson, E., & Higgs, S. (2013). Making food choices in the presence of ‘healthy’ and ‘unhealthy’ companions. British Journal of Nutrition, 109(4), 765–771. Robinson, E., Oldham, M., Sharps, M., Cunliffe, A., Scott, J., & Clark, E. (2016). Social imitation of alcohol consumption and ingratiation motives in young adults. Psychology of Addictive Behaviors, 30(4), 442. Robinson, E., Tobias, T., Shaw, L., Freeman, E., & Higgs, S. (2011). Social matching of food intake and the need for social acceptance. Appetite, 56(3), 747–752. Rolls, B. J., Roe, L. S., & Meengs, J. S. (2007). The effect of large portion sizes on energy intake is sustained for 11 days. Obesity, 15(6), 1535–1543. Roth, D. A., Herman, C. P., Polivy, J., & Pliner, P. (2001). Self-presentational conflict in social eating situations: A normative perspective. Appetite, 36(2), 165–171. Rothgerber, H. (2013). Real men don’t eat (vegetable) quiche. Masculinity and the justification of meat consumption. Psychology of Men and Masculinity, 14, 363–375. Rozin, P. (1996). The socio-cultural context of eating and food choice. In H. L. Meiselman & H. J. H. MacFie (Eds.), Food choice, acceptance and consumption (pp. 83–104). US: Springer. Rozin, P., Hormes, J. M., Faith, M. S., & Wansink, B. (2012). Is meat male? A quantitative multimethod framework to establish metaphoric relationships. Journal of Consumer Research, 39, 629–643. Ruby, M. B., & Heine, S. J. (2011). Meat, morals, and masculinity. Appetite, 56(2), 447–450. Salmon, S. J., De Vet, E., Adriaanse, M. A., Fennis, B. M., Veltkamp, M., & De Ridder, D. T. (2015). Social proof in the supermarket: Promoting healthy choices under low self-control conditions. Food Quality and Preference, 45, 113–120. Salvy, S. J., Vartanian, L. R., Coelho, J. S., Jarrin, D., & Pliner, P. P. (2008). The role of familiarity on modeling of eating and food consumption in children. Appetite, 50(2–3), 514–518. Scholkopf, B., & Janzing, D. (2018). Elements of causal inference: Foundations and learning algorithms. Cambridge, Massachusetts: MIT Press. Schultz, P. W. (1999). Changing behavior with normative feedback interventions: A field experiment on curbside recycling. Basic and Applied Social Psychology, 21, 25–38. Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J., & Griskevicius, V. (2007). The constructive, destructive, and reconstructive power of social norms. Psychological Science, 18(5), 429–434. Sch€uz, B., Bower, J., & Ferguson, S. G. (2015). Stimulus control and affect in dietary behaviours. An intensive longitudinal study. Appetite, 87, 310–317. Sch€uz, B., Revell, S., Hills, A. P., Sch€uz, N., & Ferguson, S. G. (2017). Higher BMI is associated with stronger effects of social cues on everyday snacking behaviour. Appetite, 114, 1–5. Sharps, M., Higgs, S., Blissett, J., Nouwen, A., Chechlacz, M., Allen, H. A., et al. (2015). Examining evidence for behavioural mimicry of parental eating by adolescent females. An observational study. Appetite, 89, 56–61. Sommer, R., & Sommer, B. A. (1989). Social facilitation effects in coffeehouses. Environment and Behavior, 21(6), 651–666. Sommer, R., Wynes, M., & Brinkley, G. (1992). Social facilitation effects in shopping behavior. Environment and Behavior, 24(3), 285–297. Stein, R. I., & Nemeroff, C. J. (1995). Moral overtones of food: Judgments of others based on what they eat. Personality and Social Psychology Bulletin, 21(5), 480–490.

38

Context

Stok, F. M., Ridder, D. T., Vet, E., & Wit, J. B. (2014). Don’t tell me what I should do, but what others do: The influence of descriptive and injunctive peer norms on fruit consumption in adolescents. British Journal of Health Psychology, 19(1), 52–64. Tanner, R. J., Ferraro, R., Chartrand, T. L., Bettman, J. R., & Van Baaren, R. (2008). Of chameleons and consumption: the impact of mimicry on choice and preferences. Journal of Consumer Research, 34, 754–766. Thomas, J. M., Ursell, A., Robinson, E. L., Aveyard, P., Jebb, S. A., Herman, C. P., et al. (2017). Using a descriptive social norm to increase vegetable selection in workplace restaurant settings. Health Psychology, 36(11), 1026. Vartanian, L. R., Herman, C. P., & Polivy, J. (2007). Consumption stereotypes and impression management: How you are what you eat. Appetite, 48(3), 265–277. Vartanian, L. R., Sokol, N., Herman, C. P., & Polivy, J. (2013). Social models provide a norm of appropriate food intake for young women. PLoS One, 8(11), e79268. Vartanian, L. R., Spanos, S., Herman, C. P., & Polivy, J. (2015). Modeling of food intake: A meta-analytic review. Social Influence, 10(3), 119–136. Versluis, I., & Papies, E. K. (2016). The role of social norms in the portion size effect: Reducing normative relevance reduces the effect of portion size on consumption decisions. Frontiers in Psychology, 7, 756. White, K., & Dahl, D. W. (2006). To be or not be? The influence of dissociative reference groups on consumer preferences. Journal of Consumer Psychology, 16, 404–414. Wood, W., & Hayes, T. (2012). Social Influence on consumer decisions: Motives, modes, and consequences. Journal of Consumer Psychology, 22(3), 324–328. ´ . H., & Van Herpen, E. (2017). Effects of front-of-pack social norm Zandstra, E. H., Carvalho, A messages on food choice and liking. Food Quality and Preference, 58, 85–93.

Further reading Cialdini, R. B., & Goldstein, N. J. (2004). Social influence: Compliance and conformity. Annual Review of Psychology, 55, 591–621. Feunekes, G. I., de Graaf, C., Meyboom, S., & van Staveren, W. A. (1998). Food choice and fat intake of adolescents and adults: Associations of intakes within social networks. Preventive Medicine, 27(5), 645–656. Higgs, S. (2015). Social norms and their influence on eating behaviours. Appetite, 86, 38–44. Lakin, J. L., & Chartrand, T. L. (2003). Using nonconscious behavioral mimicry to create affiliation and rapport. Psychological Science, 14(4), 334–339. Robinson, E., Fleming, A., & Higgs, S. (2014). Prompting healthier eating: Comparing the use of health and social norm based messages. Health Psychology, 33, 1057–1064. Tavoularis, G., & Mathe, T. (2010). Le mode`le alimentaire franc¸ais contribue a` limiter le risque d’obesite. Consommation et modes de vie, 232.

Context effects at the level of the sip and bite

3

Armand V. Cardello U.S. Army Natick RD&E Center, Natick, MA, United States

3.1

Introduction

3.1.1 Defining context and its dimensions In a recent review, Cardello and Meiselman (2018) examined the role of context in sensory and hedonic evaluations of food. This review discussed an organizational framework established for different types of context effects in sensory and consumer science that was proposed by Rozin and Tuorila (1993). The latter investigators organized contextual effects into those that are simultaneous (the contextual and target stimuli occur at the same time) or temporal (the contextual stimuli appear before or after the target stimulus). They also categorized the effects into those that are intrinsic (being a part of the target stimulus) and those that are extrinsic (external to the target stimulus). Lastly, they grouped the effects into different levels of analysis, ranging from the macroscopic to the microscopic. These units of analysis included the meal, the dish, the food, and, at the most fundamental level, the sip or bite. The review by Cardello and Meiselman (2018) focused on extrinsic effects (those arising from sources external to the target stimulus) and those occurring at the level of the meal, dish, or food. Intrinsic effects that occur primarily within a sip or bite were not addressed, because they constitute a different level and form of contextual effect, one that requires a careful analysis of the sensory, psychological, and affective mechanisms that underlie the changes in perception that may occur when one ingredient, flavor compound, or sensory attribute is perceived within a context that includes other ingredients, flavor compounds, or sensory attributes, either simultaneously, sequentially, or in a continuous cascade of temporal interrelationships among them. The latter types of “contextual” effects require a more nuanced discussion of the concept of “context.”

3.1.2 Context effects at the level of the sip or bite In this chapter, I focus on context effects that occur within or between individual bites of food or sips of beverages. Such effects are “intrinsic” effects, because they originate within constituents of the “target stimulus,” that is, the food or beverage in the mouth. Classically, context is defined as “the situation in which something happens; the group of conditions that exist where and when something happens” (Merriam-Webster’s Professional Reference Dictionary (2016); see also the Introduction to this book by Context. https://doi.org/10.1016/B978-0-12-814495-4.00003-9 Copyright © 2019 Elsevier Inc. All rights reserved.

40

Context

Meiselman). In psychology, context is defined similarly, so that for sensory stimuli, context refers to “the state or scenario wherein a certain sensation happens” (Nugent, 2013). Thus, within the intrinsic milieu of a sip or bite of food, sensory context refers to how the sensory experience of one target element/ingredient/constituent changes as a function of other “contextual” elements/ingredients/constituents that comprise the functional stimulus in the mouth. Further, this context may be established through simultaneous occurrence of the different ingredients or through sequential (temporal) occurrences, as may occur in a sequence of bites and sips. As an early example of such contextual interactions at the level of the sip or bite, Rosemary Pangborn, who was concerned about the degree to which psychophysical phenomenon were the same or different when studied within the confines of chemosensory laboratories versus real world food applications (see Lahne, 2016, for a modern version of this perspective), studied such contextual phenomena as how thresholds and relative taste intensities of sucrose and fructose changed when mixed in either water or pear nectar, or with or without added citric acid (Pangborn, 1963). As the following sections will show, numerous changes in the perception of a specific stimulus element may occur, depending on the other elements in which it is presented. Hyde and Witherly (1993) addressed the issue of stimulus context as it relates to combinations of taste, smell, somesthetic and other stimuli that comprise a food or beverage being consumed. These investigators used the example of ice cream to underscore the rapidly changing set of interactions that occur between any one element in a food and other elements during the consumption process. Thus, while frozen ice cream is relatively tasteless, as it begins to melt, flavors associated with sweetness and creamy mouthfeel sensations begin to develop. To avoid the extreme cold on the mouth and tongue, the ice cream is moved by the tongue, where it stimulates other receptors. Because these receptors and associated tongue areas often differ in their sensitivity to sweetness or somatosensory sensations, there is a cascade of changing temporal and spatial sensations in which each of the constituents of the ice cream are perceived. As the tongue cools with further consumption, sensitivity to taste compounds decrease, creating further complexity in the interactions of sensory (and hedonic) responses. Finally, when the melted ice cream is swallowed, a whole new set of receptor surfaces are stimulated on the back of the tongue, throat, and glottis. The net result of this process is a continuing cascade of changing relationships among the perceived taste, odor, and somatosensory attributes of the ice cream, creating an ever changing contextual background for the perceptions occurring at any given instant. Hyde and Witherly (1993) went on to argue how this dynamic set of changing sensations can have a dramatic effect on hedonics through a phenomenon they termed “dynamic contrast.”

3.1.3 Context and related concepts in sensory science Even though the term “context” is applicable when discussing a stimulus that is presented either simultaneously with another stimulus, introduced into an existing stimulus environment (e.g., a target odor presented into a room containing another

Context effects at the level of the sip and bite

41

ambient odor), or presented within a temporal series of other stimuli (e.g., a target odor is presented following another specific odor or within a set of several other odorants), sensory scientists often use different terminology to describe the mutual effects of stimuli that are combined together in a mixture. Spence (2011) discussed these various terms, including “multisensory perception,” “multisensory integration,” “cross-modal correspondence,” “synesthetic associations,” and “binding.” He concludes that the differences in terminology are related to the putative neural (or other) mechanisms underlying their effect. In the case of “context,” the word would seem to imply a purely psychological effect, because it treats the target stimulus as distinct and separate from its surrounding or temporally distinct contextual stimuli. However, many context effects, for example, visual contrast, are first assumed to be purely psychological in nature, until such time as neural mechanisms that underlie the phenomena are discovered, for example, lateral inhibition in the case of visual contrast.

3.2

Objectives of the chapter

In this chapter, “context” and “context effect” are taken to reflect general terms that refer to what happens when one stimulus element (e.g., ingredient, food, or beverage) is presented in simultaneous or temporal combination with other stimulus elements (other ingredients, foods, or beverages). The role of nonfood elements, for example, saliva, will not be discussed. Neither will the underlying mechanisms of interaction be belabored or discussed, primarily because the chapter is targeted to an audience that is interested in the wider variety of “contextual” effects in food, whether they be simultaneous or temporal, intrinsic or extrinsic, and whether at the level of the sip or bite, the food, the dish, or the meal. Conceptually, “context” is used here along the lines that Korzen and Lassen (2010) described: “as an interpretative frame that can be used as a basis upon which to pose relevant research questions or to interpret data.” The purpose of the chapter is to offer completeness to other chapters in this volume that examine how more macroscopic changes in the environment can alter the perceptions of both the sensory and hedonic characteristics of individual elements of food and meal situations.

3.3

Simultaneous contextual effects in taste and smell

3.3.1 Ingredient interactions As noted in the Introduction, the most elemental contextual effects that occur in foods and beverages are the result of the interactions among different food and beverage ingredients that comprise the food. As also noted, the perceived flavor of a food or beverage is the result of the integration of a vast array of sensory inputs from multiple sensory modalities (appearance, taste, smell, texture, audition, etc.), each evoked by one or more constituents of the product. Such integration is responsible for the often reported confusions of taste and smell, and the fact that the tastes and smells of foods and beverages produce an experience that is often more than the sum of the sensations

42

Context

arising from the individual modalities stimulated (Auvray & Spence, 2008; Rozin, 1982; Small & Prescott, 2005). Taste and smell psychophysicists, flavorists, and neurocognitive psychologists have known about and studied these mutual interactive effects for well over a century (see Delwiche, 2004; Small & Prescott, 2005, Verhagen & Engelen, 2006; Auvray & Spence, 2008; Spence, 2011; ThomasDanguin, Sinding, Tournier, & Saint-Eve, 2016; Stevenson, 2016; PiquerasFiszman & Spence, 2016 for recent reviews and books). In one such review, Delwiche (2004) summarized these various interactions graphically (Fig. 3.1). As can be seen, the taste, smell, temperature, texture, color, and even the irritation of food constituents can all interact to influence one or more other sensory modalities. In turn, by influencing perception, these different sensory-based contexts can also influence other evaluative aspects of the stimulus, for example, liking and other affective responses.

3.3.2 Taste-taste and odor-odor interactions Although not shown in Fig. 3.1, even two ingredients that evoke the same sensory quality can have mutual interactive influences. For example, combining two different high intensity sweeteners can produce enhancement (synergy) of their sweetness levels (Ayya & Lawless, 1992; Hyvonen, Kurkela, Koivistoinen, & Ratilainen, 1978; Moskowitz, 1973a, 1974; Schiffman et al., 1995). Similarly, combining two ingredients that elicit the umami taste, for example, MSG and 50 ribonucleotides, Temperature

Texture

Perceptual & physical interactions Perceptual interaction

Perceptual interactions

Taste

Cognitive Integration

Perceptual interactions

Color

Smell

Perceptual interactions

Irritation

Fig. 3.1 Summary of perceptual interactions evoked during ingestion. Arrowhead indicates a modality that has been demonstrated to interact with another modality (from Delwiche, 2004).

Context effects at the level of the sip and bite

43

can produce enhancement of this basic taste (Rifkin and Bartoshuk, 1980; Schiffman, Frey, Luboski, Foster, & Erickson, 1991). However, enhancement, or synergy, is most common only at low concentrations. At higher concentrations, enhancement is reduced, and “suppression” is more likely (Breslin & Beauchamp, 1995; Keast & Breslin, 2003), such as when NaCl and KCl are mixed (Breslin & Beauchamp, 1995). When different tastes are combined, such as when sugar is placed into coffee, suppression most often occurs, that is, both the sweetness of the sugar and the bitterness of the coffee are suppressed (see papers by Lawless, 1987; Kroeze, 1990; Breslin & Beauchamp, 1995; Keast & Breslin, 2003). However, the latter effect does not occur equally across different taste qualities or different concentrations. For example, Breslin (1996) examined the interactions that occur between bitter, salty, and sour tastes, and concluded that salty and sour tastes suppress each other at high concentrations, but enhance each other at lower concentrations. Similarly, bitter and sour tastes can either suppress or enhance one another, while salty and bitter tastes usually have the result of suppressing the bitterness, but leaving the salty taste unaffected (Breslin, 1996). Similar complex interactions of one taste presented simultaneously with another have been shown repeatedly, beginning with research summarized in the early review by Fabian and Blum (1943) and the early research of Anderson (1950), BeebeCenter, Rogers, Atkinson, and O’connell (1959), Kamen (1959), Kamen, Pilgrim, Gutman, and Kroll (1961), and Pangborn (1960a, 1960b, 1962), and continuing through to the present (Breslin & Beauchamp, 1995; Calvino, Garcı´a-Medina, & Cometto-Muniz, 1990; Diamond, Breslin, Doolittle, Nagata, & Dalton, 2005; Gillette, 1985; Keast, 2008; Keast, Dalton, & Breslin, 2004; Kemp & Beauchamp, 1994; Kroeze, 1990; Kroeze & Bartoshuk, 1985; Lawless, 1979, 1982; Schifferstein & Frijters, 1993; Schifferstein & Kleykers, 1996; Schiffman et al., 1994; Schiffman, Sattely-Miller, Graham, Booth, & Gibes, 2000; Stevens, 1995). With respect to odor-odor combinations, early research on two-component combinations showed that the mixture generally smells less intense than the sum of the intensities of the individual components, but more intense than the weaker of the two, that the individual components smell weaker than when unmixed, and that the stronger of the two components reduces the intensity of the weaker component more than the weaker one reduces the stronger (Cain, Schiet, Olsson, & de Wijk, 1995; Laing, Panhuber, Willcox, & Pittman, 1984; Lawless, 1997). With more complex mixtures of 3 or 4 components, adding additional odors has little impact on overall intensity (Moskowitz & Barbe, 1977). In addition, interactions depend, to some extent, on the overall intensity of the odors (Cain et al., 1995; Cometto-Mun˜iz, Cain, & Abraham, 2005; Laing et al., 1984; Laska, Hudson, & Distel, 1990). As with taste mixtures, odorant mixtures can be analyzed into their components. However, while taste qualities can be discerned in 5-6 component mixtures (Marshall, Laing, Jinks, & Hutchinson, 2006), complex odor mixtures often form holistic sensations, and even expert perfumers fail to correctly identify all of the constituents of mixtures that consist of more than about three to five chemicals (Livermore & Laing, 1998), supporting the fact that olfaction is a synthetic sense. Although most sensory research shows that the perceived qualities of odor mixtures generally fall between those of their unmixed components, with no radically new

44

Context

notes emerging, some research leaves the possibility of emergent qualities open ( Jinks & Laing, 2001; Moskowitz & Barbe, 1977; Zou & Buck, 2006). In addition, some odors tend to have an impact on the quality of mixtures disproportionate to their individual perceived intensities (Schifferstein & Kleykers, 1996). For example, woody odors tend to dominate fruity ones (Atanasova et al., 2005). In addition, it has been shown that the similarity of the constituent odors can make it more likely that the combination is perceived holistically versus elementally (Wiltrout, Dogra, & Linster, 2003).

3.3.3 Interactions between tastes and smells Perhaps the most ubiquitous contextual interactions in food are taste-odor interactions. Early research (Garcia-Medina, 1981; Gillan, 1983; Murphy & Cain, 1980) showed that the overall intensity of taste-odor mixtures are perceived to be close to a simple additive function of the individual components, and that increasing the intensity of either component increases the intensity of the other (see also research by Bonnans & Noble, 1993; Philipsen, Clydesdale, Griffin, & Stern, 1995; Salles, 2006). Enhancement or synergy between odors and taste has often been shown, for example, when such odors as strawberry or vanilla are paired with sweet tastes (Cliff & Noble, 1990; Frank & Byram, 1988; Frank, Ducheny, & Mize, 1989; Schifferstein & Verlegh, 1996; Stevenson, Prescott, & Boakes, 1999). However, sweetness suppression has also been reported (Linscott & Lim, 2016; Stevenson et al., 1999). The extent of enhancement has been shown to depend on the specific odor and taste combinations. For example, the combination of peanut odor with sucrose does not enhance sweet taste (Frank & Byram, 1988). Rather, mutual interactive effects occur most often with more qualitatively similar tastes and odors of those often paired in nature (Dalton, Doolittle, Nagata, & Breslin, 2000; Delwiche & Heffelfinger, 2005; Frank & Byram, 1988; Lim, Fujimaru, & Linscott, 2014; Linscott & Lim, 2016; Nguyen, Valentin, Ly, Chrea, & Sauvageot, 2002; Schifferstein & Verlegh, 1996; Stevenson et al., 1999). This phenomenon has been attributed to associative learning (Frank & Byram, 1988; Rolls, 2011; Stevenson, Boakes, & Prescott, 1998), leading some researchers to utilize this phenomenon to produce odor-induced enhancement of saltiness in order to reduce sodium content in foods (Lawrence, Salles, Septier, Busch, & Thomas-Danguin, 2009; Nasri, Septier, Beno, Salles, & Thomas-Danguin, 2013). In one example, the odor of sardines (a food associated with high saltiness) was shown to enhance the perceived saltiness of water to the equivalence of 20 mM NaCl. Clearly, the associative effects between odors and tastes can be quite pronounced, and can occur with only a single odor-taste pairing (Prescott, Johnstone, & Francis, 2004). It appears that much of our perception of novel foods and beverages results from a comparison of their taste/odor to cognitive elements in memory, that is, associations with previously encountered foods, tastes, odors, and so forth (Auvray & Spence, 2008; Stevenson & Boakes, 2004; Valentin, Chrea, & Nguyen, 2006), producing a form of “learned” synesthesia. As noted in the Introduction, Rose Marie Pangborn conducted a wide variety of studies examining how the experience of basic tastants were influenced by the media

Context effects at the level of the sip and bite

45

(often real foods and beverages) in which they were dispersed or dissolved, for example, sugar(s) in lemonade versus orange drinks (Pangborn, Marsh, Channell, & Campbell, 1960), salts and acids in water or green pea puree (Pangborn & Trabue, 1967), sugar(s) in vanilla or chocolate milk (Pangborn, 1980), sweet, salty, sour, and bitter in ethyl alcohol (Martin & Pangborn, 1970), salt in differing fat levels of sour cream dips (Pangborn, 1988), and salt, acid, and sugar in tomato soup (Pangborn & Chrisp, 1964). In addition to Pangborn’s early work, McBride and Johnson (1987) studied the enhancement of perceived lemon flavor with increasing levels of sucrose and citric acid in lemon juices, while Perng and McDaniel (1989) showed that sucrose enhanced the fruit flavor intensity of blackberry juice. Other investigators have shown that aspartame enhances and prolongs the fruitiness of fruit-flavored solutions more than does sucrose (e.g., Bonnans & Noble, 1993). A number of additional studies using food and beverage-related stimuli have also shown the strong interactions of taste and smell experience in real foods (Labbe, Damevin, Vaccher, Morgenegg, & Martin, 2006; Linscott & Lim, 2016). Although for taste combinations, most research points to the fact that the individual components are perceived separately, in keeping with the fact that taste is an “analytic” (as opposed to synthetic) sense, this is not necessarily the case for taste/odor mixtures. For example, many odors can induce perceptions associated with taste, for example, vanilla is often described as being sweet, as is banana odor (Burdach, Kroeze, & K€ oster, 1984; Stevenson et al., 1999). This effect was demonstrated in research on cocoa beverages by Labbe et al. (2006). These authors added either cacao or vanilla flavorants (odors) to the beverage, and found that the addition of cocoa increased the bitter taste of the beverage and decreased its sweetness, while adding vanilla increased sweetness. The fact that odors can induce taste sensations is most dramatically noticed with the removal of retronasal odors (e.g., by closing the nares), which has the result of making foods “tasteless.” Prescott et al. (2004) has shown that such cross-modal influences can be altered by manipulating the perceptual strategy that the consumer uses when making judgments, that is, through encouragement of a more “analytic” or “synthetic” perception of the odor-taste pairings. When a more analytic perceptual approach is facilitated, odor has far less influence on taste than when a more synthetic approach is facilitated (Prescott et al., 2004). Relevant reviews of taste/odor interactions and their physiological bases can be found in Small and Prescott (2005), Thomas-Danguin et al. (2014); and Stevenson, 2016). Of some special relevance to the important contextual influences of taste and smell on one another, it should be noted that Djordjevic, Zatorre, & Jones-Gotman, 2004 compared detection accuracy of sucrose in water during simultaneous presentation of real or imagined soy sauce and strawberry odors. The accuracy of detecting sweetness was found to be enhanced in the presence of actual strawberry odor, and depressed in the presence of actual soy sauce odor. However, almost identical results were found when subjects merely imagined these odors. This effect speaks to the importance of central mechanisms (memory, attention, etc.) on flavor perception and has important practical relevance for the use of “evoked contexts” that involve imagined scenarios employed in the laboratory in order to better mimic contextual influences that occur in real-life food and consumption situations (Hein, Hamid, Jaeger, & Delahunty, 2010).

46

3.4

Context

Contextual interactions of trigeminal (irritation) and somatosensory (texture/thermal) sensations with taste and smell

Although less studied than interactions between taste and odor, trigeminal sensations (irritation) and somatosensory sensations (tactile/thermal) can also alter our flavor experiences of foods. Of course, we must be careful to dissociate the “contextual” effects of these sensations from any direct stimulation of taste or smell receptors. That is, some tactile oral stimulants such as carbonation can act directly on taste receptors, that is, sour receptors (Chandrashekar et al., 2009). Similarly, the perception of fat in the mouth is often believed to be due to tactile perception, when in fact, it has been reported that fat can have a direct effect on taste receptors (Keast & Costanzo, 2015; Mattes, 2009). Such direct actions on taste must be differentiated from their crossmodal interactive effects. Several researchers have shown that a mouth rinse of either piperine or capsaicin reduces the perceived intensity of sweet, sour, salty, and bitter tastes (Gilmore & Green, 1993; Lawless, Rozin, & Shenker, 1985; Lawless & Stevens, 1984). Capsaicin has also been shown to reduce the perceived sweetness of sucrose solutions in both tomato soup and solutions having orange and vanilla flavors (Prescott, Allen, & Stephens, 1993; Prescott & Stevenson, 1995a, 1995b), although the reverse effect, that is, sucrose reducing the effect of capsaicin (pepper) burn (Nasrawi & Pangborn, 1989; Prescott et al., 1993; Stevens & Lawless, 1986), has not been found. When odorant compounds are mixed with irritant compounds, early work by Cain and Murphy (1980) showed that the nasal irritant, CO2, suppresses the perceived intensity of amyl butyrate odors. Moreover, by presenting the irritant to one nostril and the odor to the other, it was possible to show that the locus of interaction occurs centrally in the nervous system (Cain & Murphy, 1980). However, in the case of taste perception, Yau and McDaniel (1992) and Cometto-Mun˜iz, Garcı´a-Medina, Calvin˜o, and Noriega (1987), using the same irritant, found either no effects on taste, or only minor suppression, for example, for salty taste. Other research on trigeminal effects by Lawless and Stevens (1984) and Lawless et al. (1985) found both taste and odor suppression from pre-treatment of the mouth with capsaicin (temporal effect), but no effects on flavor identification, while studies of capsaicin in solution with tastants (simultaneous effect) failed to show suppression (Cowart, 1987), except at high concentrations (Prescott et al., 1993; Prescott & Stevenson, 1995a, 1995b). The reader is referred to Carstens et al. (2002) for a short review of other interactive effects within and among trigeminal sensations, and the underlying mechanisms that may account for these effects and their role in food preferences. The perceived texture of foods and beverages in the mouth can also influence perceived taste and odor. For example, increasing the viscosity (thickness) of a solution lowers its perceived taste and flavor intensity (Arabie & Moskowitz, 1971; Kokini, 1985, 1987; Pangborn, Gibbs, & Tassan, 1978), especially at higher viscosity levels (Cook, Hollowood, Linforth, & Taylor, 2003; Hollowood, Linforth, & Taylor, 2002). Furthermore, these effects do not seem to be simply due to a slowing of the diffusion of

Context effects at the level of the sip and bite

47

taste or odor molecules to the receptors (Cook et al., 2003), although both gelling agents and fat slow the release of volatile and non-volatile components (Brauss, Linforth, Cayeux, Harvey, & Taylor, 1999; Kinsella, 1990; Overbosch, Afterof, & Haring, 1991). Rather, the interactive effects of viscosity, taste, and odor appear also to be due to a form of physiological convergence in the nervous system (CerfDucastel, Van de Moortele, MacLeod, Le Bihan, & Faurion, 2001), establishing a true, psychological cross-modal effect (Bult, de Wijk, & Hummel, 2007; Kutter, Hanesch, Rauh, & Delgado, 2011). Another interesting example of the effect of somatosensory sensations on taste is found in the work of Mosca, van de Velde, Bult, van Boekel, and Stieger (2012), who showed that soft textures in the mouth produce higher ratings of sweetness. As far as the inverse interaction is concerned, for example, taste experience affecting viscosity or other somatosensory aspects of food experience, research by Christensen (1983) showed that increasing the sweet taste of solutions resulted in an increase in the experience of their perceived thickness or viscosity. Other tastes, for example, salts and acids, have also been shown to increase the perceived viscosity of oil in water emulsions (Martin & Mela, 1994). In addition, more recent food-related research has provided support to the notion that chemosensory stimuli, for example, creamy odors (and especially retro-nasal odors), can increase the perceived thickness of foods and beverages in the mouth (Bult et al., 2007; Sundqvist, Stevenson, & Bishop, 2006; Tournier et al., 2009; Weenen, Jellema, & De Wijk, 2005). The reader is referred to a recent review of cross-modal tactile-taste interactions in foods by Slocombe, Carmichael, and Simner (2016). The temperature of foods, beverages, and tastants/odorants can also have an influence on perceived taste and odor (Bartoshuk, Rennert, Rodin, & Stevens, 1982; Calvin˜o, 1986; McBurney, Collings, & Glanz, 1973; Moskowitz, 1973b). The mechanism for this effect is often a physical one, that is, that cooler temperatures slow the release of volatile components in solution (e.g., ice cream is sweeter and more flavorful when it melts), while warmer temperature produces an increase in the release of volatiles, so that odors are more likely to be detected and more intense at higher temperatures, (e.g., warm beer is more bitter than cold beer) [see Olson, Caporaso, and Mandigo (1980) and Voirol and Daget (1989) for the case of meats]. However, it has also been demonstrated that direct cooling of the tongue can reduce the sweetness and bitterness of sucrose and caffeine (Green & Frankmann, 1987), an effect that is independent of the molecular activity of the tastant solution. In general, in these studies, the maximum sensitivity for chemosensory sensitivity occurs around room temperature. With regard to hedonics and other emotional responses, several studies (Cardello & Maller, 1982; Lester & Kramer, 1991; Pangborn, Chrisp, & Bertolero, 1970; Pramudya & Seo, 2018; Zellner, Stewart, Rozin, & Brown, 1988) have shown differences in the acceptability of foods, beverages, and taste solutions as a function of their temperature. Here, it has been found that the most acceptable temperatures for foods are those at which they are normally served and eaten (lasagna is preferred hot, carbonated beverages are preferred cold, and coffee is preferred either hot or ice cold, but not at room temperature), suggesting a learned association. Interestingly, it has been shown that warm ambient temperatures have a positive influence on most

48

Context

product assessments through a proposed activation of the concept of emotional warmth that then becomes associated with the test product (Zwebner, Lee, & Goldenberg, 2014). Whether warm foods in the mouth induce a similar phenomenon is an interesting issue for further study. With regard to trigeminal sensations (irritation, pungency), it has been shown that increases in temperature increase the perceived intensity of irritant stimuli (e.g., capsaicin), while cooling reduces it (Green, 1986), although the effects for CO2 irritation appear to evoke the opposite effects (Green, 1992). In a parallel manner, increases in irritant intensity have been shown to increase the perceived temperature of a food (Green, 1986). The role of somatosensory experience in multimodal food and beverage perception has recently been reviewed by Spence and Piqueras-Fiszman (2016).

3.5

Contextual interactions of color and chemosensory perception

The visual aspects of food are important drivers of both our sensory experiences of foods and beverages and our hedonic appreciation of them. The primary visual attribute that has been investigated for interactive effects with chemosensory experience is color. Although certain interactions among different chemosensory sensations within a solution or food matrix can be attributed to peripheral physiological interactions, the effect of color on taste and odor perception is a purely central phenomenon. Many mutual color-odor correspondences have been identified, for example, the color brown and the odors of caramel/chocolate, or the color red and the odor of strawberry (Dematte, Sanabria, & Spence, 2006; Dematte`, Sanabria, & Spence, 2008; Gilbert, Martin, & Kemp, 1996; Schifferstein & Tanudjaja, 2004). Although many of these correspondences are common in the population, many have been shown to be culturally-dependent (Shankar, Levitan, & Spence, 2010). Some of the earliest research on the role of color on taste/flavor was conducted by Moir (1936) and Duncker (1939), who showed a strong association between the color of a food and an individual’s ability to identify it. Since that time, many other studies have shown significant effects of color on taste recognition and/or taste intensity (e.g., Christensen, 1983; Johnson, Dzendolet, & Clydesdale, 1983; Kostyla, Clydesdale, & McDaniel, 1978; Maga, 1974; Pangborn, 1960a, 1960b; Roth, Radle, Gifford, & Clydesdale, 1988; Zellner & Durlach, 2003) and on flavor detection and identification (DuBose, Cardello, & Maller, 1980; Ferna´ndez-Va´zquez et al., 2013; Garber Jr, Hyatt, & Starr Jr, 2000; Lavin & Lawless, 1998; Philipsen et al., 1995; Teerling, 1992; Urbanyi, 1982; Zampini, Sanabria, Phillips, & Spence, 2007; Zampini, Wantling, Phillips, & Spence, 2008; Zellner, Bartoli, & Eckard, 1991). In a careful review of this literature, Spence, Levitan, Shankar, and Zampini (2010) have documented that the effect of color on taste identification appears to be more robust than on taste intensity. Similar effects have been demonstrated on odor/flavor identification and discrimination when common odors/flavors in solution /foods were either colored, uncolored, or inappropriately colored (Blackwell, 1995; Garber Jr et al., 2000;

Context effects at the level of the sip and bite

49

Levitan, Zampini, Li, & Spence, 2008; Morrot, Brochet, & Dubourdieu, 2001; Parr, Geoffrey White, & Heatherbell, 2003; Stevenson & Oaten, 2008; Zampini et al., 2007, 2008; Zellner et al., 1991; Zellner & Whitten, 1999). In one such study (Morrot et al., 2001), simply adding red color to white wine was sufficient to induce a red wine flavor into the white wine. With regard to odor/ flavor intensity, it has been shown that increases in the color intensity of the stimulus increases taste/odor/flavor intensity judgments (Bayarri, Calvo, Costell, & Dura´n, 2001; Chan & Kane-Martinelli, 1997; Clydesdale, Gover, & Fugardi, 1992; Johnson et al., 1983; Johnson, Dzendolet, Damon, Sawyer, & Clydesdale, 1982). In general, color-flavor interactions are dependent upon the specific color-odor pairing, with colors that are psychologically congruent with the food enhancing flavor perception (and liking) (Bayarri et al., 2001; Garber Jr et al., 2000; Zampini et al., 2007, 2008). For example, a red color will increase perceptions of strawberry flavor more than a green color (see Zellner and Kautz (1990) for an exception). Shankar et al. (2010) have argued that many color-flavor interactions are not strictly due to multisensory integration, but are mediated by higher level cognitive processes, such as expectations. Expectations have been implicated in a wide variety of extrinsic effects on sensory and hedonic experiences, and the reader is referred to reviews of these effects on food and beverage experience (Cardello, 1994, 2007; Piqueras-Fiszman & Spence, 2015). As a final point, from a mechanistic point of view, the effects of color on odor/flavor judgments may well be dependent upon whether the stimulus is perceived orthonasally or retronasally, with orthonasal judgments resulting in increases in the perceived intensity of the odors, but retronasal judgments resulting in decreases (Koza, Cilmi, Dolese, & Zellner, 2005). Recently, Zellner (2013) has proposed a perceptual model of color-odor effects that accounts for a wide variety of findings in the literature of the effects of color on odor identification, discrimination, intensity, and even hedonics, and Spence (2015a) has provided a comprehensive review of the effects of color on food perception and liking.

3.6

Contextual interactions of sound and chemosensory perception

A number of studies have demonstrated the influence of sound and hearing on food perception in the mouth. Much of this research comes from the studies by Vickers on the perception of crispness (Christensen & Vickers, 1981; Vickers, 1985, 1991; Vickers & Bourne, 1976a, 1976b), in which it was shown that the sound made by crisp food being eaten contributes significantly to the perception of the food’s crispness. Other studies have shown similar effects of the relationship of sound to food perception (Dematte` et al., 2014; Duizer, 2001; Luyten, Plijter, & Van Vliet, 2004; Masuda & Okajima, 2011; Spence & Zampini, 2006; Szczesniak, 1988; Verhagen & Engelen, 2006; Zampini & Spence, 2004, 2005). A recent and interesting report in Bacci and Melcher (2011) cited an experiment conducted by Charles Spence and Heston Blumenthal in which it was shown that the sound of frying bacon increased

50

Context

the perceived intensity of bacon flavor in egg flavored ice cream, and recent reviews of this area by Spence and Zampini (2006), Spence and Shankar (2010) and Spence (2015b) have delineated the important role of auditory context in food and beverage perception.

3.7

Temporal contextual effects in chemosensory perception

All of the contextual and interaction effects outlined above involve the effect of one sensory stimulus, or attribute on another when these stimuli/attributes occur simultaneously. However, context effects also occur when stimuli are temporally separated. For example, the same sensory stimulus, when presented following a more intense stimulus, or together in a series of higher intensity stimuli, will be perceived as less intense than when that same stimulus is presented following, or within, a series of less intense stimuli. This is commonly referred to as “contextual contrast,” that is, the standard stimulus is contrasted with the greater or less intense stimuli in the temporal series. Such contrast effects are commonly reported with taste stimuli (Conner, Land, & Booth, 1987; Hallowell, Parikh, Veldhuizen, & Marks, 2016; Lawless, 1983; Lawless, Horne, & Spiers, 2000; Lee & O’Mahony, 2007; Marks, Shepard, Burger, & Chakwin, 2012; Rankin & Marks, 1991; Riskey, 1982; Schifferstein & Frijters, 1992; Schifferstein & Oudejans, 1996), odor stimuli (Nakano & AyabeKanamura, 2017; Pol, Hijman, Baare, & van Ree, 1998), auditory stimuli (Marks and Warner, 1991; Arieh & Marks, 2003, 2011), and visual stimuli (Arieh and Marks (2002). In addition, this effect can occur with extremely long intervals between the standard and contextual stimuli, that is, up to 25 minutes in the case of odor (Pol et al., 1998), and even across test sessions in the case of sucrose intensity (Vollmecke, 1987), capsaicin burn (Stevenson & Prescott, 1994), and the flavor of raspberry cordials (Walter & Boakes, 2009). A related contrast effect, known as condensation, has also been demonstrated, in which less intense stimuli within a context of more intense stimuli are less discriminable from one another (Parker, Murphy, & Schneider, 2002; Ward, Armstrong, & Golestani, 1996). Contrast effects are often explained in terms of adaptation-level theory (Helson, 1964), which postulates that each stimulus in a series is compared relative to an average of all of the stimuli in the contextual range. Thus, a control stimulus presented within a range of high-intensity stimuli will be judged lower than when the same stimulus is presented within a range of lower-intensity stimuli. Alternatively, contrast effects can be explained through range-frequency theory (Parducci, 1963, 1983), which postulates that such effects result from a combination of subjects dividing the entire stimulus range into a finite set of equal subjective intervals (range principle), and subjects assigning the same number of stimuli to each perceptual category (frequency principle). However, contrast effects are not the only type of temporal contextual effects that have been reported. In certain cases, assimilation has been reported, that is, the standard stimulus is perceived as being more intense when presented within a series of greater intensity stimuli than when presented

Context effects at the level of the sip and bite

51

within a series of less intense stimuli. Assimilation effects are relatively rare in taste, but have been reported (Schifferstein & Kuiper, 1997). However, assimilation effects are common in other sensory modalities, such as in hearing and vision (Stewart, Brown, & Chater, 2005; Ward, 1979, 1982, 1985). Assimilation effects can be explained using integration theory (Anderson, 1981), which states that the rating for any stimulus is a weighted average of its own value, and the value of all other stimuli in its contextual range. One important characteristic of the relationship between the standard stimulus and the contextual stimuli that seems to influence the likelihood of the occurrence of a context effect and whether or not contrast or assimilation occurs is the similarity between the standard stimulus and the contextual stimuli (Coren & Enns, 1993; Marks & Warner, 1991; Rankin & Marks, 1991). To the extent that the standard stimulus is perceived to be similar to, a part of, or within the same category as the contextual series, a contrast effect is more likely to occur (Marks & Warner, 1991; Schifferstein, 1995a, 1995b; Schifferstein & Oudejans, 1996). To the extent that the standard stimulus is different from, not perceived as part of, or in a different category as the contextual stimuli, this effect is less likely to occur. Before concluding this section, it is worth mentioning another form of temporal contextual effect. This is the phenomenon of perceptual priming. For example, in the case of odors, it has been shown that presenting a series of food odors to subjects will predispose the subjects to more quickly perceive those same odors as being associated with foods (Koenig, Bourron, & Royet, 2000). This effect is presumed to occur by the first stimulus activating neural patterns that are stored in neural memory and are then more readily activated when the same stimulus is presented a second time. The interested reader is referred to Smeets and Dijksterhuis (2014) and Dijksterhuis (2016) for a detailed discussion of multisensory flavor priming.

3.8

Contextual effects on hedonics

3.8.1 Simultaneous effects Most research on contextual effects in the hedonic domain have examined the role of extrinsic variables on liking (see Cardello and Meiselman (2018) for a recent review). The reason for this focus on extrinsic effects is that intrinsic effects at the level of the bite or sip are too numerous to quantify, because every combination of two or more sensory stimuli within a food or beverage matrix will produce changes in the overall sensory profile of the product, which will, in turn, affect its hedonic character. That is, one could combine two well-liked chemosensory stimuli/ingredients, for example, a strawberry odor and a salty taste, or a pork flavor and a fish flavor, but produce a flavor combination that is disliked; or you might combine two disliked sensations, for example, the bitterness of coffee and the thick mouthfeel of whole cream to produce a well-liked combination. As such, determining specific combinatorial rules in hedonics is highly dependent on the stimulus elements, the matrices in which they appear, and, of course, the individual, because liking/disliking is heavily determined by past experience.

52

Context

This said, many researchers have examined how liking/disliking of specific chemosensory stimuli or ingredients are affected by being paired with other chemosensory stimuli or ingredients. For example, early research by Pangborn et al. (1960) showed differences in preferences for sweetener concentrations when the sweetener was added to either lemonade or orange juice drink. Moskowitz and Klarman (1975) looked at pairwise mixtures of artificial sweeteners and found that hedonic ratings did not add arithmetically, but that the unpleasantness of the mixtures was often greater than the unpleasantness of the components. Lawless (1977) found that the pleasantness of both taste and smell mixtures could be predicted by a weighted additive model of their individual pleasantness ratings. When looking at taste-odor combinations, von Sydow, Moskowitz, Jacobs, and Meiselman (1974) demonstrated that adding sugar to fruit juice increased pleasant odors from the juice and decreased unpleasant odors. In other research, Frank and Archambo (1986) found that the sweetness of sucrose tended to eliminate the unpleasantness of high citric acid concentrations, but not the unpleasantness of high sodium chloride concentrations, concluding that the hedonic integration of liking and disliking from two separate taste components is related to the underlying taste suppression that occurs in those components. This conclusion is consistent with a broader conceptualization of all hedonic combinations, that is, that the pleasantness of a stimulus may be different when that stimulus is combined/mixed with other sensory stimuli, and that the resultant hedonic character will be dependent upon the specific sensory interactions that occur among the stimuli.

3.8.2 Temporal effects Perhaps more useful for the purpose of developing general principles of the role of context in hedonic judgments at the level of the bite or sip are studies that examine temporal contextual effects in hedonics. Contextual effects in hedonics have been known since the time of Fechner (1898), who formulated his Law of Hedonic Contrast to account for situations in which test stimuli are better liked when presented within the context of less-liked stimuli (positive hedonic contrast), and less well liked when presented within the context of better-liked stimuli (negative hedonic contrast): “that which gives pleasure gives more pleasure the more it enters into contrast with sources of displeasure and a corresponding proposition holds for that which gives displeasure (Beebe-Center, 1932—reprinted 1965, p. 222).This maxim was confirmed early in the food literature by Kamenetzy (1959), who showed that the preference for poor quality foods was reduced when preceded by good quality food versus when preceded by other poor quality foods. Similarly, Riskey, Parducci, and Beauchamp (1979) demonstrated that, at low concentrations, the pleasantness of soft drinks were greater when presented within a series of predominantly lower sweetness (less pleasant) drinks; however, at higher concentrations, where greater sweetness is associated with lower pleasantness, pleasantness was shown to be lower within this same series (i.e., judged less pleasant within a context of better liked stimuli). Other, similar context effects on affect have been reported by McBride (1982) for sweetened milk products, and for artificial orange drinks (McBride, 1985), by Schifferstein (1995a, 1995b) for tastant

Context effects at the level of the sip and bite

53

solutions embedded in a series of predominantly bitter versus sweet contextual stimuli, and by Cardello, Melnick, and Rowan (1996) for sweetened orange drinks. Perhaps the greatest amount of research on this topic in recent years has been done by Zellner and colleagues. In one study that has illuminated important underlying dimensions of these effects, Zellner et al. (2003, Exp. 1) had two groups of consumers rate their liking for a series of 10 different fruit-based beverages. For one of these groups, all 10 samples were presented simultaneously. The first eight stimuli were standard versions of the fruit beverages, while the last two were diluted versions. The consumers in this group rated each one sequentially for liking. In the second group, only the first eight (of 10) beverages were presented for rating. However, after the eighth sample, the last two test samples were presented, but with the consumers being told that these last two were “commercial drinks being tested for another company.” Two control groups (Groups 3 and 4) received only the two diluted samples, and were given the same information about the samples (“juices” vs. “commercial drink for another company”) as the consumers in Groups 1 and 2. Results showed that consumers in the first group who received the eight full strength (better tasting) beverages first, followed by the diluted beverages, rated their liking for the last two “juices” much lower than the consumers who rated these two samples alone (Group 3), thus demonstrating a strong hedonic contrast effect (the first eight better-tasting samples depressed the liking ratings for the two diluted samples). However, this effect was severely diminished in Group 2, who were told that the last two samples were not part of the same set of stimuli. Based on this research and a series of related studies using a wide range of chemosensory and other hedonic stimuli (Forsythe, Zellner, Cogan, & Parker, 2014; Zellner et al., 2010; Zellner, Allen, Henley, & Parker, 2006; Zellner, Hoer, & Feldman, 2014; Zellner, Kern, & Parker, 2002; Zellner, Mattingly, & Parker, 2009), Zellner has demonstrated (1) that hedonic contrasts effects are dependent upon the degree to which the target stimulus is perceived to be a part of the contextual stimuli (a counterpart to similar findings for sensory context—see Section 3.6), (2) that preferences for one sample over another are reduced following the presentation of more preferred samples (hedonic condensation), but (3) these same preferences are increased following a series of disliked samples (hedonic expansion). In her most recent studies, contrast effects have been demonstrated in real-world eating situations. In particular, it was demonstrated that a good versus bad appetizer can produce a contrast effect on the main entree (Lahne & Zellner, 2015), but that such effects only occur when foods are actually eaten, not when foods are presented only visually ( Jimenez et al., 2015).

3.9

Conclusions

This chapter has highlighted the many different types of contextual interactions that can occur among different sensory components that comprise a unitary food, beverage, or other stimulus placed in the mouth during a single bite or sip. The bases of these effects can be both psychological as well as physiological, and in some case, may be purely physical. Nevertheless, they all underscore the importance that one stimulus

54

Context

component has on another, whether they are presented simultaneously or sequentially. While it may be debated whether some of these effects should truly be described as “contextual” in nature, what is not in debate is that the effect of any single component in a food or beverage cannot be treated in an absolute manner for its effect on sensory experience. Rather, the experience of any component of a bite of food or sip of beverage will depend on the sensory experiences arising from other sensory elements within that bite or sip, or from those bites or sips preceding and/or following it. The sensory experience of foods and beverages is not a simple sum of the individual sensory experiences arising from its varied components, but is a holistic interpretation that is dependent on the contribution of all of the components together. With regard to general principles of these contextual effects, the literature shows that the effects on sensory experience are dependent on the specific combinations of tastes, smells, somesthetic, or other sensations being combined, with some general rules being established for when suppression or enhancement occurs. In addition, it is clear that sensations that are congruent, similar, or associated with one another in memory, have a greater likelihood to interact with one another. This is especially true for temporal contextual effects, where similarity of the stimuli or their perceptual attributes is a well-established requirement in order for a contextual effect to occur. Product developers and others interested in how one sensory stimulus or sensation will be experienced when combined with another sensory stimulus or sensation, would do well to carefully study the available literature on the specific combinations under consideration, in order to best enable a prediction of the contextual effects that may result. In this way, it will be possible to make holistic predictions and to make sense out of the experiential coalition that, otherwise, would be what William James (1890) described as the “blooming, buzzing confusion” of individual sensations.

References Anderson, N. H. (1981). Foundations of information integration theory. New York: Academic. Anderson, R. J. (1950). Taste thresholds in stimulus mixtures. (Doctoral dissertation). Arabie, P., & Moskowitz, H. R. (1971). The effects of viscosity upon perceived sweetness. Attention, Perception, & Psychophysics, 9(5), 410–412. Arieh, Y., & Marks, L. E. (2002). Context effects in visual length perception: Role of ocular, retinal, and spatial location. Attention, Perception, & Psychophysics, 64(3), 478–492. Arieh, Y., & Marks, L. E. (2003). Recalibrating the auditory system: A speed-accuracy analysis of intensity perception. Journal of Experimental Psychology: Human Perception and Performance, 29(3), 523. Arieh, Y., & Marks, L. E. (2011). Measurement of loudness, Part II: Context effects. In Loudness (pp. 57–87). New York: Springer. Atanasova, B., Thomas-Danguin, T., Chabanet, C., Langlois, D., Nicklaus, S., & Etievant, P. (2005). Perceptual interactions in odour mixtures: Odour quality in binary mixtures of woody and fruity wine odorants. Chemical Senses, 30(3), 209–217. Auvray, M., & Spence, C. (2008). The multisensory perception of flavor. Consciousness and Cognition, 17(3), 1016–1031. Ayya, N., & Lawless, H. T. (1992). Quantitative and qualitative evaluation of high-intensity sweeteners and sweetener mixtures. Chemical Senses, 17(3), 245–259.

Context effects at the level of the sip and bite

55

Bacci, F., & Melcher, D. (2011). Art and the senses. Oxford, UK: Oxford University Press. Bartoshuk, L. M., Rennert, K., Rodin, J., & Stevens, J. C. (1982). Effects of temperature on the perceived sweetness of sucrose. Physiology & Behavior, 28(5), 905–910. Bayarri, S., Calvo, C., Costell, E., & Dura´n, L. (2001). Influence of color on perception of sweetness and fruit flavor of fruit drinks. Revista de Agaroquimica y Tecnologia de Alimentos, 7(5), 399–404. Beebe-Center, J. G. (1932). The Psychology of pleasantness and unpleasantness, reprinted. Russell & Russell 1965. Beebe-Center, J. G., Rogers, M. S., Atkinson, W. H., & O’connell, D. N. (1959). Sweetness and saltiness of compound solutions of sucrose and NaCl as a function of concentration of solutes. Journal of Experimental Psychology, 57(4), 231. Blackwell, L. (1995). Visual cues and their effects on odour assessment. Nutrition & Food Science, 95(5), 24–28. Bonnans, S., & Noble, A. C. (1993). Effect of sweetener type and of sweetener and acid levels on temporal perception of sweetness, sourness and fruitiness. Chemical Senses, 18(3), 273–283. Brauss, M. S., Linforth, R. S., Cayeux, I., Harvey, B., & Taylor, A. J. (1999). Altering the fat content affects flavor release in a model yogurt system. Journal of Agricultural and Food Chemistry, 47(5), 2055–2059. Breslin, P. A. (1996). Interactions among salty, sour and bitter compounds. Trends in Food Science & Technology, 7(12), 390–399. Breslin, P. A. S., & Beauchamp, G. K. (1995). Suppression of bitterness by sodium: variation among bitter taste stimuli. Chemical Senses, 20(6), 609–623. Bult, J. H., de Wijk, R. A., & Hummel, T. (2007). Investigations on multimodal sensory integration: texture, taste, and ortho-and retronasal olfactory stimuli in concert. Neuroscience Letters, 411(1), 6–10. Burdach, K. J., Kroeze, J. H., & K€oster, E. P. (1984). Nasal, retronasal, and gustatory perception: An experimental comparison. Perception & Psychophysics, 36(3), 205–208. Cain, W. S., & Murphy, C. L. (1980). Interaction between chemoreceptive modalities of odour and irritation. Nature, 284(5753), 255–257. Cain, W. S., Schiet, F. T., Olsson, M. J., & de Wijk, R. A. (1995). Comparison of models of odor interaction. Chemical Senses, 20(6), 625–637. Calvin˜o, A. M. (1986). Perception of sweetness: The effects of concentration and temperature. Physiology & Behavior, 36(6), 1021–1028. Calvino, A. M., Garcı´a-Medina, M. R., & Cometto-Muniz, J. E. (1990). Interactions in caffeine– sucrose and coffee–sucrose mixtures: Evidence of taste and flavor suppression. Chemical Senses, 15(5), 505–519. Cardello, A. V. (1994). Consumer expectations and their role in food acceptance. In Measurement of food preferences (pp. 253–297). Boston, MA: Springer. Cardello, A. V. (2007). Measuring consumer expectations to improve food product development. In Consumer-led food product development (pp. 223–261). Cardello, A. V., & Maller, O. (1982). Acceptability of water, selected beverages and foods as a function of serving temperature. Journal of Food Science, 47(5), 1549–1552. Cardello, A. V., & Meiselman, H. L. (2018). Contextual influences on consumer responses to food products. In Methods in consumer research, Vol. 2: Alternative approaches and special applications (pp. 3–54). Kidlington, UK: Woodhead Publishing Series in Food Science, Technology and Nutrition. Cardello, A. V., Melnick, S. M., & Rowan, P. A. (1996). Expectations as a mediating variable in context effects. In: Proceedings of the food preservation 2000 conference, vol. 1 (pp. 259–309). Hampton, VA: Science & Technology Corporation, pp. 259–309.

56

Context

Carstens, E., Carstens, M. I., Dessirier, J. M., O’Mahony, M., Simons, C. T., Sudo, M., et al. (2002). It hurts so good: Oral irritation by spices and carbonated drinks and the underlying neural mechanisms. Food Quality and Preference, 13(7), 431–443. Cerf-Ducastel, B., Van de Moortele, P. F., MacLeod, P., Le Bihan, D., & Faurion, A. (2001). Interaction of gustatory and lingual somatosensory perceptions at the cortical level in the human: A functional magnetic resonance imaging study. Chemical Senses, 26(4), 371–383. Chan, M. M., & Kane-Martinelli, C. (1997). The effect of color on perceived flavor intensity and acceptance of foods by young adults and elderly adults. Journal of the American Dietetic Association, 97(6), 657–659. Chandrashekar, J., Yarmolinsky, D., von Buchholtz, L., Oka, Y., Sly, W., Ryba, N. J., et al. (2009). The taste of carbonation. Science, 326(5951), 443–445. Christensen, C. M. (1983). Effects of color on aroma, flavor and texture judgments of foods. Journal of Food Science, 48(3), 787–790. Christensen, C. M., & Vickers, Z. M. (1981). Relationships of chewing sounds to judgments of food crispness. Journal of Food Science, 46(2), 574–578. Cliff, M., & Noble, A. C. (1990). Time-intensity evaluation of sweetness and fruitiness and their Interaction in a model solution. Journal of Food Science, 55(2), 450–454. Clydesdale, F. M., Gover, R., & Fugardi, C. (1992). The effect of color on thirst quenching, sweetness, acceptability and flavor intensity in fruit punch flavored beverages. Journal of Food Quality, 15(1), 19–38. Cometto-Mun˜iz, J. E., Cain, W. S., & Abraham, M. H. (2005). Odor detection of single chemicals and binary mixtures. Behavioural Brain Research, 156(1), 115–123. Cometto-Mun˜iz, J. E., Garcı´a-Medina, M. R., Calvin˜o, A. M., & Noriega, G. (1987). Interactions between CO2 oral pungency and taste. Perception, 16(5), 629–640. Conner, M. T., Land, D. G., & Booth, D. A. (1987). Effect of stimulus range on judgements of sweetness intensity in a lime drink. British Journal of Psychology, 78(3), 357–364. Cook, D. J., Hollowood, T. A., Linforth, R. S., & Taylor, A. J. (2003). Oral shear stress predicts flavour perception in viscous solutions. Chemical Senses, 28(1), 11–23. Coren, S., & Enns, J. T. (1993). Size contrast as a function of conceptual similarity between test and inducers. Attention, Perception, & Psychophysics, 54(5), 579–588. Cowart, B. J. (1987). Oral chemical irritation: does it reduce perceived taste intensity? Chemical Senses, 12(3), 467–479. Dalton, P., Doolittle, N., Nagata, H., & Breslin, P. A. S. (2000). The merging of the senses: Integration of subthreshold taste and smell. Nature Neuroscience, 3(5), 431–432. Delwiche, J. (2004). The impact of perceptual interactions on perceived flavor. Food Quality and Preference, 15(2), 137–146. Delwiche, J. F., & Heffelfinger, A. L. (2005). Cross-modal additivity of taste and smell. Journal of Sensory Studies, 20(6), 512–525. Dematte`, M. L., Pojer, N., Endrizzi, I., Corollaro, M. L., Betta, E., Aprea, E., et al. (2014). Effects of the sound of the bite on apple perceived crispness and hardness. Food Quality and Preference, 38, 58–64. Dematte, M. L., Sanabria, D., & Spence, C. (2006). Cross-modal associations between odors and colors. Chemical Senses, 31(6), 531–538. Dematte`, M. L., Sanabria, D., & Spence, C. (2008). Olfactory discrimination: When vision matters? Chemical Senses, 34(2), 103–109. Diamond, J., Breslin, P. A., Doolittle, N., Nagata, H., & Dalton, P. (2005). Flavor processing: perceptual and cognitive factors in multi-modal integration. Chemical Senses, 30 (suppl 1), i232–i233.

Context effects at the level of the sip and bite

57

Dijksterhuis, G. (2016). Multisensory flavor priming. In Multisensory flavor perception (pp. 133–153). Woodhead Publishing. Djordjevic, J., Zatorre, R. J., & Jones-Gotman, M. (2004). Odor-induced changes in taste perception. Experimental Brain Research, 159(3), 405–408. DuBose, C. N., Cardello, A. V., & Maller, O. (1980). Effects of colorants and flavorants on identification, perceived flavor intensity, and hedonic quality of fruit-flavored beverages and cake. Journal of Food Science, 45(5), 1393–1399. Duizer, L. (2001). A review of acoustic research for studying the sensory perception of crisp, crunchy and crackly textures. Trends in Food Science & Technology, 12(1), 17–24. Duncker, K. (1939). The influence of past experience upon perceptual properties. The American Journal of Psychology, 52(2), 255–265. Fabian, F. W., & Blum, H. B. (1943). Relative taste potency of some basic food constituents and their competitive and compensatory action. Journal of Food Science, 8(3), 179–193. Fechner, G. T. (1898). (2nd ed.). Vorschule der Aesthetik II Leipzig: Breitkopf & Ha¨rtel. Ferna´ndez-Va´zquez, R., Hewson, L., Fisk, I., Vila, D. H., Mira, F. J. H., Vicario, I. M., et al. (2013). Colour influences sensory perception and liking of orange juice. Flavour, 3(1), 1. Forsythe, M., Zellner, D., Cogan, E., & Parker, S. (2014). Attractiveness difference magnitude affected by context, range, and categorization. Perception, 43(1), 59–69. Frank, R. A., & Archambo, G. (1986). Intensity and hedonic judgments of taste mixtures: an information integration analysis. Chemical Senses, 11(4), 427–438. Frank, R. A., & Byram, J. (1988). Taste–smell interactions are tastant and odorant dependent. Chemical Senses, 13(3), 445–455. Frank, R. A., Ducheny, K., & Mize, S. J. S. (1989). Strawberry odor, but not red color, enhances the sweetness of sucrose solutions. Chemical Senses, 14(3), 371–377. Garber, L. L., Jr., Hyatt, E. M., & Starr, R. G., Jr. (2000). The effects of food color on perceived flavor. Journal of Marketing Theory and Practice, 8(4), 59–72. Garcia-Medina, M. R. (1981). Flavor-odor taste interactions in solutions of acetic acid and coffee. Chemical Senses, 6(1), 13–22. Gilbert, A. N., Martin, R., & Kemp, S. E. (1996). Cross-modal correspondence between vision and olfaction: The color of smells. The American Journal of Psychology, 335–351. Gillan, D. J. (1983). Taste-taste, odor-odor, and taste-odor mixtures: Greater suppression within than between modalities. Perception & Psychophysics, 33(2), 183–185. Gillette, M. (1985). Flavor effects of sodium chloride. Food Technology, 39, 47–56. Gilmore, M. M., & Green, B. G. (1993). Sensory irritation and taste produced by NaCl and citric acid: effects of capsaicin desensitization. Chemical Senses, 18(3), 257–272. Green, B. G. (1986). Sensory interactions between capsaicin and temperature in the oral cavity. Chemical Senses, 11(3), 371–382. Green, B. G. (1992). The effects of temperature and concentration on the perceived intensity and quality of carbonation. Chemical Senses, 17(4), 435–450. Green, B. G., & Frankmann, S. P. (1987). The effect of cooling the tongue on the perceived intensity of taste. Chemical Senses, 12(4), 609–619. Hallowell, E. S., Parikh, R., Veldhuizen, M. G., & Marks, L. E. (2016). Flavor identification and intensity: Effects of stimulus context. Chemical Senses, 41(3), 249–259. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2010). Application of a written scenario to evoke a consumption context in a laboratory setting: Effects on hedonic ratings. Food Quality and Preference, 21(4), 410–416. Helson, H. (1964). Adaptation-level theory: An experimental and systematic approach to behavior. New York: Harper and Row.

58

Context

Hollowood, T. A., Linforth, R. S. T., & Taylor, A. J. (2002). The effect of viscosity on the perception of flavour. Chemical Senses, 27(7), 583–591. Hyde, R. J., & Witherly, S. A. (1993). Dynamic contrast: A sensory contribution to palatability. Appetite, 21(1), 1–16. Hyvonen, L., Kurkela, R., Koivistoinen, P., & Ratilainen, A. (1978). Fructose-saccharine and xylitol-saccharin synergism. Journal of Food Science, 43(1), 251–254. James, W. (1890). The Principles of psychology. 1. New York: Henry Holt. Jimenez, M., Rodriguez, D., Greene, N., Zellner, D. A., Cardello, A. V., & Nestrud, M. (2015). Seeing a meal is not eating it: Hedonic context effects differ for visually presented and actually eaten foods. Food Quality and Preference, 41, 96–102. Jinks, A., & Laing, D. G. (2001). The analysis of odor mixtures by humans: Evidence for a configurational process. Physiology & behavior, 72(1), 51–63. Johnson, J. L., Dzendolet, E., & Clydesdale, F. M. (1983). Psychophysical relationship between sweetness and redness in strawberry-flavored drinks. Journal of Food Protection, 46(1), 21–25. Johnson, J. L., Dzendolet, E., Damon, R., Sawyer, M., & Clydesdale, F. M. (1982). Psychophysical relationships between perceived sweetness and color in cherry-flavored beverages. Journal of Food Protection, 45(7), 601–606. Kamen, J. M. (1959). Interaction of sucrose and calcium cyclamate on perceived intensity of sweetness. Journal of Food Science, 24(3), 279–282. Kamen, J. M., Pilgrim, F. J., Gutman, N. J., & Kroll, B. J. (1961). Interactions of suprathereshold taste stimuli. Journal of Experimental Psychology, 62(4), 348. Kamenetzy, J. (1959). Contrast and convergence effects in ratings of foods. Journal of Applied Psychology, 43(1), 47–52. Keast, R. S. (2008). Modification of the bitterness of caffeine. Food Quality and Preference, 19 (5), 465–472. Keast, R. S., & Breslin, P. A. (2003). An overview of binary taste–taste interactions. Food Quality and Preference, 14(2), 111–124. Keast, R. S., & Costanzo, A. (2015). Is fat the sixth taste primary? Evidence and implications. Flavour, 4(1), 5. Keast, R. S., Dalton, P. H., & Breslin, P. A. (2004). Flavor interactions at the sensory level. Flavor Perception, 228–255. Kemp, S. E., & Beauchamp, G. K. (1994). Flavor modification by sodium chloride and monosodium glutamate. Journal of Food Science, 59(3), 682–686. Kinsella, J. E. (1990). Flavor perception and binding. International News on Fats, Oils and Related Materials (USA), 1, 215–226. Koenig, O., Bourron, G., & Royet, J. P. (2000). Evidence for separate perceptive and semantic memories for odours: A priming experiment. Chemical Senses, 25(6), 703–708. Kokini, J. L. (1985). Fluid and semi-solid food texture and texture-taste interactions. Food Technology, 39(11), 86–92 94. Kokini, J. L. (1987). The physical basis of liquid food texture and texture-taste interactions. Journal of Food Engineering, 6(1), 51–81. Korzen, S., & Lassen, J. (2010). Meat in context. On the relation between perceptions and contexts. Appetite, 54(2), 274–281. Kostyla, A. S., Clydesdale, F. M., & McDaniel, M. R. (1978). The psychophysical relationships between color and flavor∗. Critical Reviews in Food Science & Nutrition, 10(3), 303–321. Koza, B. J., Cilmi, A., Dolese, M., & Zellner, D. A. (2005). Color enhances orthonasal olfactory intensity and reduces retronasal olfactory intensity. Chemical Senses, 30(8), 643–649. Kroeze, J. H. (1990). The perception of complex taste stimuli. In Psychological Basis of sensory evaluation (pp. 41–68). London: Elsevier.

Context effects at the level of the sip and bite

59

Kroeze, J. H., & Bartoshuk, L. M. (1985). Bitterness suppression as revealed by split-tongue taste stimulation in humans. Physiology & Behavior, 35(5), 779–783. Kutter, A., Hanesch, C., Rauh, C., & Delgado, A. (2011). Impact of proprioception and tactile sensations in the mouth on the perceived thickness of semi-solid foods. Food Quality and Preference, 22(2), 193–197. Labbe, D., Damevin, L., Vaccher, C., Morgenegg, C., & Martin, N. (2006). Modulation of perceived taste by olfaction in familiar and unfamiliar beverages. Food Quality and Preference, 17(7), 582–589. Lahne, J. (2016). Sensory science, the food industry, and the objectification of taste. Anthropology of Food, 10, . Lahne, J., & Zellner, D. A. (2015). The great is the enemy of the good: Hedonic contrast in a coursed meal. Food Quality and Preference, 45, 70–74. Laing, D. G., Panhuber, H., Willcox, M. E., & Pittman, E. A. (1984). Quality and intensity of binary odor mixtures. Physiology & Behavior, 33(2), 309–319. Laska, M., Hudson, R., & Distel, H. (1990). Sensitivity to biologically relevant odours may exceed the sum of component thresholds. Chemoecology, 1(3), 139–141. Lavin, J. G., & Lawless, H. T. (1998). Effects of color and odor on judgments of sweetness among children and adults. Food Quality and Preference, 9(4), 283–289. Lawless, H. (1982). Adapting efficiency of salt-sucrose mixtures. Attention, Perception, & Psychophysics, 32(5), 419–422. Lawless, H. (1983). Contextual effects in category ratings. Journal of Testing and Evaluation, 11(5), 346–349. Lawless, H., Rozin, P., & Shenker, J. (1985). Effects of oral capsaicin on gustatory, olfactory and irritant sensations and flavor identification in humans who regularly or rarely consume chili pepper. Chemical Senses, 10(4), 579–589. Lawless, H., & Stevens, D. A. (1984). Effects of oral chemical irritation on taste. Physiology & Behavior, 32(6), 995–998. Lawless, H. T. (1977). The pleasantness of mixtures in taste and olfaction. Sensory Processes, 1, 227–237. Lawless, H. T. (1979). Evidence for neural inhibition in bittersweet taste mixtures. Journal of Comparative and Physiological Psychology, 93(3), 538. Lawless, H. T. (1987). Gustatory psychophysics. In Neurobiology of taste and smell (pp. 401– 420). New York: Wiley. Lawless, H. T. (1997). Olfactory psychophysics. In Tasting and smelling: Handbook of perception and cognition (pp. 125–174). London: Elsevier. Lawless, H. T., Horne, J., & Spiers, W. (2000). Contrast and range effects for category, magnitude and labeled magnitude scales in judgements of sweetness intensity. Chemical Senses, 25(1), 85–92. Lawrence, G., Salles, C., Septier, C., Busch, J., & Thomas-Danguin, T. (2009). Odour–taste interactions: A way to enhance saltiness in low-salt content solutions. Food Quality and Preference, 20(3), 241–248. Lee, H. S., & O’Mahony, M. (2007). Difference test sensitivity: Cognitive contrast effects. Journal of Sensory Studies, 22(1), 17–33. Lester, L. S., & Kramer, F. M. (1991). The effects of heating on food acceptability and consumption. Journal of Foodservice, 6(2), 69–87. Levitan, C. A., Zampini, M., Li, R., & Spence, C. (2008). Assessing the role of color cues and people’s beliefs about color–flavor associations on the discrimination of the flavor of sugar-coated chocolates. Chemical Senses, 33(5), 415–423. Lim, J., Fujimaru, T., & Linscott, T. D. (2014). The role of congruency in taste–odor interactions. Food Quality and Preference, 34, 5–13.

60

Context

Linscott, T. D., & Lim, J. (2016). Retronasal odor enhancement by salty and umami tastes. Food Quality and Preference, 48, 1–10. Livermore, A., & Laing, D. G. (1998). The influence of odor type on the discrimination and identification of odorants in multicomponent odor mixtures. Physiology & Behavior, 65 (2), 311–320. Luyten, H., Plijter, J., & Van Vliet, T. O. N. (2004). Crispy/crunchy crusts of cellular solid foods: A literature review with discussion. Journal of Texture Studies, 35(5), 445–492. Maga, J. A. (1974). Influence of color on taste thresholds. Chemical Senses, 1(1), 115–119. Marks, L. E., Shepard, T. G., Burger, K., & Chakwin, E. M. (2012). Flavor-intensity perception: Effects of stimulus context. Physiology & Behavior, 105(2), 443–450. Marks, L. E., & Warner, E. (1991). Slippery context effect and critical bands. Journal of Experimental Psychology: Human Perception and Performance, 17(4), 986–996. Marshall, K., Laing, D. G., Jinks, A. L., & Hutchinson, I. (2006). The capacity of humans to identify components in complex odor–taste mixtures. Chemical Senses, 31(6), 539–545. Martin, N. B. P. A., & Mela, D. J. (1994). Perception of taste and viscosity of oil-in-water and water-in-oil emulsions. Journal of Food Science, 59(6), 1318–1321. Martin, S., & Pangborn, R. M. (1970). Taste interaction of ethyl alcohol with sweet, salty, sour and bitter compounds. Journal of the Science of Food and Agriculture, 21(12), 653–655. Masuda, M., & Okajima, K. (2011). Added mastication sound affects food texture and pleasantness. i-Perception, 2(8), 949. Mattes, R. D. (2009). Is there a fatty acid taste? Annual Review of Nutrition, 29, 305–327. McBride, R. L. (1982). Range bias in sensory evaluation. International Journal of Food Science & Technology, 17(3), 405–410. McBride, R. L. (1985). Stimulus range influences intensity and hedonic ratings of flavour. Appetite, 6(2), 125–131. McBride, R. L., & Johnson, R. L. (1987). Perception of sugar-acid mixtures in lemon juice drink. International Journal of Food Science & Technology, 22(4), 399–408. McBurney, D. H., Collings, V. B., & Glanz, L. M. (1973). Temperature dependence of human taste responses. Physiology & Behavior, 11(1), 89–94. Merriam-Webster.com (2016). “Context,” Merriam-Webster. (Accessed 10 January 2018). Moir, H. C. (1936). Some observations on the appreciation of flavour in foodstuffs. Journal of Chemical Technology and Biotechnology, 55(8), 145–148. Morrot, G., Brochet, F., & Dubourdieu, D. (2001). The color of odors. Brain and Language, 79 (2), 309–320. Mosca, A. C., van de Velde, F., Bult, J. H., van Boekel, M. A., & Stieger, M. (2012). Effect of gel texture and sucrose spatial distribution on sweetness perception. LWT-Food Science and Technology, 46(1), 183–188. Moskowitz, H. R. (1973a). Effects of solution temperature on taste intensity in humans. Physiology & Behavior, 10(2), 289–292. Moskowitz, H. R. (1973b). Models of sweetness additivity. Journal of Experimental Psychology, 99(1), 88. Moskowitz, H. R. (1974). Models of additivity for sugar sweetness. In Sensation and measurement (pp. 379–388). Springer: Dordrecht. Moskowitz, H. R., & Barbe, C. D. (1977). Profiling of odor components and their mixtures. Sensory Processes, 1, 212–226. Moskowitz, H. R., & Klarman, L. (1975). The hedonic tones of artificial sweeteners and their mixtures. Chemical Senses, 1(4), 423–429. Murphy, C., & Cain, W. S. (1980). Taste and olfaction: Independence vs interaction. Physiology & Behavior, 24(3), 601–605.

Context effects at the level of the sip and bite

61

Nakano, S., & Ayabe-Kanamura, S. (2017). The influence of olfactory contexts on the sequential rating of odor pleasantness. Perception, 46(3-4), 393–405. Nasrawi, C. W., & Pangborn, R. (1989). The influence of tastants on oral irritation by capsaicin. Journal of Sensory Studies, 3(4), 287–294. Nasri, N., Septier, C., Beno, N., Salles, C., & Thomas-Danguin, T. (2013). Enhancing salty taste through odour–taste–taste interactions: Influence of odour intensity and salty tastants’ nature. Food Quality and Preference, 28(1), 134–140. Nguyen, D. H., Valentin, D., Ly, M. H., Chrea, C., & Sauvageot, F. (2002). When does smell enhance taste? Effect of culture and odorant/tastant relationship. In: Paper presented at European chemoreception research organization conference, Erlangen Germany, 23–27 July, 2002. Nugent, P. M. S. (2013). CONTEXT. In PsychologyDictionary.org. April 7, https://psycho logydictionary.org/context/. Accessed 10 January 2018. Olson, D. G., Caporaso, F., & Mandigo, R. W. (1980). Effects of serving temperature on sensory evaluation of beef steaks from different muscles and carcass maturities. Journal of Food Science, 45(3), 627–628. Overbosch, P., Afterof, W. G. M., & Haring, P. G. M. (1991). Flavor release in the mouth. Food Reviews International, 7(2), 137–184. Pangborn, R. (1960b). Taste interrelationships. Journal of Food Science, 25(2), 245–256. Pangborn, R. (1962). Taste Interrelationships. Journal of Food Science, 27(5), 495–500. Pangborn, R. (1963). Relative taste intensities of selected sugars and organic acids. Journal of Food Science, 28(6), 726–733. Pangborn, R. M. (1980). Critical analysis of sensory responses to sweetness. In: P. Koivisto & L. Hyvhonen (Eds.), Carbohydrate sweeteners in foods and nutrition, NJF Symposium held 29.11.78–1.12.78, Helsinki-Espoo, Finland, London: Academic Press. Pangborn, R., & Chrisp, R. B. (1964). Taste interrelationships. VI. Sucrose, sodium chloride, and citric acid in canned tomato juice. Journal of Food Science, 29(4), 490–498. Pangborn, R., Gibbs, Z. M., & Tassan, C. (1978). Effect of hydrocolloids on apparent viscosity and sensory properties of selected beverages. Journal of Texture Studies, 9(4), 415–436. Pangborn, R. M. (1960a). Influence of color on the discrimination of sweetness. The American Journal of Psychology, 73(2), 229–238. Pangborn, R. M. (1988). Sensory attributes and acceptance of fat, sugar, and salt in dairy products. In Food acceptability (pp. 413–429). London: Elsevier. Pangborn, R. M., Chrisp, R. B., & Bertolero, L. L. (1970). Gustatory, salivary, and oral thermal responses to solutions of sodium chloride at four temperatures. Attention, Perception, & Psychophysics, 8(2), 69–75. Pangborn, R. M., Marsh, G. L., Channell, W. R., & Campbell, H. (1960). Consumer opinion of sweeteners in frozen concentrated lemonade and orange juice drink. Food Technology, 14 (10), 515–520. Pangborn, R. M., & Trabue, I. M. (1967). Detection and apparent taste intensity of salt—acid mixtures in two media. Attention, Perception, & Psychophysics, 2(11), 503–509. Parducci, A. (1963). Range-frequency compromise in judgment. Psychological Monographs: General and Applied, 77(2), 1. Parducci, A. (1983). Category ratings and the relational character of judgment. Advances in Psychology, 11, 262–282. Parker, S., Murphy, D. R., & Schneider, B. A. (2002). Top-down gain control in the auditory system: Evidence from identification and discrimination experiments. Attention, Perception, & Psychophysics, 64(4), 598–615. Parr, W. V., Geoffrey White, K., & Heatherbell, D. A. (2003). The nose knows: influence of colour on perception of wine aroma. Journal of Wine Research, 14(2-3), 79–101.

62

Context

Perng, C. M., & McDaniel, M. R. (1989). Optimization of a blackberry juice drink using response surface methodology. In: Institute of Food Technologists, Program and Abstracts, Annual Meeting. Institute of Food Technologists, Chicago, IL216. Philipsen, D. H., Clydesdale, F. M., Griffin, R. W., & Stern, P. (1995). Consumer age affects response to sensory characteristics of a cherry flavored beverage. Journal of Food Science, 60(2), 364–368. Piqueras-Fiszman, B., & Spence, C. (2015). Sensory expectations based on product-extrinsic food cues: An interdisciplinary review of the empirical evidence and theoretical accounts. Food Quality and Preference, 40, 165–179. Piqueras-Fiszman, B., & Spence, C. (Eds.), (2016). Multisensory flavor perception: From fundamental neuroscience through to the marketplace. London: Woodhead Publishing. Pol, H. E., Hijman, R., Baare, W. F., & van Ree, J. M. (1998). Effects of context on judgements of odor intensities in humans. Chemical Senses, 23(2), 131–135. Pramudya, R. C., & Seo, H.-S. (2018). Influences of product temperature on emotional responses to, and sensory attributes of, coffee and green tea beverages. Frontiers in Psychology, 8, 2264. Prescott, J., Allen, S., & Stephens, L. (1993). Interactions between oral chemical irritation, taste and temperature. Chemical Senses, 18(4), 389–404. Prescott, J., Johnstone, V., & Francis, J. (2004). Odor–taste interactions: Effects of attentional strategies during exposure. Chemical Senses, 29(4), 331–340. Prescott, J., & Stevenson, R. J. (1995a). Effects of oral chemical irritation on tastes and flavors in frequent and infrequent users of chili. Physiology & Behavior, 58(6), 1117–1127. Prescott, J., & Stevenson, R. J. (1995b). Pungency in food perception and preference. Food Reviews International, 11(4), 665–698. Rankin, K. M., & Marks, L. E. (1991). Differential context effects in taste perception. Chemical Senses, 16(6), 617–629. Rifkin, B., & Bartoshuk, L. M (1980). Taste synergism between monosodium glutamate and disodium 50 -guanylate. Physiology & behavior, 24(6), 1169–1172. Riskey, D. R. (1982). Effects of context and interstimulus procedures in judgments of saltiness and pleasantness. In J. T. Kuznicki, R. A. Johnson, & A. F. Rutkiewic (Eds.), Selected sensory methods: Problems and approaches to measuring hedonics (pp. 71–83). Philadelphia: American Society for Testing and Materials. Riskey, D. R., Parducci, A., & Beauchamp, G. K. (1979). Effects of context in judgments of sweetness and pleasantness. Attention, Perception, & Psychophysics, 26(3), 171–176. Rolls, E. T. (2011). Chemosensory learning in the cortex. Frontiers in Systems Neuroscience, 5, 78. https://doi.org/10.3389/fnsys.2011.00078. Roth, H. A., Radle, L. J., Gifford, S. R., & Clydesdale, F. M. (1988). Psychophysical relationships between perceived sweetness and color in lemon- and lime-flavored drinks. Journal of Food Science, 53(4), 1116–1119. Rozin, P. (1982). “Taste-smell confusions” and the duality of the olfactory sense. Attention, Perception, & Psychophysics, 31(4), 397–401. Rozin, P., & Tuorila, H. (1993). Simultaneous and temporal contextual influences on food acceptance. Food Quality and Preference, 4(1-2), 11–20. Salles, C. (2006). Odour–taste interactions in flavour perception. In Flavour in food (pp. 345–368). London: Woodhead Publishing. Schifferstein, H. N. (1995a). Contextual effects in difference judgments. Attention, Perception, & Psychophysics, 57(1), 56–70. Schifferstein, H. N. (1995b). Contextual shifts in hedonic judgments. Journal of Sensory Studies, 10(4), 381–392.

Context effects at the level of the sip and bite

63

Schifferstein, H. N., & Frijters, J. E. (1992). Contextual and sequential effects on judgments of sweetness intensity. Attention, Perception, & Psychophysics, 52(3), 243–255. Schifferstein, H. N., & Frijters, J. E. (1993). Perceptual integration in heterogeneous taste percepts. Journal of Experimental Psychology: Human Perception and Performance, 19 (3), 661. Schifferstein, H. N., & Kleykers, R. W. (1996). An empirical test of Olsson’s interaction model using mixtures of tastants. Chemical Senses, 21(3), 283–291. Schifferstein, H. N., & Kuiper, W. E. (1997). Sequence effects in hedonic judgments of taste stimuli. Attention, Perception, & Psychophysics, 59(6), 900–912. Schifferstein, H. N., & Oudejans, I. M. (1996). Determinants of cumulative successive contrast in saltiness intensity judgments. Attention, Perception, & Psychophysics, 58(5), 713–724. Schifferstein, H. N., & Tanudjaja, I. (2004). Visualising fragrances through colours: the mediating role of emotions. Perception, 33(10), 1249–1266. Schifferstein, H. N., & Verlegh, P. W. (1996). The role of congruency and pleasantness in odor-induced taste enhancement. Acta Psychologica, 94(1), 87–105. Schiffman, S. S., Booth, B. J., Carr, B. T., Losee, M. L., Sattely-Miller, E. A., & Graham, B. G. (1995). Investigation of synergism in binary mixtures of sweeteners. Brain Research Bulletin, 38(2), 105–120. Schiffman, S. S., Frey, A. E., Luboski, J. A., Foster, M. A., & Erickson, R. P. (1991). Taste of glutamate salts in young and elderly subjects: role of inosine 50 -monophosphate and ions. Physiology & Behavior, 49(5), 843–854. Schiffman, S. S., Gatlin, L. A., Sattely-Miller, E. A., Graham, B. G., Heiman, S. A., Stagner, W. C., et al. (1994). The effect of sweeteners on bitter taste in young and elderly subjects. Brain Research Bulletin, 35(3), 189–204. Schiffman, S. S., Sattely-Miller, E. A., Graham, B. G., Booth, B. J., & Gibes, K. M. (2000). Synergism among ternary mixtures of fourteen sweeteners. Chemical Senses, 25(2), 131–140. Shankar, M. U., Levitan, C. A., & Spence, C. (2010). Grape expectations: The role of cognitive influences in color–flavor interactions. Consciousness and Cognition, 19(1), 380–390. Slocombe, B. G., Carmichael, D. A., & Simner, J. (2016). Cross-modal tactile–taste interactions in food evaluations. Neuropsychologia, 88, 58–64. Small, D. M., & Prescott, J. (2005). Odor/taste integration and the perception of flavor. Experimental Brain Research, 166(3-4), 345–357. Smeets, M. A., & Dijksterhuis, G. B. (2014). Smelly primes–when olfactory primes do or do not work. Frontiers in Psychology, 5, 96. Spence, C. (2011). Crossmodal correspondences: A tutorial review. Attention, Perception, & Psychophysics, 73(4), 971–995. Spence, C. (2015a). On the psychological impact of food colour. Flavour, 4(1), 21. Spence, C. (2015b). Eating with our ears: assessing the importance of the sounds of consumption on our perception and enjoyment of multisensory flavour experiences. Flavour, 4(1), 3. Spence, C., Levitan, C. A., Shankar, M. U., & Zampini, M. (2010). Does food color influence taste and flavor perception in humans? Chemosensory Perception, 3(1), 68–84. Spence, C., & Piqueras-Fiszman, B. (2016). Oral-somatosensory contributions to flavor perception and the appreciation of food and drink. In Multisensory flavor perception (pp. 59–79). Woodhead Publishing. Spence, C., & Shankar, M. U. (2010). The influence of auditory cues on the perception of, and responses to, food and drink. Journal of Sensory Studies, 25(3), 406–430. Spence, C., & Zampini, M. (2006). Auditory contributions to multisensory product perception. Acta Acustica united with Acustica, 92(6), 1009–1025.

64

Context

Stevens, D. A., & Lawless, H. T. (1986). Putting out the fire: Effects of tastants on oral chemical irritation. Attention, Perception, & Psychophysics, 39(5), 346–350. Stevens, J. C. (1995). Detection of heteroquality taste mixtures. Perception & Psychophysics, 57(1), 18–26. Stevenson, R. J. (2016). Attention and flavor binding. In B. Fiszman & C. Spence (Eds.), Multisensory flavor perception (pp. 15–35). Woodhead Publishing. Stevenson, R. J., & Boakes, R. A. (2004). Sweet and sour smells: Learned synesthesia between the senses of taste and smell. In G. A. Calvert, C. Spence, & B. E. Stein (Eds.), The handbook of multisensory processing (pp. 69–83). Cambridge, MA: MIT Press. Stevenson, R. J., Boakes, R. A., & Prescott, J. (1998). Changes in odor sweetness resulting from implicit learning of a simultaneous odor-sweetness association: An example of learned synesthesia. Learning and Motivation, 29(2), 113–132. Stevenson, R. J., & Oaten, M. (2008). The effect of appropriate and inappropriate stimulus color on odor discrimination. Perception & Psychophysics, 70(4), 640–646. Stevenson, R. J., & Prescott, J. (1994). The effects of prior experience with capsaicin on ratings of its burn. Chemical Senses, 19(6), 651–656. Stevenson, R. J., Prescott, J., & Boakes, R. A. (1999). Confusing tastes and smells: how odours can influence the perception of sweet and sour tastes. Chemical Senses, 24(6), 627–635. Stewart, N., Brown, G. D., & Chater, N. (2005). Absolute identification by relative judgment. Psychological Review, 112(4), 881. Sundqvist, N. C., Stevenson, R. J., & Bishop, I. R. (2006). Can odours acquire fat-like properties? Appetite, 47(1), 91–99. Szczesniak, A. S. (1988). The meaning of textural characteristics – Crispness. Journal of Texture Studies, 19(1), 51–59. Teerling, A. (1992). The colour of taste. Chemical Senses, 17(6), 886. Thomas-Danguin, T., Sinding, C., Romagny, S., El Mountassir, F., Atanasova, B., Le Berre, E., et al. (2014). The perception of odor objects in everyday life: A review on the processing of odor mixtures. Frontiers in Psychology, 5, 504. Thomas-Danguin, T., Sinding, C., Tournier, C., & Saint-Eve, A. (2016). Multimodal interactions. P. Etievant, E. Guichard, C. Salles, & A. Voilley (Eds.), Flavor From Food to Behaviors, Wellbeing and HealthWoodhead Publishing Series in Food Science, Technology and Nutrition, pp. 121–141. Tournier, C., Sulmont-Rosse, C., Semon, E., Vignon, A., Issanchou, S., & Guichard, E. (2009). A study on texture–taste–aroma interactions: Physico-chemical and cognitive mechanisms. International Dairy Journal, 19(8), 450–458. Urbanyi, G. (1982). Investigation into the interaction of different properties in the course of sensory evaluation. I. The effect of colour upon the evaluation of taste in fruit and vegetable products. Acta Alimentaria, 11, 233–243. Valentin, D., Chrea, C., & Nguyen, D. H. (2006). Taste-odour interactions in sweet taste perception. In Optimising sweet taste in foods (pp. 66–84). London: Woodhead. Verhagen, J. V., & Engelen, L. (2006). The neurocognitive bases of human multimodal food perception: Sensory integration. Neuroscience & Biobehavioral Reviews, 30(5), 613–650. Vickers, Z. (1991). Sound perception and food quality. Journal of Food Quality, 14(1), 87–96. Vickers, Z., & Bourne, M. C. (1976a). A psychoacoustical theory of crispness. Journal of Food Science, 41(5), 1158–1164. Vickers, Z., & Bourne, M. C. (1976b). Crispness in foods – A review. Journal of Food Science, 41(5), 1153–1157. Vickers, Z. M. (1985). The relationships of pitch, loudness and eating technique to judgments of the crispness and crunchiness of food sounds. Journal of Texture Studies, 16(1), 85–95.

Context effects at the level of the sip and bite

65

Voirol, E., & Daget, N. (1989). Direct nasal and oronasal profiling of a meat flavouring: Influence of temperature, concentration and additives. Lebensmittel-Wissenschaft + Technologie, 22(6), 399–405. Vollmecke, T. A. (1987). Long-lasting effects of context on sweetness evaluation. Annals of the New York Academy of Sciences, 510(1), 695–697. von Sydow, E., Moskowitz, H., Jacobs, H., & Meiselman, H. (1974). Odor-taste interaction in fruit juices. Lebensmittel-Wissenschaft und Technologie, 7(1), 18–24. Walter, F., & Boakes, R. A. (2009). Long-term range effects in hedonic ratings. Food Quality and Preference, 20(6), 440–449. Ward, L. M. (1979). Stimulus information and sequential dependencies in magnitude estimation and cross-modality matching. Journal of Experimental Psychology: Human Perception and Performance, 5(3), 444. Ward, L. M. (1982). Mixed-modality psychophysical scaling: Sequential dependencies and other properties. Attention, Perception, & Psychophysics, 31(1), 53–62. Ward, L. M. (1985). Mixed-modality psychophysical scaling: Inter-and intramodality sequential dependencies as a function of lag. Attention, Perception, & Psychophysics, 38(6), 512–522. Ward, L. M., Armstrong, J., & Golestani, N. (1996). Intensity resolution and subjective magnitude in psychophysical scaling. Attention, Perception, & Psychophysics, 58(5), 793–801. Weenen, H., Jellema, R. H., & De Wijk, R. A. (2005). Sensory sub-attributes of creamy mouthfeel in commercial mayonnaises, custard desserts and sauces. Food Quality and Preference, 16(2), 163–170. Wiltrout, C., Dogra, S., & Linster, C. (2003). Configurational and nonconfigurational interactions between odorants in binary mixtures. Behavioral Neuroscience, 117(2), 236. Yau, N. J. N., & McDaniel, M. R. (1992). Carbonation interactions with sweetness and sourness. Journal of Food Science, 57(6), 1412–1416. Zampini, M., Sanabria, D., Phillips, N., & Spence, C. (2007). The multisensory perception of flavor: Assessing the influence of color cues on flavor discrimination responses. Food Quality and Preference, 18(7), 975–984. Zampini, M., & Spence, C. (2004). The role of auditory cues in modulating the perceived crispness and staleness of potato chips. Journal of Sensory Studies, 19(5), 347–363. Zampini, M., & Spence, C. (2005). Modifying the multisensory perception of a carbonated beverage using auditory cues. Food Quality and Preference, 16(7), 632–641. Zampini, M., Wantling, E., Phillips, N., & Spence, C. (2008). Multisensory flavor perception: Assessing the influence of fruit acids and color cues on the perception of fruit-flavored beverages. Food Quality and Preference, 19(3), 335–343. Zellner, D. A. (2013). Color–odor interactions: A review and model. Chemosensory Perception, 6(4), 155–169. Zellner, D. A., Allen, D., Henley, M., & Parker, S. (2006). Hedonic contrast and condensation: Good stimuli make mediocre stimuli less good and less different. Psychonomic Bulletin & Review, 13(2), 235–239. Zellner, D. A., Bartoli, A. M., & Eckard, R. (1991). Influence of color on odor identification and liking ratings. The American Journal of Psychology, 547–561. Zellner, D. A., & Durlach, P. (2003). Effect of color on expected and experienced refreshment, intensity, and liking of beverages. The American Journal of Psychology, 116(4), 633–647. Zellner, D. A., Hoer, K., & Feldman, J. (2014). Labels affect both liking and preference: The better the stimuli, the bigger the preference. Attention, Perception, & Psychophysics, 76(8), 2189–2192. Zellner, D. A., Jones, K., Morino, J., Cogan, E. S., Jennings, E. M., & Parker, S. (2010). Increased hedonic differences despite increases in hedonic range. Attention, Perception, & Psychophysics, 72(5), 1261–1265.

66

Context

Zellner, D. A., & Kautz, M. A. (1990). Color affects perceived odor intensity. Journal of Experimental Psychology: Human Perception and Performance, 16(2), 391. Zellner, D. A., Kern, B. B., & Parker, S. (2002). Protection for the good: Subcategorization reduces hedonic contrast. Appetite, 38(3), 175–180. Zellner, D. A., Mattingly, M. C., & Parker, S. (2009). Categorization reduces the effect of context on hedonic preference. Attention, Perception, & Psychophysics, 71(6), 1228–1232. Zellner, D. A., Rohm, E. A., Bassetti, T. L., & Parker, S. (2003). Compared to what? Effects of categorization on hedonic contrast. Psychonomic Bulletin & Review, 10(2), 468–473. Zellner, D. A., Stewart, W. F., Rozin, P., & Brown, J. M. (1988). Effect of temperature and expectations on liking for beverages. Physiology & Behavior, 44(1), 61–68. Zellner, D. A., & Whitten, L. A. (1999). The effect of color intensity and appropriateness on color-induced odor enhancement. The American Journal of Psychology, 112(4), 585. Zou, Z., & Buck, L. B. (2006). Combinatorial effects of odorant mixes in olfactory cortex. Science, 311(5766), 1477–1481. Zwebner, Y., Lee, L., & Goldenberg, J. (2014). The temperature premium: Warm temperatures increase product valuation. Journal of Consumer Psychology, 24(2), 251–259.

In-home testing

4


Lion* Elizabeth H. Zandstra*,†, Rene *Consumer Science, Unilever R&D Vlaardingen, Vlaardingen, The Netherlands, †Division of Human Nutrition and Health, Wageningen University, Wageningen, The Netherlands

4.1

In-home testing—Bringing lab results to the real world

4.1.1 Qualitative observational consumer research 4.1.1.1 Aim and expected outcome In-home qualitative consumer research aims to understand consumers’ behaviors during food preparation and consumption, including food products, beverages, meals, recipes, ingredients, herbs and spices, and the use of kitchen equipment and cooking utensils (Richardson, 1996). Observing people in their natural environments (i.e., while at home) provides a rich understanding of the dynamics of food preparation and consumption behavior (cf. Fine & Elsbach, 2000): How do they carry out cooking and meal preparations in natural settings, how do they use products and why, who eats what, when, and where? It is important to do this in a natural setting, as people have often found ways to deal with minor inconveniences (e.g., lids that are difficult to open, use of bouillon cubes), and much of this behavior is habitual. As habitual behaviors are automatic and mostly occur outside conscious awareness, they are less amenable to introspection (Verplanken, 2006; Verplanken & Wood, 2006) and are less likely to surface in individual or group interviews, which depend on conscious retrospective introspection of one’s cooking experience. Qualitative observational consumer research therefore supplements data obtained from individual and group interviews. It provides detailed information about how people use products—what specific actions do they perform, what workarounds have they created, what annoys them, what do they find pleasing about the experience? Moreover, it provides valuable input for quantitative consumer research studies on what to ask and in what format. Thus, there are two ways in which qualitative observational consumer studies can add value: (1) basic consumer understanding to identify opportunities for innovation and help refine products/concepts, and (2) identifying key steps and language used for further quantitative studies.

4.1.1.2 Basic consumer understanding to aid innovation Understanding how and why consumers carry out their actions during cooking stimulates hypotheses as to which needs have not been adequately met, and where innovative products may have real consumer appeal. This is especially true in new Context. https://doi.org/10.1016/B978-0-12-814495-4.00004-0 Copyright © 2019 Elsevier Inc. All rights reserved.

68

Context

markets, or when new products have been launched in a market. For example, how do people currently use bouillon cubes? When do they add them? How do they add them? What do they notice? What are negative elements in the cooking process, what are positive elements? How can product development be used to improve the experience? Thus, observational consumer research studies provide fine-grained information about people’s actual actions, the details of which people are often unaware of. There are a number of theoretical considerations that need to be taken into account when conducting qualitative observational consumer research. A consumer may change her behavior when she is being observed, as she may be more conscious than normal of what she is doing and might be inclined to create a positive impression of herself (Leary & Kowalski, 1990; Pliner & Chaiken, 1990; Vartanian, 2015; Vartanian, Herman, & Polivy, 2007). Social norms regarding food preparation may then change her behaviors during the observation process, for example, she might make smaller portions or cook with less fat. Emphasizing that the interviewee should really cook the way she usually does and asking the right questions may help limit this bias. Another consideration is the context in which the respondent is performing the activity. For example, in foods, it is known that behavior and judgment criteria are different according to the time of the day, for example, people predicted that they would like a cold cereal at lunch and a pizza at breakfast less, as they did not feel it was appropriate for that situation and time of the day (Cardello, Schutz, Snow, & Lesher, 2000). It is therefore important that observations take place at a time when a respondent would usually perform the activity. For similar reasons, the observation should usually take place in the respondent’s own home with their own equipment. In addition, it is important to sample a range of behaviors likely to be relevant for the food product of interest, and include these as recruitment criteria. Also, there is often a need to understand not only those who use the food product of interest (users), but also those who have done so in the past (lapsed users), and those who might do so in the future (potential users). It is always difficult to give an indication of samples size, but based on our experience, a small sample of about 6 to 8 respondents per target group will provide useful information, and will yield saturation (i.e., an additional interview will not provide much more new information).

4.1.1.3 Groundwork for quantitative studies Two other reasons for conducting qualitative observational research is to (1) gain technical data to set up relevant (laboratory) evaluation measurement techniques, and (2) to obtain information on how a new product might be used at home in real life, including both positive and negative aspects, so that a quantitative assessment measures the appropriate steps, actions, benefits and so forth. Thus, exploratory qualitative observational consumer research helps identify which questions to ask and which steps to measure. Without a good insight into these behaviors, a quantitative study runs the risk of asking irrelevant or incomprehensible questions. The results of qualitative research are potential drivers of brand choice and product performance (both

In-home testing

69

observed and consumer perceived), consumer language describing the sensory experience of a product used at home, consumer-stated motivations for choice of food products or brands, and consumers’ perceptions of constraints, problems, needs, likes, and dislikes.

4.1.1.4 Measurement There is no one-size-fits-all methodology for conducting qualitative research (Maxwell, 2009). However, for conducting in-home observation tests for food products, we have found that closely monitoring the preparation and cooking behaviors provides valuable information. In doing so, there are two styles: “interactive observation” and “passive observation.” The choice between the techniques depends on the objectives of the study. The interactive observation technique provides direct feedback on perceptions and motivations during the natural process of preparation and cooking. The interviewer probes the respondent’s thoughts, feelings, and actions while the respondent is performing the behavior. Ideally, the cooking behavior is taped on video. The consumer data for analysis consists of videos of the activity and transcriptions of the interview. Although the drawback is that the respondent could be interrupted during the process, the advantage is that the data is generated as it occurs, enabling a more accurate representation of actual thoughts and feelings. The passive observation technique is used to record the natural processes as unobtrusively as possible. The respondent is not diverted from the natural flow of the activity, and is more likely to perform it according to her normal routine. Video recording is also useful for later analysis. The analysis consists of the breakdown of the process into a series of activities and items of interest, for example, opening a bottle (from “pick up” to “hand leaving cap”), and dosing (from “finish of opening” to “bottle put down”). It is important to be as unobtrusive as possible during the observation, and the respondent should feel relaxed and comfortable. Following the observation, a qualitative individual interview can help identify the respondents’ goals during the process, her motivations for different actions, her perceptions, likes/dislikes/problems, and what would be her ideal outcome and process. By playing back the video recording of the behaviors to the respondent during this interview, additional information can be gathered (e.g., “What caused you to do that?”), be it that respondents reflect on this in retrospect, rather than as it occurs. Fig. 4.1 shows an example of an observation at a consumer’s home to obtain a full dish understanding on what she cooks and how she cooks. As the whole cooking process is captured without interruption, it can be analyzed in a very structured way from the start of the cooking, up to the evaluation of the end dish. The recipe and ingredients that are used, which parts of the ingredients are used (e.g., small pieces, large pieces in blocks or stripes), how they are prepared (e.g., boiling, frying), and what consumers do (e.g., chopping, stirring) can all be captured. Depending on the specific objective of the study, variations can be applied to the study designs: for example, longer observations (i.e., “living in” with respondent for 1–2 days), a fixed video camera at home activated by a trigger, such as picking up the product package, respondents using their own video camera with another family

70

Context

Fig. 4.1 Example of an observational study at a consumer’s home, from opening the pack, to preparing the dish, and evaluation of the end dish.

member recording the process, respondents providing a running commentary of what they are doing and/or perceiving during the video, respondents using the product for a week and then being video-taped as they use the product at a central location.

4.1.2 Quantitative in-home test 4.1.2.1 Aim and expected outcome Traditional product tests are mostly restricted to first impressions at central locations, that is, consumers are asked to taste and rate how much they like a sample just once. However, data obtained from central location single measurements in which consumers test small food samples can differ considerably from data obtained from repeated in-home consumption of larger or meal-sized portions (Boutrolle, Delarue, Arranz, Rogeaux, & K€ oster, 2007; Holthuysen, Vrijhof, De Wijk, & Kremer, 2017). The advantage of an in-home test is that the data are a better reflection of a daily life situation in which people repeatedly consume and use the product over a longer period of time. Thus, people may modify the product through the cooking methods they use, add other ingredients, or use the product in a personalized manner.

4.1.2.2 Design There are essentially two approaches to in home product testing: (1) monadic testing, where respondents test only one product, and (2) comparative testing, where respondents test 2 or more products and compare them with each other. Monadic testing is more representative of real life, in which consumers try one product at a time, and compare them in their own minds to other products they have tried. As a consequence, this test is less likely to pick up differences compared with a comparative test. If differences are picked up, it is likely they will be picked up in real life as well. This test is typically used to assess acceptability of novel products and product improvements of existing products, and ensures that products that are launched have noticeable

In-home testing

71

improvements or benefits. In contrast, a comparative test is more likely to pick up differences between products as they are directly compared with one another. If no differences are picked up, one can be reasonably sure these will not be picked up in real life. The advantage of comparative testing is that more options can be tested at once, and that drivers of liking can be identified, as well as key differences between the products. Care should be taken to avoid order effects, for example where the tasting of the first product impacts the tasting of the second product, leading to unwanted influences on the rating of the second product. In that case, it should be monadically tested. The comparative test is typically used for product development purposes when a high sensitivity is needed, for example, when product formulation changes should not be noticed by consumers. Product testing should be conducted among a representative sample of the (intended) target group. A minimum of 200 respondents per monadic cell is recommended, with a minimum of 75 respondents for each sub-group (e.g., 75 light users vs. 75 heavy users). Respondents should be given sufficient time to assess the product adequately, based on the number of times that the product is usually used or intended to be used. Typically, this is for a period of several days in a row (up to weeks or even months), depending on the type of product (Hetherington, Bell, & Rolls, 2000; Porcherot & Issanchou, 1998; Stubenitsky, Aaron, Catt, & Mela, 1999; Zandstra, De Graaf, & Van Trijp, 2000; Zandstra, Weegels, Van Spronsen, & Klerk, 2004). In the next sections, we will describe the underlying mechanisms involved in changes in liking over time, and further considerations that need to be taken into account when designing an in-home test.

4.1.2.3 Underlying mechanisms One of the key outcome measures in in-home use testing is liking, which can be very dynamic. Researchers have shown an increased liking with repeated consumption, especially relevant for novel, unfamiliar flavors and products (Crandall, 1984; Porcherot & Issanchou, 1998), a sustained liking (Hetherington et al., 2000; Kahneman & Snell, 1992; L
evy & K€ oster, 1999; Willems, Van Hout, Zijlstra, & Zandstra, 2014; Zandstra et al., 2004), and a decreased liking (Porcherot & Issanchou, 1998; Stubenitsky et al., 1999; Zandstra et al., 2000). For example, Hetherington, Pirie, and Nabb (2002) showed a significant decline in pleasantness for chocolate, but not for bread and butter, when it was eaten daily at the laboratory for 22 days (see Fig. 4.2). Overall, it can be concluded that changes in liking over repeated exposure very much depend on the type of food product. Staple foods, such as crackers, milk, bread, and butter, seem to be resistant to boredom, as their liking ratings remain stable over repeated consumption, both at high and low frequencies of consumption (Zandstra et al., 2004). Other food products either increase or decrease in liking over repeated consumption. How do these changes in liking over time come about? We know that liking is, to a large extent, learned (Yeomans, 2006). The underlying learning mechanisms are: (1) “mere” exposure effects, (2) associative conditioning, or (3) social conditioning.

72

Context 100

Rating (mm)

80

60

40

20

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15 16 17 18 19 20 21 22 Time (days)

Fig. 4.2 Mean daily pleasantness ratings measured across 22 days at the laboratory: example of decreased liking (( ) chocolate) and sustained liking ((O) bread). Arrows indicate test days when intake was ad libitum. Source: Hetherington, M. M., Pirie, L. M., & Nabb, S. (2002). Stimulus satiation: Effects of repeated exposure to foods on pleasantness and intake. Appetite 38, 19–28.

▪

“Mere exposure” effects—“Mere exposure” is the phenomenon whereby people’s liking for novel things increases the more they are exposed to them (Bornstein, 1989; Stang, 1975; Zajonc, 1968). This “mere exposure” effect has been shown with different stimuli (e.g., pictures, sounds, paintings, human faces), and the results are fairly consistent (Bornstein, 1989). With food products, mere exposure effects are especially relevant for novel, unfamiliar flavors and products (Pliner, 1982). The underlying mechanisms are not yet clear, but repeatedly exposing consumers to new flavors may result in reduced feelings of uncertainty, due to an increase in perceptual fluency. There is also the possibility that covert associative conditioning processes are responsible for changes in liking after repeat exposure. Associative conditioning—Liking (but also disliking) of food can be formed through a learned association between the context in which the food is eaten, and/ or physiological consequences of ingesting foods with the sensory characteristics of those foods (Bolles, Hayward, & Crandall, 1981; Booth, Mather, & Fuller, 1982; Kuenzel et al., 2011; Kuenzel, Zandstra, El-Deredy, Blanchette, & Thomas, 2011; Mela, 2000). This is called associative conditioning. The associative effect for which we have the most evidence is taste aversion (Bernstein & Webster, 1980; Pelchat & Rozin, 1982). A flavor that is associated with something that produces nausea/vomiting comes to be disliked. This learned taste aversion usually requires only one exposure to the food, and, as a result, the food may be avoided for many years. In a similar manner, consumers learn to like foods that are associated with positive postingestional effects, for example, fat and sugar (Booth et al., 1982), but only when sufficient cognitive resources are available (Davies, El-Deredy, Zandstra, & Blanchette, 2012), and in the absence of explicit expectations of actual nutrient content of the product (Brunstrom, 2005; Gould, Zandstra, & Yeomans,

In-home testing

73

2017; Zandstra & El-Deredy, 2011). The knowledge available on conditioned preferences is mostly based on laboratory research with children, which showed that flavors associated with higher energy density come to be preferred. However, we should be careful with this outcome: there are also several unpublished studies that were unsuccessful in demonstrating conditioned preferences in children (Yeomans, 2012). For adults, evidence for conditioned preferences is mixed: it is very difficult to influence the development of flavor preferences with foods under realistic eating conditions at home (Stubenitsky et al., 2000). Social conditioning—Learned associations might also develop through the social environment. We eat our meals with other people, which can have a positive or negative effect on liking. And when we eat, the social context can influence our mood, which might subsequently affect our liking. The knowledge available on the effect of social influences on food liking is mostly based on research with children. A successful example is the research program of Lowe, Horne, Tapper, Bowdery, and Egerton (2004) that has shown large-scale and long-lasting increases in children’s consumption of fruit and vegetables. In this research program, a combination of peer modeling (children seeing their “Food Dudes” heroes eating vegetables) and consumption rewards (e.g., stickers, pencils) resulted in a significant increase in total intake of fruit served (from 20% to 69%), and vegetables served (from 35% to 55%) at school, but also an increased liking for the fruits and vegetables. The gadgets (stickers, pens, etc.) were the reward that motivated the kids to begin to taste the foods. After a while, the taste of the fruit and vegetables became the reward, as the kids developed a taste for them. Four months later (and in the absence of modeling and reward), children were still eating almost three times as much fruit as at baseline (i.e., 55%), and significantly more vegetables (maintained at 53%) than they were before the study.

4.1.2.4 Factors influencing product evaluations Appropriateness and context of use One of the factors that influences liking is the perceived appropriateness of the food for the eating situation (see also Chapter 6 by Giacalone). Perceived appropriateness refers to the degree to which a food is perceived suitable for a specific context (time, place, and person; Schutz, 1994). For example, you may like ice cream a lot, but feel no desire to eat this when it is served for breakfast. Cardello et al. (2000) showed a strong association between food appropriateness and expected liking/disliking in a laboratory study. For example, consumers predicted that cold cereals consumed at lunch would be liked less, as it was judged inappropriate for that situation. In fact, they did like the products when consumed in this situation, but participants were less satisfied with the overall experience at the end, and this could be attributed to their appropriateness judgments. The latter has potentially important implications for understanding and managing product acceptance and repeat usage: a food that perfectly matches the situation will probably result in higher liking scores. Social context has been reported to impact liking and food consumption as well (Stroebele & De Castro, 2004; see also Chapter 2 by Higgs). Social context comprises

74

Context

mainly the people involved, that is, whether foods are consumed alone or in a group, or whether they are consumed together with family or with strangers (De Castro, 1990; De Castro & De Castro, 1989). This also has implications for the design of the in-home test. The evaluation and use of the product should be in a context that is as similar as possible to how the product would usually be used by consumers, and this should be included in the instructions to the respondents.

Expectations Expectations play an important role in food evaluation. Consumer’s evaluations of a food product depends not only on the actual sensory properties, but also, to a large extent, on the product information that is available to them, for example, appearance, claims, and packaging. This information creates expectations that drive the perception and liking of a product (Cardello, 1994; Deliza & MacFie, 1996; Kuenzel, Blanchette, et al., 2011; Kuenzel, Zandstra, et al., 2011; Willems et al., 2014). For example, seeing a brown, carbonated beverage may evoke, by association, the expectation that it will taste like Coca Cola. Liking can move in the direction of positive expectations and enhance the perception of the product, a so-called “assimilation effect,” but if the actual taste is too different compared with expectations, it can create a so-called “contrast effect” (Deliza & MacFie, 1996). An example of an assimilation effect: if you expect a labeled/branded product to taste great (e.g., 8 on 9-point scale), but the actual product scores average when tasted blind (e.g., 6), you would still perceive it to taste pretty good (e.g., 7). Of course, the difference between expected and actual product taste should not be too large. In that case, the promise of a good product when the quality is poor could lead to a decrease instead of an increase in consumer satisfaction (i.e., “contrast effect”). As consumers buy brands but use products, it is important to understand which aspects of product performance is influenced (biased) by branding. In blind testing, consumers receive the product without knowing the product’s brand, and the advantage is that it is a clear evaluation of the product, without any influence (bias) of the brand. In branded testing, consumers receive and test the product knowing which brand it is. The product’s score can then be influenced by the brand, but the advantage is that you get an understanding on whether the product and brand fit together or not. It is therefore important to reflect and decide on what type of information is provided about the product or not, e.g., will it be tasted blind, with a concept, or branded.

Amount and frequency of consumption In a central location test, the amount of food consumed is usually fixed, and portions tend to be relatively small, with the advantage that measurements are conducted under standardized conditions. In contrast, in an in-home test, larger or full portions of the food are prepared and consumed (Boutrolle et al., 2007). These fixed and smaller portions can give a biased estimation of the optimal taste intensity of foods (Bellisle, Giachetti, & Tournier, 1988; P
erez, Dalix, Guy-grand, & Bellisle, 1994; Zandstra, De Graaf, Van Trijp, & Van Staveren, 1999). For example, in a central location test, Zandstra et al. (1999) showed that the optimal sucrose concentration in sweetened

In-home testing

75

10

10

9

9

8

8

7

7

6

6

Liking

Desire

yogurts was higher in situations in which subjects consumed a small fixed portion of the yogurt (50 mL), compared with ad libitum consumption, in which subjects were allowed to consume as much yogurt as they wished from a bowl of 300 g (i.e., to satiety). An explanation could be that the small, fixed portion did not include the postingestive rewarding effects (i.e., the presence or absence of satiety) of ingesting substantial amounts of yogurt with sucrose (Sclafani, 1997). However, other laboratory studies did not find a difference in optimal concentrations as a function of amounts of food consumed (Daillant & Issanchou, 1991; Popper, Maller, & Cardello, 1989; Shepherd, Farleigh, & Warf, 1991). Another variable that is important for liking is the frequency of consumption. A consumption frequency that is perceived as undesirable or too high could result in boredom with the product. It is therefore important to consider different consumption frequencies when testing new products of which the “normal” consumption frequency is not yet known. For example, in an in-home test, we investigated the impact of repeated exposure on desire and liking of three different sweet-flavored spreads. All participants (n ¼ 50 per group) were asked to consume the spread on a slice of bread at home every weekday for 3 weeks in a row. We found a significant decline in desire to eat (P < .05) and liking (P < .05) (see Fig. 4.3). However, no significant differences were found in the magnitude of the decrease in desire and liking between the three different-flavored spreads (P > .05). These results have clear implications, as it appeared that these products were very sensitive to boredom, suggesting the products were not appropriate for daily consumption. Therefore, when in doubt about the envisaged usage or positioning of the tested product, it is important to vary the frequency of consumption in the in-home use test.

5

5

4

4

3

3

2

2

1

1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Time (days) Flavor A

Flavor B

Time (days) Flavor C

Flavor A

Flavor B

Flavor C

Fig. 4.3 Desire-to-eat (left) and liking (right) ratings of sweet flavored spreads over time (3 weeks, five exposures per week), measured on 10-point scales at home (from 1 ¼ “no desire at all” to 10 ¼ “very much desire” and 1 ¼ “don’t like it at all” to 10 ¼ “like it extremely”).

76

Context

Variety and choice of products There are individual differences in how quickly people get bored with a product. Variety seeking might partially explain these individual differences in the long-term acceptability of foods. Seeking variety is a major feature of food choice behavior (Rozin & Markwith, 1991). Both person-related and context-related determinants have been shown to influence the intensity of variety-seeking in product choice behavior (Van Trijp, 1994; Van Trijp, L€ahteenm€aki, & Tuorila, 1992; Van Trijp & Steenkamp, 1992). To what extent repeated consumption affects food acceptance depends on the availability of different varieties of particular foods, and the degree of freedom of choice (Deci & Ryan, 1987; Kamen & Peryam, 1961; Meiselman, De Graaf, & Lesher, 2000; Meiselman, Johnson, Reeve, & Crouch, 2000; Pliner, Polivy, Herman, & Zakalusny, 1980). In most studies that used an in-home test, liking was measured in a nonchoice situation, as participants had to consume what the research protocol prescribed. In real life, people are free to choose what they eat on a given day, and in that perspective, these studies might not have been a good reflection of a daily life situation. In fact, Zandstra et al. (2000) showed that consumers liked a product most when they could choose among three alternatives, compared with those who were not allowed to choose. In addition, Kramer, Lesher, and Meiselman (2001) suggested that the repeat consumption of the same food with or without choice might be a different phenomenon. De Graaf et al. (2005) found that liking ratings obtained under laboratory conditions in which participants had some degree of choice among foods improved correlations with liking ratings obtained under realistic field conditions. Although one could argue that at home, only the food preparer determines what is eaten at dinner, and the family does have the opportunity to influence what is prepared, which is typically not the case in in-home tests with a fixed protocol. Future work in liking research should therefore address choice to better understand and improve the validity of liking ratings, both in in-home tests and central location tests.

4.1.2.5 Measurement Preference, liking, and desire Different measurements have been used to investigate food acceptance, such as: preference, liking, pleasantness or appreciation, desire to eat, actual intake, perceived boredom, probability of choosing a food, and product interestingness. Preference refers to a choice of one item rather than another (Mela, 2000). Preference judgments are often a mixture of many different factors, including price, availability, and beliefs (e.g., concerns about health, nutrition, body weight). Liking is an affective response to foods. Comparing liking scores of two products is therefore not the same as asking a preference between two products. The latter describes a choice, whereas the “like” statement describes an experience of pleasure (Mela, 2000; Zandstra & El-Deredy, 2011). It is quite possible to “prefer” one item to another, while not liking either. Berridge (1996) was the first to distinguish between liking and desire to eat. As mentioned, liking is an affective response to a food, for example, “it tastes nice”

In-home testing

77

and is assumed to be independent from the context in which it is consumed. Desire refers to (immediately) wanting or intending to consume a food, and is influenced by general liking for a food, one’s current physiological state (e.g., being hungry), and the situational context (time of the day, perceived appropriateness). For example, you may like red wine, but feel no desire to drink it at breakfast (Mela, 2000). These two types of phenomena require two different approaches to solve them if there is a problem with it. If there is a decrease in liking (i.e., boredom with the product), the formulation of the product will need to be adjusted to make it less susceptible to this effect, whereas a decrease in desire (i.e., boredom with the concept) can potentially be addressed through the marketing and positioning of the product (e.g., adjust the concept, situation or range; Zandstra et al., 2004). It is possible that the liking for a product is quite high and stable, and that the desire to eat this product varies considerably, or vice versa. It is therefore important to influence both the “desire to eat” a food and “liking” for a food, and to manage the expectations in the direction of a favorable evaluation of the product. In a laboratory experiment, we investigated the impact of labeling and repeated consumption on desire and liking of freshly prepared chicken bouillons (Zandstra, 2003). All subjects (n ¼ 100) received the same bouillon over the whole period. For each group, half of the group tasted the bouillon blind, and the other half tasted the bouillon labeled. When the bouillon was tasted with labeling, it was labeled as a “freshly home-made chicken bouillon.” Subjects drank 100 mL of the bouillon 12 times over a period of 4 weeks. Desire and liking were measured at each consumption. Liking of the bouillons was quite high and remained constant over time. There were no differences in liking among the bouillons (both blind and labeled; P > .05). However, the label “fresh” had a positive impact on the desire ratings, both initially and over repeated consumption (see Fig. 4.4). The freshly prepared chicken bouillon with the label “fresh” was significantly more desired than the freshly prepared chicken bouillon when tasted without labeling (P > .05). The label “fresh” did not influence actual product liking (P < .05). This study demonstrates the power of expectations as measured with desire: a product labeled as “fresh” is more desired by consumers, even though the product itself is not liked more. In addition to measuring desire, liking, and/or preference, it is important to carry out sensory profiling of products as well to obtain technical product factors. This is also essential to ensure that the products used in the in-home test significantly differ on those attributes that drive liking (otherwise, no difference can be measured).

4.1.2.6 Home use test versus central location test The context in which food is selected and consumed is an important factor in its acceptability, choice, and consumption (De Graaf et al., 2005; Edwards, Meiselman, Edwards, & Lesher, 2003; King, Meiselman, Hottenstein, Work, & Cronk, 2007; Meiselman, De Graaf, & Lesher, 2000; Meiselman, Johnson, et al., 2000). For example, Edwards et al. (2003) showed that products are experienced differently in different contexts: context contributed significantly to overall acceptability of a dish with chicken and rice, with even a one-point difference on a nine-point hedonic scale between the highest (upscale restaurant) and lowest ratings

Context 7

7

6

6

5

5 Liking

Desire

78

4

4

3

3

2

2

1

1 1 2 3 4 5 6 7 8 9 10 11 12

1 2 3 4 5 6 7 8 9 10 11 12

Time (days)

Time (days)

Blind

Labeled 'fresh'

Blind

Labeled 'fresh'

Fig. 4.4 Mean desire-to-eat (left) and liking (right) ratings of chicken bouillons over time (4 weeks, three exposures per week), tasted blind or labeled as “freshly home-made chicken bouillon”, measured on 7-point scales at a central location (from 1 ¼ “no desire at all” to 7 ¼ “very much desire” and 1 ¼ “don’t like it at all” to 7 ¼ “like it extremely”).

(army training camp) of the dish. In a similar vein, hedonic ratings elicited in an in-home test in which consumers use the product in a realistic consumption context have shown to be consistently higher than those elicited in central location tests under controlled sensory laboratory conditions (Boutrolle, Arranz, Rogeaux, & Delarue, 2005; Holthuysen et al., 2017; Petit & Sieffermann, 2007; Willems et al., 2014), although this appears to be dependent on product type as well (Boutrolle et al., 2007). Consequently, it is strongly recommended to assess foods in realistic contexts to increase the external validity of the results of these consumer tests (Meiselman, 2013). However, accurate context testing as an integral part of the new product development cycle comes at a price, as these in-home tests are rather time consuming, relatively burdensome, and therefore expensive to conduct compared with classical central location tests. Also, there is little control on how the product is being tested (i.e., the experimenters do not know who really tasted and rated the product and under what circumstances). A solution would be to simulate these contexts in the laboratory, which would not only be cost-effective, but would also combine the increased experimental control of the laboratory with the increased realism of the simulated context (Holthuysen et al., 2017). Recently, the incorporation of context into sensory and consumer testing has received more and more attention. For example, situational simulation methods have tried to evoke a consumption context either by means of providing a descriptive text (Hein, Hamid, Jaeger, & Delahunty, 2010; Hein, Hamid, Jaeger, & Delahunty, 2012;

In-home testing

79

L
evy & K€ oster, 1999), a picture (Hersleth, Monteleone, Segtnan, & Naes, 2012), or a combination of a descriptive text and picture (Pisqueras-Fiszman & Jaeger, 2014). In these studies, making use of evoked consumption contexts, differences in mean hedonic ratings, and an increased sample discrimination have been observed (see also Kim, Lee, & Kim, 2016). A new and promising development to create immersive simulated contexts is the use of augmented and virtual reality in sensory and consumer testing (see Section 4.2).

4.2

Bring the world into the lab—creating standardized, real-life contexts

4.2.1 Consumer testing in immersive simulated contexts 4.2.1.1 The future: virtual reality for early-stage product testing Virtual reality (VR) can be defined as “computer technology that uses headsets or multi-projected environments, sometimes in combination with physical environments or props to generate images, sounds, and other sensations that simulate a user’s physical presence in a virtual or imaginary environment” (Wikipedia, 2018). Augmented reality (AR) can be defined as “a combination of real and virtual objects in a real environment, interactively, and in real time, and aligning real and virtual objects” (Azuma et al., 2001). Recent technological advancements in computing power, display technology, and computer generated imagery (CGI) have created breakthroughs in the development of VR/AR devices, with large tech companies such as Google, Facebook, Sony, HTC, and Samsung heavily investing in VR/AR technologies and software. Virtual reality (VR), or augmented reality (AR) technologies offer increasingly realistic ways to interact with virtual products, virtual people, and virtual situations, and have already started to make inroads into gaming and marketing applications (see also Chapter 16 by Hartmann and Siegrist and Chapter 23 by Hehn et al.). VR allows researchers to create more realistic and immersive, yet highly controlled environments (e.g., recreating a coffee bar to test different kinds of coffee, as opposed to a central testing location; Bangcuyo et al., 2015). VR therefore offers unique opportunities for in-context research, especially where “real” products or situations are expensive (e.g., making expensive prototypes) or difficult to control (e.g., a consumer test in a coffee shop). One of the key benefits of VR/AR is that the context of use can be more realistic and immersive compared with mock-ups or 2D-visualizations, and participants’ behaviors are likely to be more predictive of people’s responses in real life, compared with standard consumer research or training settings (Bordegoni & Ferrise, 2013; Bordegoni, Ferrise, & Lizaranzu, 2011; Carulli, Bordegoni, & Cugini, 2016; Li, Daugherty, & Biocca, 2001). However, there are also some large hurdles that need to be overcome, before VR/ AR can really fulfill their promises. Although the major advantage would be that the whole stimulus environment—the context and the products—is under complete control of the experimenter, a key hurdle lies in the fact that many foods are soft,

80

Context

deformable, or fluid. Creating computer-generated interactive deformable or fluid objects, however, is very computationally intensive, and therefore still relatively limited in its realism – even though big steps are being made. Furthermore, the haptics involved in virtual reality, let alone with deformable objects or fluids, are a still relatively uninvestigated area (e.g., Zhang & Liu, 2017). Another area that warrants further research regards the validity of research conducted in a VR/AR setting: does research with real objects and experiences provide the same results as research with virtual objects and experiences (Bangcuyo et al., 2015; Kim et al., 2016; Luigi, Massimiliano, Aniello, Gennaro, & Virginia, 2015; Ung, Menozzi, Hartmann, & Siegrist, 2018). One key aspect in that is how realistic the objects should be for the outcome of this research to be equivalent to research with real objects (Portman, Natapov, & Fisher-Gewirtzman, 2015). Although the ideal would be to have products that are indistinguishable, this will not be possible in the near future. Thus, how realistic should the prototypes be? When does people’s sense of “embodiment”—the feeling that one’s virtual presence is one’s own—break down (Kilteni, Groten, & Slater, 2012)?

4.3

Conclusion

This chapter summarized and reviewed current knowledge and practical aspects of in-home testing of foods. We described different ways of measuring in-home behavior, as well as underlying mechanisms and practical considerations that should be taken into account when designing a study. The intention of this chapter is not to give an exhaustive description on conducting in-home tests—rather, we hope that this chapter made clear that there is no one-size-fits-all methodology for conducting in-home testing: each new product testing situation requires careful consideration of the (dis)advantages of different study designs and measurement decisions. Furthermore, we have highlighted a few areas for future research. There are two key benefits of conducting in-home tests. The first is that products are evaluated by consumers in a real-life setting. Furthermore, the evaluation is typically based on multiple exposures over weeks or months. Both aspects enhance the ecological validity of the results and can prevent the launch of products that are doomed to fail. We have shown a number of aspects and mechanisms that should be taken into consideration when designing a study, such as learned liking (an increase in liking over time), boredom (a decrease in liking over time), appropriateness and context of use, expectations, and different ways of measuring consumers’ evaluations (liking, desire, preference). In highlighting the different aspects that should be taken into consideration, it is clear that there are also important gaps for further research. A key area that warrants further research is how liking evolves dynamically, and what factors influence this, to provide a fundamental understanding of how people process and respond to certain stimuli over time in ecologically valid “real-life” contexts. Furthermore, although we know that expectations play a key role in people’s evaluations, we do not fully understand what drives assimilation and contrast effects, making it difficult to fully

In-home testing

81

control for that in designing a stimulus set. How can we ensure that we understand what role contrast/assimilation effects have in influencing the test scores? What (statistical) measures can be deployed to deal with this? The ecological validity is one of the main benefits of an in-home study. However, they also make the test relatively expensive, and there is little control over whether respondents follow instructions. Thus, one of the areas that is being investigated is the use of VR, which promises to provide a realistic context that is under the control of the researcher. As these tools become more realistic and practical to deploy, we will undoubtedly gain new insights that will help create healthy, tasty, and sustainable foods that are liked again and again.

References Azuma, R., Baillot, Y., Behringer, R., Feiner, S., Julier, S., & MacIntyre, B. (2001). Recent advances in augmented reality. IEEE Computer Graphics and Applications, 21(6), 34–47. Bangcuyo, R. G., Smith, K. J., Zumach, J. L., Pierce, A., Guttman, G. A., & Simons, C. (2015). The use of immersive technologies to improve consumer testing: The role of ecological validity, context and engagement in evaluating coffee. Food Quality and Preference, 41, 84–95. Bellisle, F., Giachetti, I., & Tournier, A. (1988). Comment d
eterminer la salinit
e d’un aliment: comparaison de l’
evaluation sensorielle et des tests de consummation. Sciences des Aliments, 8, 557–564. Bernstein, I. L., & Webster, M. M. (1980). Learned taste aversions in humans. Physiology & Behavior, 25, 363–366. Berridge, K. C. (1996). Food reward: Brain substrates of wanting and liking. Neuroscience and Biobehavioural Reviews, 20, 1–25. Bolles, R. C., Hayward, L., & Crandall, C. (1981). Conditioned taste preferences based on caloric density. Journal of Experimental Psychology: Animal Behavior Processes, 7, 59–69. Booth, D. A., Mather, P., & Fuller, J. (1982). Starch content of ordinary foods associatively conditions human appetite and satiation, indexed by intake and eating pleasantness of starch-paired flavours. Appetite, 3, 163–184. Bordegoni, M., & Ferrise, F. (2013). Designing interaction with consumer products in a multisensory VR environment. Virtual and Physical Prototyping, 8, 51–64. Bordegoni, M., Ferrise, F., & Lizaranzu, J. (2011). The use of interactive Virtual Prototypes for product design specifications: A pilot study on consumer products. In: Proceedings of the IEEE virtual reality conference (VR), pp. 197–198. Bornstein, R. F. (1989). Exposure and affect: Overview and meta-analysis of research, 1968–1987. Psychological Bulletin, 106, 265–289. Boutrolle, I., Arranz, D., Rogeaux, M., & Delarue, J. (2005). Comparing central location test and home use test results: Application of a new criterion. Food Quality & Preference, 16, 704–713. Boutrolle, I., Delarue, J., Arranz, D., Rogeaux, M., & K€oster, E. P. (2007). Central location test vs. home use test: Contrasting results depending on product type. Food Quality & Preference, 18, 490–499. Brunstrom, J. M. (2005). Dietary learning in humans: Directions for future research. Physiology & Behavior, 85, 57–65.

82

Context

Cardello, A. V. (1994). Consumer expectations and their role in food acceptance. In H. J. H. MacFie & D. M. H. Thomson (Eds.), Measurement of food preferences (pp. 253–297). London: Blackie academic. Cardello, A. V., Schutz, H., Snow, C., & Lesher, L. (2000). Predictors of food acceptance, consumption and satisfaction in specific eating situations. Food Quality and Preference, 11, 201–216. Carulli, M., Bordegoni, M., & Cugini, U. (2016). Integrating scents simulation in virtual reality multisensory environment for industrial product evaluation. Computer-Aided Design & Applications, 13, 320–328. Crandall, C. S. (1984). The liking of foods as a result of exposure: Eating doughnuts in Alaska. Journal of Social Psychology, 125, 187–194. Daillant, B., & Issanchou, S. (1991). Most preferred level of sugar: Rapid measure and consumption test. Journal of Sensory Studies, 6, 131–144. Davies, S. R., El-Deredy, W., Zandstra, E. H., & Blanchette, I. (2012). Evidence for the role of cognitive resources in flavour-flavour evaluative conditioning. Quarterly Journal of Experimental Psychology, 65(12), 2297–2308. De Castro, J. M. (1990). Social facilitation of duration and size but not rate of the spontaneous meal intake in humans. Physiology & Behavior, 47, 1129–1135. De Castro, J. M., & De Castro, E. S. (1989). Spontaneous meal patterns in humans: Influence of the presence of other people. American Journal of Clinical Nutrition, 50, 237–247. De Graaf, C., Cardello, A. V., Kramer, M., Fisher, L. L., Meiselman, H. L., & Schutz, H. G. (2005). A comparison between liking ratings obtained under laboratory and field conditions: The role of choice. Appetite, 44, 15–22. Deci, E. L., & Ryan, R. M. (1987). The support of autonomy and the control of behavior. Journal of Personality and Social Psychology, 53, 1024–1037. Deliza, R., & MacFie, H. J. H. (1996). The generation of sensory expectation by external cues and its effect on sensory perception and hedonic ratings: A review. Journal of Sensory Studies, 11, 103–128. Edwards, J. S. A., Meiselman, H. L., Edwards, A., & Lesher, L. (2003). The influence of eating location on the acceptability of identically prepared foods. Food Quality and Preference, 14, 647–652. Fine, G. A., & Elsbach, K. D. (2000). Ethography and experiment in social psychological theory building: Tactics for integrating qualitative field data with quantitative lab data. Journal of Experimental Social Psychology, 36, 51–76. Gould, N. J., Zandstra, E. H., & Yeomans, M. R. (2017). Knowing too much: Knowledge of energy content prevents liking change through flavour-nutrient associations. Quarterly Journal of Experimental Psychology, 1–10. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2010). Application of a written scenario to evoke a consumption context in a laboratory setting: Effects on hedonic ratings. Food Quality & Preference, 21, 410–416. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2012). Effects of evoked consumption contexts on hedonic ratings: A case study with two fruit beverages. Food Quality & Preference, 26, 35–44. Hersleth, M., Monteleone, E., Segtnan, A., & Naes, T. (2012). Consumers’ acceptance and probability of buying dry cured ham in two different meal contexts – Effects of intrinsic and extrinsic product attributes. In: Oral presentation at the fifth European conference on sensory and consumer research, Bern, Switzerland, 9–12 September 2012. Hetherington, M. M., Bell, A., & Rolls, B. J. (2000). Effects of repeat consumption on pleasantness, preference and intake. British Food Journal, 102, 507–521.

In-home testing

83

Hetherington, M. M., Pirie, L. M., & Nabb, S. (2002). Stimulus satiation: Effects of repeated exposure to foods on pleasantness and intake. Appetite, 38, 19–28. Holthuysen, N. T. E., Vrijhof, M. N., De Wijk, R. A., & Kremer, S. (2017). Welcome on board’: Overall liking and just-about-right ratings of airplane meals in three different consumption contexts – Laboratory, re-created airplane, and actual airplane’. Journal of Sensory Studies, 32, e12254. Kahneman, D., & Snell, J. (1992). Predicting a changing taste: Do people know what they will like? Journal of Behavioral Decision Making, 5, 187–200. Kamen, J. M., & Peryam, D. R. (1961). Acceptability of repetitive diets. Food Technology, 15, 173–177. Kilteni, K., Groten, R., & Slater, M. (2012). The sense of embodiment in virtual reality. Presence, 21, 373–387. Kim, S. E., Lee, S. M., & Kim, K. O. (2016). Consumer acceptability of coffee as affected by situational conditions and involvement. Food Quality and Preference, 52, 124–132. King, S. C., Meiselman, H. L., Hottenstein, A. W., Work, T. M., & Cronk, V. (2007). The effects of contextual variables on food acceptability: A confirmatory study. Food Quality & Preference, 18, 58–65. Kramer, F. M., Lesher, L. L., & Meiselman, H. L. (2001). Monotony and choice: Repeated serving of the same item to soldiers under field conditions. Appetite, 36, 239–240. Kuenzel, J., Blanchette, I., Lion, R., Zandstra, E. H., Thomas, A., & El-Deredy, W. (2011). Conditioning specific positive states to unfamiliar flavours influences flavour liking. Food Quality & Preference, 22, 397–403. Kuenzel, J., Zandstra, E. H., El-Deredy, W., Blanchette, I., & Thomas, A. (2011). Expecting yoghurt drinks to taste sweet or pleasant increases liking. Appetite, 56, 122–127. Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review. Psychological Bulletin, 10(1), 34–47. L
evy, C. M., & K€oster, E. P. (1999). The relevance of initial hedonic judgements in the prediction of subtle food choices. Food Quality and Preference, 10, 185–200. Li, H., Daugherty, T., & Biocca, F. (2001). Characteristics of virtual experience in electronic commerce: A protocol analysis. Journal of Interactive Marketing, 15(3), 13–30. Lowe, C. F., Horne, P., Tapper, K., Bowdery, M., & Egerton, C. (2004). Effects of a peer modelling and rewards-based intervention to increase fruit and vegetable consumption in children. European Journal of Clinical Nutrition, 58, 510–522. Luigi, M., Massimiliano, M., Aniello, P., Gennaro, R., & Virginia, P. R. (2015). On the validity of immersive virtual reality as tool for multisensory evaluation of urban spaces. Energy Procedia, 78, 471–476. Maxwell, J. A. (2009). Designing a qualitative study. In L. Bickman & D. J. Eog (Eds.), The SAGE handbook of applied social research methods (pp. 214–253). Thousand Oaks CA: SAGE publications Ltd. Meiselman, H. L. (2013). The future in sensory/consumer research:… evolving to a better science. Food Quality & Preference, 27, 208–214. Meiselman, H. L., De Graaf, C., & Lesher, L. L. (2000). The effects of variety and monotony on food acceptance and intake in a midday meal. Physiology & Behaviour, 70, 119–125. Meiselman, H. L., Johnson, J. L., Reeve, W., & Crouch, J. E. (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35, 231–237. Mela, D. J. (2000). Why do we like what we like? Journal of the Science of Food and Agriculture, 81, 10–16. Pelchat, M. L., & Rozin, P. (1982). The special role of nausea in the acquisition of food likes in humans. Appetite, 3, 341–351.

84

Context

P
erez, C., Dalix, A., Guy-grand, B., & Bellisle, F. (1994). Human responses to five concentrations of sucrose in a dairy product: Immediate and delayed palatability effects. Appetite, 23, 165–178. Petit, C., & Sieffermann, J. M. (2007). Testing consumer preferences for iced-coffee: Does the drinking environment have any influence? Food Quality & Preference, 18, 161–172. Pisqueras-Fiszman, B., & Jaeger, S. R. (2014). Emotion responses under evoked consumption contexts: A focus on the consumers’ frequency of product consumption and the stability of responses. Food Quality & Preference, 35, 24–31. Pliner, P. (1982). The effects of mere exposure on liking for edible substances. Appetite, 3, 283–290. Pliner, P., & Chaiken, S. (1990). Eating, social motives, and self-presentation in women and men. Journal of Experimental Social Psychology, 26(3), 240–254. Pliner, P., Polivy, J., Herman, C. P., & Zakalusny, I. (1980). Short-term intake of overweight individuals and normal weight dieters and non-dieters with and without choice among a variety of foods. Appetite, 1, 203–213. Popper, R., Maller, O., & Cardello, A. V. (1989). Hedonic ratings and consumption as measures of preference for a salted entr
ee. Chemical Senses, 14, 739. Porcherot, C., & Issanchou, S. (1998). Dynamics of liking for flavoured crackers: Test of predictive value of a boredom test. Food Quality and Preference, 9, 21–29. Portman, M. E., Natapov, A., & Fisher-Gewirtzman, D. (2015). To go where no man has gone before: Virtual reality in architecture, landscape architecture and environmental planning. Computers, Environment and Urban Systems, 54, 376–384. Richardson, J. T. E. (1996). Handbook of qualitative research methods for psychology and the social sciences. Leicester: PBS Books. Rozin, P., & Markwith, M. (1991). Cross-domain variety seeking in human food choice. Appetite, 16, 57–59. Schutz, H. G. (1994). Appropriateness as a measure of the cognitive-contextual aspects of food acceptance. In H. J. H. M. Fie & D. M. H. Thomson (Eds.), Measurement of food preferences (pp. 25–50). London, UK: Blackie Academic and Professional. Sclafani, A. (1997). Learned controls of ingestive behaviour. Appetite, 29, 153–158. Shepherd, R., Farleigh, C. A., & Warf, S. G. (1991). Effect of quantity consumed on measures of liking for salt concentrations in soup. Journal of Sensory Studies, 6, 227–238. Stang, D. J. (1975). Effects of mere exposure on learning and affect. Journal of Personality and Social Psychology, 31, 7–12. Stroebele, N., & De Castro, J. M. (2004). Effect of ambience on food intake and food choice. Nutrition, 20, 821–838. Stubenitsky, K., Aaron, J. I., Catt, S. L., & Mela, D. J. (1999). Effect of information and extended use on the acceptance of reduced-fat products. Food Quality and Preference, 10, 367–376. Stubenitsky, K., Zandstra, L. H., Elliman, N. A., de Graaf, C., Smit, H. J., Rogers, P. J., et al. (2000). No development of energy-based conditioned flavour preferences in human adults under realistic eating conditions. Appetite, 35, 311. Ung, C. Y., Menozzi, M., Hartmann, C., & Siegrist, M. (2018). Innovations in consumer research: The virtual food buffet. Food Quality and Preference, 63, 12–17. Van Trijp, H. C. M. (1994). Product-related determinants of variety-seeking behavior for foods. Appetite, 22, 1–10. Van Trijp, H. C. M., L€ahteenm€aki, L., & Tuorila, H. (1992). Variety seeking in the consumption of spread and cheese. Appetite, 18, 155–164. Van Trijp, H. C. M., & Steenkamp, J.-B. E. M. (1992). Consumers’ variety seeking tendency with respect to foods: Measurement and managerial implications. European Review of Agricultural Economics, 19, 181–195.

In-home testing

85

Vartanian, L. R. (2015). Impression management and food intake. Current directions in research. Appetite, 86, 74–80. Vartanian, L. R., Herman, C. P., & Polivy, J. (2007). Consumption stereotypes and impression management: How you are what you eat. Appetite, 48, 265–277. Verplanken, B. (2006). Beyond frequency: Habit as a mental construct. British Journal of Social Psychology, 45, 639–656. Verplanken, B., & Wood, W. (2006). Interventions to break and create consumer habits. Journal of Public Policy and Marketing, 25, 90–103. Wikipedia (2018). Virtual reality. https://en.wikipedia.org/wiki/Virtual_reality. Willems, A. A., Van Hout, D. A., Zijlstra, N., & Zandstra, E. H. (2014). Effect of salt labelling and repeated in-home consumption on long-term liking of reduced-salt soups. Public Health Nutrition, 17(5), 1130–1137. Yeomans, M. R. (2006). The role of learning in development of food preferences. In R. Shepherd & M. Raats (Eds.), Psychology of food choice (pp. 93–112). Oxford, UK: Centre for Agriculture and Bioscience International. Yeomans, M. R. (2012). Flavour-nutrient learning in humans: An elusive phenomenon? Physiology & Behavior, 106, 345–355. Zajonc, R. B. (1968). Attitudinal effects of mere exposure. Journal of Personality and Social Psychology Monograph Supplement, 9, 1–27. Zandstra, E. H. (2003). Effects of labelling and repeated consumption on desire and liking over time. In: Oral presentation at the 5th Pangborn sensory science symposium, Boston (USA) abstract book. Zandstra, E. H., De Graaf, C., & Van Trijp, J. C. M. (2000). Effects of variety and repeated in-home consumption on product acceptance. Appetite, 35, 113–119. Zandstra, E. H., De Graaf, C., Van Trijp, J. C. M., & Van Staveren, W. A. (1999). Laboratory hedonic ratings as predictors of consumption. Food Quality and Preference, 10, 411–418. Zandstra, E. H., & El-Deredy, W. (2011). Effects of energy conditioning on food preferences and choice. Appetite, 57, 45–49. Zandstra, E. H., Weegels, M. F., Van Spronsen, A. A., & Klerk, M. (2004). Scoring or boring? Predicting boredom through repeated in-home consumption. Food Quality & Preference, 15, 549–557. Zhang, X., & Liu, S. (2017). SPH haptic interaction with multiple-fluid simulation. Virtual Reality, 21, 165–175.

Further reading Mattes, R. D., & Mela, D. J. (1986). Relationships between and among selected measures of sweet-taste preference and dietary intake. Chemical Senses, 11(4), 523–539. Moskowitz, H. R. (2000). Engineering out food boredom: A product development approach that combines home use tests and time-preference analysis. Food Quality and Preference, 11, 445–456. Zandstra, E. H., De Graaf, C., & Van Staveren, W. A. (2001). Influence of health and taste attitudes on consumption of low- and high-fat foods. Food Quality & Preference, 12, 75–82.

This page intentionally left blank

Useful observational research Brian Wansink Cornell University, Ithaca, NY, United States

5.1

5

Introduction

People do not always know what they want. When it comes to improving a consumer service, experience, or product in a specific way, asking consumers for their advice is not always helpful—nor does it lead to breakthrough solutions. For example, Disney World vacationers might not know how to make the Epcot Center attraction they just saw be more educational, and McDonald’s lovers might not have any new ideas on how a Happy Meal could be both healthier and happier. Similarly, they cannot imagine how a European hotel chain could serve more environmentally responsible breakfast buffets, or how a library could help people study better. Henry Ford famously said, “If I’d have asked my customers what they wanted, they would have told me ‘A faster horse.’” You cannot always ask consumers what they want, but you can watch them. Using a type of iterative observation, called storytelling research, this chapter shows how you can design, redesign, or modify a service, product, or experience so that it is better for them (Abrams, 2000). You can make it more educational, more healthy, more fun, or more valuable. This method can show how small changes in the context of where a person shops, eats, works, or plays can more predictably guide their behavior (Crosby, Salazar, & DiClemente, 2015). Consider all-you-can-eat buffets. Although they are allegedly notorious for encouraging gluttony, at least a third of the people eating at buffets are very slim (Wansink, 2014). If we knew what they did differently than heavy people, there might be useful insights for both diners who want to eat less, as well as for buffet owners who want to serve and waste less food. The problem is that you cannot simply ask a slim buffetgoer what they think they do differently than a heavier diner, because they would answer either “I don’t know” or “I guess I eat less.” They may know that they should not only eat fried or fatty foods, but they may not know what else they can do. Eating at a buffet can be a very “mindless,” or low-involvement eating situation for many people (Wansink, Just, & Payne, 2009). Although they cannot tell us what they do, they can show us. Observation research can be conducted qualitatively or quantitatively (Girard & Cohn, 2016). Although it can be quantitative, its results are only correlational, and not causal (Fowler & Montagnes, 2015; Gerber, Green, & Kaplan, 2014). Qualitative observation involves observing, inferring, and speculating in a way that generates inferences or hypotheses (Merriam & Tisdell, 2015). It might be observed that a lot of slim people seem to scout out all the food on the buffet before picking up a plate, or that they sit far away from the buffet. Quantitative observations involve collecting Context. https://doi.org/10.1016/B978-0-12-814495-4.00005-2 Copyright © 2019 Brian Wansink. Published by Elsevier Inc. All rights reserved.

88

Context

these inferences and hypotheses together into a coding sheet and observing dozens to hundreds of consumers to confirm or disconfirm this hunch (Brannen, 2017). Useful observations are sometimes elusive or difficult to find because researchers often do not know what to observe in the first place (Abrams, 2000). How does a researcher know what they should be looking for? Two common methods are to use unstructured observations, and to use hypothesis-based observation (hypotheses from prior studies or brainstorming). An often overlooked, but very useful approach is a storytelling method. In all of these problem-focused cases, there is a three-step procedure to observational research that involves answering three questions (see Fig. 5.1): 1. What specific segment is most relevant to this problem? 2. What is influencing these people to behave? 3. What specific behaviors are most important?

If the problem is clearly and specifically stated, the first and third questions will seem initially easy to answer. For the first question, there might be multiple segments, and they might become more narrowly defined once observations begin, but at least it is somewhat clear where to start. Similarly, for the third question, the problem usually contains a specific behavior to change—decreasing the frequency that people use illegal drugs at a train station, or decreasing how much food people waste at a buffet. Although these questions seem straightforward, an insightful researcher knows that there are also specific gateway behaviors that they would also observe (such as how many hours people stay in the train station or whether they stay overnight). Avoiding these gateway behaviors may very well be an easier solution to solving the problem. While the first and third questions are reasonably easy to answer when beginning observational research, the difficulty is in breaking open the Black Box that tells us these non-obvious gateway behaviors or environmental features that will really help solve this problem.

Fig. 5.1 Useful observational coding requires you to ask three questions.

Useful observational research

89

To try and open this Black Box, one approach is to have an existing hypothesis and observe this behavior (see Fig. 5.1). A researcher might hypothesize that European train stations that are surrounded with retail service shops that have long hours (such as cafes, shopping, bookstores, and so on) also have the least drug use. A second approach would be to passively observe people and to look for emerging patterns. This can take a patient, experienced, and highly trained researcher. While this is a more academic approach, the ideas are not always actionable, nor the occurrence satisfactorily explainable to those untrained in that field. A third approach (Option C in Fig. 5.1) is to use active observation—and storytelling—to develop a wide range of actionable hypotheses for coding. This will be the focus of the next section of this chapter. After showing how to generate insights for coding, the next section shows how to develop a coding sheet, and how to collect and analyze this data. In the final part of this chapter, a real-life example (including web links to videotaped behaviors) enables readers to practice how they can use observational analysis to help correct a failed Disney attraction.

5.2

Generating insights for coding categories

This chapter focuses on unobtrusive observational research—observing people from a distance, on tape, or as a “mystery shopper.” Observation research offers a promising way to better understand the unconscious or “mindless” actions of car shoppers, app users, home do-it-yourselfers, or school lunch diners. l

l

l

l

How How How How

could a library be redesigned to encourage more efficient studying? can a stadium be changed to encourage more season ticket holders? can a website be altered to encourage more purchasing? can an app be altered to be more engaging?

What has disappointed many companies and researchers on this promise from observational research (as well as with other types of research) has been its inability to consistently deliver actionable insights. Part of this may be because it is too unstructured, because it is not often quantified, and because most researchers cannot use it to provide actionable solutions. Observational research takes different forms. From a cultural anthropology perspective, observational research can help provide deeper meaning to behavior in cultures and subcultures. This research is often built on grounded theory, and has an academic priority. A second type of observational research is focused on problem solving. That is, after identifying a specific population, how can observing this population give us insights on potential solutions to problems in a wide range of contexts. Public Health. This type of problem solving could be focused on problems relevant to public health (how can buffets encourage people to eat less food, how can drug use in European train stations be discouraged among backpackers, or how can fast food kids’ meals be both healthy and “Happy”?). Business. This type of problem solving could be focused on problems relevant to businesses (how can the shopping mall encourage people to shop longer or how can Disney attractions be more engaging for adults?).

90

Context

Nonprofit Services. This type of problem solving could be focused on helping a nonprofit service organization be more effective (how can the local library help patrons focus better when they study or how can the Interlaken Food Bank make it easier for guests to choose more balanced meals?). It is important to realize that a problem-solving approach can be synergistic with a theoretical approach. An academic or theoretical starting point can provide a practical solution, and a problem-solving stating point can generate academic and theoretical insights and articles.

5.2.1 Using the storytelling method to generate insights for coding One way to overcome these barriers is to use the storytelling method to actively understand a person’s unstated needs. It is then used to explore how previous actions they make might lead to the next action (which may be the one we want to prevent or to change). This active process of storytelling involves developing an evolving “story” about the observed situation by focusing on an attention to detail that would otherwise go unnoticed with traditional methods. It has proven useful in generating more creative observations. It has also proven useful in generating more creative potential solutions. The storytelling method can be made even more useful if multiple people are observed separately by multiple observers. Again, the purpose of this is not to confirm an objective truth. It is to develop insights that can be used to create quantitative coding sheets that can then be used by a larger number of observers to code and to confirm or disconfirm these initial insights. The basic procedure of using storytelling is as follows: 1. Select one person in the target context and target population to unobtrusively observe for a period of time. 2. Begin the observations with a careful description of the subject’s appearance—especially easy to miss details, such as the condition of wear of their shoes or how they are accessorized. The key purpose of this storytelling method, however, is in speculating why they do what they do. This will provide insight into potential motivations and values. At this initial observation level, this is done by asking yourself speculative questions regarding why they are dressed a certain way, or look the way they look. In addition to noting how they are accessorized, ask “Why are they wearing colored earrings?” “Why did they get a tattoo of a person’s name?” Additionally, try and speculate what has occurred in their day until the moment you began observing. What did they do when they woke up? What did they have for breakfast, or what did they listen to or do during their commute or during their travel? 3. Try and detail each of their distinct behaviors or actions on the coding sheet under a heading of “observations.” What’s most important is to speculate inferences as to why they did what they did. Your answers in step 2 can help in this regard because they may have given some speculative insight into their values and motivations. In some cases, you may want to limit yourself to focusing on specific behaviors (eating french fries), but in order to do even that, the inferences about the other behaviors while ordering and eating the children’s meal should be analyzed. (For example, if kids stop eating french fries once they begin playing with the toy that comes with their meal, that can suggest key insights).

Useful observational research

91

4. Create Inferences. When time allows, go back to each of the distinct behaviors and make an initial subjective inference that explains the action. An inference explains why that action occurred instead of a different one. This provides the framework for the story. This may include descriptions of environmental cues potentially stimulating the behavior. 5. Repeat the process with each discrete observed action. 6. Creating hypotheses. Use the inferences about the customer to create a vivid hypothesis that can be tested with further research. Compared with an inference, a hypothesis is broader in scope and generalizable to other segments of the population. This may relate to the overall theoretical reason for the action of the observed consumer, and may explain what characteristic of the person caused them to behave as they did. Hypotheses may be used to target that customer. Described another way, a hypothesis is a broad theme that predicts associations between actions that can be tested with research. Hypotheses can also identify relevant variables to include in a coding sheet for a later study.

In creating inferences, the important step is to continually ask why an observed action was taken instead of a different action. Why did this person use four ketchup packets? is a less insightful question than: Why did this person use four ketchup packets instead of two packets or six packets?

5.2.2 Creating stories that explain behavior patterns Critical to the storytelling method is knowing that the creative process does not need to be fully accurate to be valuable (Fortini-Campbell, 1991). Given that it is impossible to separate the recorded observations from the biases of its researchers (Grimes & Schulz, 2002; Morgan, Pullon, Macdonald, McKinlay, & Gray, 2017), we instead force the biases to work toward our purposes. Insights naturally arise from our own experience, which we may not be able to express if we were to try and describe our own behavior. Storytelling allows a researcher to discover a theory he or she might not have been prompted to create without an external stimulus of observations. Consider the following description of a teenager in the food court of a mall. 4:15 pm- Teenage male with an older male and female, likely his parents. Orders a slice of “everything” pizza with a large drink. Began eating right as he sat down and did not remove the plate from his tray before eating. Hunched over and not talking to his parents, who are not eating anything. Ate the slice of pizza methodically; from front to back without eating any of the crust until all of the cheesy area was gone. Drank his drink only after finishing the pizza. His parents left him at the table and then returned, impatiently. Picked up napkins but did not use any; wiped his mouth on his hands once he was finished and threw away the unused napkins. Entire process took less than five minutes.

On its surface, this is a very simple story of a teen eating pizza in the afternoon. The challenge is to come up with a possible explanation or story. This can be what stimulates insights for potential solutions that can be followed up on later. Dozens of these observations will eventually be made by a number of researchers, but let us focus on this 30 minute observation in order to begin to show how the storytelling method

92

Context

works. Using the storytelling method, here was a story that was created during the observations that might explain these seemingly disconnected observed behaviors: He is a teenager who was taken to the mall by his parents, an occasion that he doesn’t particularly relish. He has no control over the day’s events and is stuck with his parents until they decide to leave. His attempt to regain a measure of control takes the form of complaining of hunger until his parents are forced to stop and feed him, which explains why they aren’t similarly eating. Also, given that it is the afternoon, this is not an actual meal for him. Although he eats a meal-like food (pizza), his behaviors are those of a snack, by not engaging in the mealtime rituals of using napkins, eating slowly, and eating food in parallel with drinking his drink.

One theoretical implication is that the appearance of hunger may be used as a legitimate way to gain control over a situation. An action for retailers to take would be to use new promotions during off-peak hours as a means to help people regain control over unpleasant routines. A second theoretical implication is that people view snacks differently than meals. This distinction is made not by the type of food eaten, but by the rituals they engage in while eating. Given that the pizza was eaten in a snack-like manner, retailers may offer regular food or healthier food as a fill-in snack, and not a meal. Instead of developing snack food as an addition to the menu, coupons may be given to promote smaller versions of regular meals as snacks. In doing so, retailers would save money on development costs of a new food type to target a customer looking for a snack.

5.2.3 Turning stories into coding categories These stories are fictional, but built on observations. They are important, however, because they are a tool to help a researcher think more broadly than they otherwise would. They can range in their detail. The more detailed they are, the more they help a researcher look for insights or hypotheses for follow-up. Table 5.1 shows how different research projects have taken the observations they have made from storytelling, and then translated them into the questions that were asked during each of the main studies. In the context of a story, observations suggest the types of hypotheses to be examined during the main study. Example 1: Developing Healthier Fast Food Kids’ Meals Because of the problems of childhood obesity, an unsponsored exploratory study was done to explore what fast food companies could do to improve the healthfulness of their kids’ meals. Observing 23 children and their parents or guardians eat these meals in three different fast food chains generated stories that then generated observations and insights that kids are eating these meals for fun (that is, they often stop eating them once they open the toy) and that the size of the fries could be reduced without them knowing or caring. Another insight was that even the healthier foods they offered, such as apple slices or carrots, were eaten without needing caramel sauce or ranch dressing as an inducement. To confirm these insights, two different coding sheets were developed. The first one examined how much of the healthy foods (such as apple slices or carrot sticks) kids ate (versus threw away), the order in which they ate them, and whether they ate these foods with caramel sauce

Useful observational research

93

Table 5.1 Developing coding questions from storytelling observations

Objective

Observation (possible implication)

1. How can fast food companies improve both the nutrition and the satisfaction of their Kids’ meals?

When kids open the toys in Kids’ meals, they stop eating, start playing, and they forget the food. (Can french fry portions be reduced in size?)

l

Caramel sauce intended for apple slices is eaten as a candy or with other foods. (If caramel sauce is eliminated, will kids still eat apple slices?)

Kids eat apple slices without caramel. It could be eliminated to save calories and improve nutrition

People using the library for work will find a work space and then kill time browsing in that area before starting work (Would less browsing make them more productive?) People tend to handle only the books that have their cover (not

Some people are easily distracted and are using the library to help motivate them to accomplish work

l

What time does the patron enter and when do they begin working? How many total minutes do they spend at the library and how much time do they spend working?

Work spaces that have nonfiction facing books

l

What types of front-facing books are most plentiful within 40-ft of

2. How can we help library patrons be more efficient and productive in their work?

Hypothesis to confirm Fun is more important than the food. Kids will be equally happy with fewer fries

Questions to include on coding sheets l

l

l

l

Amount of French fries (either percent, number, or handful) eaten before toy is opened Amount of french fries (either percent, number, or handful) eaten after toy is opened Total amount of apple slices eaten (either percent, number, or handfuls) and the amount of apple slices eaten only with caramel sauce How is the caramel sauce eaten ([] with entree, [] with french fries, [] with apple slices, [] by itself, [] taken home, [] thrown out

Continued

94

Context

Table 5.1 Continued

Objective

Observation (possible implication) their spine) facing out. (Would fontfacing nonfiction or how-to books prime patrons to work more?)

3. How can we help heavy or high BMI customers eat less at buffets?

Heavy diners sit in places where the food is most visible and most convenient. (If the food were slightly less visible and convenient would people eat less?) Heavy diners take the largest plates available and don’t browse the food before serving themselves. (Would they eat less if they saw more of the food before serving themselves?)

Hypothesis to confirm (“Get Things Done, “The Seven Habits of Highly Effective People,” etc.) near the work area will promote less browsing and more working Heavy people sit close to the food and where it is most visible. As a result, they may also believe it is more normal to eat a lot because they observe people consistently serving themselves food Instead of selecting favorite foods, heavy people might instead serve themselves “any food that’s acceptable.” This would be further facilitated by a large plate.

Questions to include on coding sheets where the person placed is sitting? [] self-help, [] reference, [] journals, [] other nonfiction, [] fiction, [] other.

l

The relative height, weight, and body type of the diners (circle the closest relevant icon for body shape)

l

Where does the diner sit? (distance to buffet, facing the buffet, table or booth) At the restaurant, what is the size of the buffet plates and where are the buffet plates located relative to food (front, side, back)?

l

or ranch dressing. A second coding sheet examined how kids ate French Fries. In this second case, by observing a large number of families eating in different fast food franchises in different locations, the results supported the notion that the amount of french fries could be decreased in kids’ meals by about 50% without any likely backlash from the children. (Most of these extra fries were eaten by parents or thrown away.) One version of the coding sheet used is shown in Fig. 5.2. These general observations of more than 100 children dining at these three fast food chains were then aggregated. The hope was that these insights would help all fast food restaurant chains confidently reduce the portion size of the french fries in their meals without the fear of

Useful observational research

95

Fig. 5.2 Can french fry portions be reduced in kids’ meals? disappointing the children who want these meals. To disseminate these observations, (a) onesheet summaries of these findings were sent to the top 25 fast food restaurant chains, (b) full discussion of this and other nutritionally relevant findings were presented to companies at the Cornell Food and Brand Lab’s Consumer Camp, and (c) the highlights of the findings were presented at a nutrition conference (Wansink, Hanks, & Stein, 2014). These discoveries helped prompt healthier changes at many top fast food chains, including an 18.8% calorie reduction in children’s meals at McDonalds (Wansink & Hanks, 2014).

Example 2: Helping Library Patrons Be More Productive Many people go to the library—especially large libraries—to work. If a library can make small changes in what it does to encourage patrons to be more productive, that would be good for both patrons and for the library’s popularity. Using the storytelling method, observations were made that some people may be using the library as a way to minimize their distractions and increase their productivity. However, the front-facing books near where they decided to work also provided a distraction, which led some people to

96

Context

spend significant time browsing or bringing the books back to their desk, and browsing them instead of completing their work. To examine this more broadly, the percentage of time a person spent working (versus browsing or taking a break to wander the library) was coded and analyzed relative to the types of front-facing books that were in the proximity. The solutions for this library system involved changing the front-facing books in popular work areas to self-help books or reference books that would prime work and not instead encourage “taking a break.”

Example 3: Helping People Eat Less at Buffets Many people believe that it is difficult to go to an all-you-can-eat buffet without overeating. Yet what we see at many buffets (such as pizza, Chinese, breakfast, or so on) are that at least onethird of the people eating there are slim. If we knew what they did differently than heavy people, dieticians and public health officials might be able to provide suggestions that help diners eat at buffets without overeating. To generate initial insights in this area, obese diners were observed along with slim diners. As Table 5.1 indicates, the storytelling method led to a wide range of observations. One of these was that the more visible (salient) and convenient the food was (and the easier it was for a diner to see other people returning to the buffet), the more normal it would be for them to overeat. If a person habitually sat in places where the food was very visible, salient, and convenient, over the years we might expect them to grow heavier and heavier than a person who always sat far away from the food, or at a location where it was not constantly visible. These observations about the factors related to convenience and visibility were translated into coding sheets that operationalized these general principles into specific hypothesized observations. For example, one coding variable related to convenience would be whether they take a big plate versus a small plate (that requires more refilling), whether they sit at a table versus a booth (that is more difficult to slide in and out of ), or how far they sit from the buffet. Similarly, a coding variable for visibility would be whether a person sits facing the food or facing away from it. To determine whether these insights (or hypotheses) are generalizable, more than 200 diners at Chinese buffets in six different US states were observed (Wansink & Payne, 2008). Fig. 5.3 shows one version of this coding sheet, which was used to discover that the lightest third of the diners (compared with the heaviest third of the diners) differed in their serving behavior (for example, they were more likely to use a small plate versus a large plate, and were more likely to browse the buffet instead of immediately serving themselves), and their eating behavior (for example, they were more likely to sit at a booth versus a table, and more likely to face away from the food instead of face it). Food is a huge expense at all-you-can-eat buffets. People not only overeat, but they often take a lot of food they eventually throw away. If a buffet owner could encourage more diners to behave the way that these slim diners acted, they might be able to save money on food expenses. One owner of more than 60 buffets used these findings to redesign his restaurants (Wansink, 2014). l

l

l

l

l

The buffet owner bought smaller plates for the serving line, downsizing from 11 in. to just under 10. Portions looked bigger, and diners never noticed. He had the waitstaff seat people farthest from the buffet. It made the food less convenient and less tempting. He set chopsticks at every place setting. If you wanted a fork, you had to ask. He put all the plates behind the buffet. People had to walk around the buffet—encouraging them to scout out their favorite dishes before serving. He strategically placed plants and folding Chinese screens between the buffet and the tables to help block some of the food from sight.

Useful observational research

97

Fig. 5.3 What do slim people do differently at buffets? Six months after these changes, one of his team members estimated the annual savings in food expenses would be more than $30,000 per restaurant, or a $1.8 million savings for all of them—each year.

5.3

Coding, collecting, and analyzing observational data

Observational data shows correlational results, not causal ones. Still, the goal of some observation research is to generate insights that may suggest and lead to possible solutions (Goodman, Schneeweiss, & Baiocchi, 2017). Sometimes these new insights can suggest direct action, but sometimes they need to be confirmed by observing a larger sample (Bernard, 2017). In these cases, there needs to be an organized method for multiple researchers to observe multiple people. To do so, coding sheets can systematize

98

Context

these observations in a way that enables them to be quantified (Chorney, McMurtry, Chambers, & Bakeman, 2014). These observation sessions last the length of the encounter in that context: The length of a fast food meal, the length of a Disney attraction, the length of a Chinese buffet dinner, the length of a study session in a library. If the purpose is to generate insights that will be tested in a quantitative survey, a couple dozen observations might be enough to generate hypotheses. If they are going to be used and analyzed quantitatively, then hundreds of observations will be necessary.

5.3.1 Developing a coding sheet To develop a coding sheet, consider the earlier example of trying to determine what slim people do differently at buffets than heavy people. Maybe heavier diners sit closer to the food, face the food (instead of facing away), use forks instead of chopsticks, and maybe they use large plates. Maybe they are also less selective about what they take so they take more different types of foods, and maybe they end up wasting more than other diners. These were all simply observations that were collected during the storytelling observations. Some might have theoretical explanations (for example, it is more convenient to serve yourself without browsing first, sit closer, or use a fork), but others might involve observations that seem to reoccur with no understandable explanation at the time. Our view is to include this item because it may be that with enough data, a pattern can provide a currently elusive explanation. Consider four types of information that you should code: (1) Demographic and identification data, (2) environmental factors, (3) independent behaviors, and (4) dependent variables. 1. Demographic and Identification data. These include the basic demographic data along with the time, location, and date it is being collected. A key feature of some research involving eating is that it can be useful to have the height and weight of people so that an approximate body mass index can be calculated. Having this data, for example, was important when conducting research on what slim diners do differently at buffets than heavy diners. A researcher or observer can be coached to better estimate heights and weights, or they could also use benchmarks (height marks on buffet lines or pressure sensitive mats). In addition, visual display cards, such as the Stunkard Visual Figures Scale can be copied onto the back of coding sheets, enabling a researcher to circle the shape of the observed person (see Fig. 5.4). 2. Environmental Factors. For example, if it is believed that the books that are in the proximity of a work area will either prime a person to work or distract them not to work, it is important to code their type, their distance, and their visibility. As the examples in both Figs. 5.2 and 5.3 show environmental factors on coding sheets can also include who a person is with and how the others around them are acting. 3. Independent Behaviors. If it is believed that the distance a diner sits from a buffet table, the size of the plate they select, or the direction they face might influence their behavior (and could be related to their overall BMI in the long term), it is important to code these. Independent behaviors are often difficult to determine because they are not

Useful observational research

99

Fig. 5.4 Stunkard’s visual figures scale.

always obvious before the fact. Using the storytelling method is one of the best ways to identify these possible independent behaviors. These should be objectively observable “Yes/No” questions whenever possible (did they use chopsticks), or objectively measured (number of feet the person is sitting from the nearest part of the buffet). 4. Dependent variables. When an observation study goes to the stage of being quantitative with a large number of variables, there is generally a specific spotlighted variable that is of interest. It could be how much money a person spends, how long they stay at the location, how much food they waste, whether they make a purchase, whether they order healthy food, and so forth. This is the most critical variable to measure accurately, and it is best if it can be collected in multiple ways. Consider the following multiple ways data can be collected: Wasting Food: What percent is wasted? What foods are wasted? How many different types are wasted? Purchase Behavior: Whether or not a purchase is made ([] Purchases; [] Does not Purchase), How much is spent? How many items are purchased?

How long should a coding questionnaire be? Some researchers only include coding variables that can be theoretically justified. This is generally too few because it only allows for confirmation, but for no new discoveries. Other researchers indiscriminately list nearly everything they can observe, but this extreme also has problems. They often do not know how to analyze this in an insightful way, and they might run the risk of false positives—something being significant by chance. There are a number of solutions of how one can minimize this and still generate unexpected insights. If they make theoretical sense and if they are consistent with other behaviors you would expect to see, then the exploratory finding is probably not a false positive.

100

Context

For example, suppose it is shown that people who face the buffet make more trips to the buffet. It may be that they do this because it is more visible (theoretical explanation), and if people who have an unobstructed view of the buffet also make more trips, this would be another consistent behavior that supports this finding.

5.3.2 Creating a coding sheet The more simple the survey, the more accurate it will probably be. An interesting example is the coding sheet that was developed to measure waste in school cafeterias after students drop off their tray and leave. The gold standard for measuring food waste is to take the amount served and subtract the amount left to provide the amount eaten. A drawback of this is the time and labor needed. Another method is to take photos of the empty tray, but this does not account for the food that is in containers and cannot be observed. A third method is for a coder to estimate what percent of the food has been eaten to the nearest 10th (10%, 20%, etc.). A fourth method, called the quarter-plate waste method, asks coders to only check whether 0%, 25%, 50%, 75%, or 100% of the food has been eaten. Whereas this quarter-plate method provides more crude estimates than the 10% method, it was more quickly done, and prone to fewer errors due to indecision or misestimation (Boschini, Falasconi, Giordano, & Alboni, 2018; Getts, 2016; Getts, Quinn, Johnson, & Otten, 2017; Hanks, Just, & Wansink, 2013). There are three factors that should govern the layout and design of a small, efficient coding sheet. First, the most important variables should be first when possible. Second, the variables should be listed in the most likely order of observation. Third, the flow and the design should be convenient to check or complete with little thought. Coding sheets should be designed in ways that can make it easiest to code people as quickly and accurately as possible. Using electronic coding sheets with a cloud based spreadsheet or a web-based survey sheet is a convenient way to collect coding. These can include using GoogleDocs or using a Dropbox folder, or an online data entry portal, such as BORIS (Friard & Gamba, 2016). They can inconspicuously be conducted with a smart phone or on a computer. There are no skipped questions, no lost coding sheets, no unclear writing, and no data entry or transcription problems. There is also the ability to quickly run analyses to see if the coding sheet needs to be changed, or to give an early reading of the results. Paper-based coding sheets can be plagued with coder errors, missing data, and data entry mistakes. They can also look out of place in a smart phone world. What they do have going for them is that they are not reliant on technology, wi-fi service, connectivity, battery strength, and interrupting calls or texts. They can also allow a person to better observe multiple people without making a mistake—mistakenly coding Person A’s action as being from Person B is less likely to occur on paper, and is easier to retrace and correct if it does happen. Most electronic coding sheets are laid out in a linear way. Question 1 is followed by Question 2, and so on. There may be branching routines or skip routines, but these often look like a long list of questions or coding categories. In contrast, paper-based coding sheets also more easily allow the layout of the sheet to be designed in a way

Useful observational research

101

that groups similar types of questions together in visual clusters that do not involve scrolling on a screen. An example of clustering variables together in a nonlinear way can be seen in the Chinese Buffet Coding Sheet in Fig. 5.3. A typical electronic or online version of this would have these listed, and it would be more difficult to quickly and accurately fill out, particularly if a number of people were being simultaneously observed. Icons can further be used to break up the page and quickly direct one’s coding.

5.3.3 Observing and coding consumers Pure observational research does not involve interacting with the people who are being observed, so it is exempt from Institutional Review Approval at most institutions. But being an invisible, or at least unobtrusive, observer is not easy. One way to do this is to use the “Mystery Shopper approach” and play the role of a customer or patron as you are doing the observations. In grocery stores, you can have a full cart of groceries with a coding sheet that is shaped to look like a shopping list. At an amusement park, your coding sheet can be designed to fit into a program or amusement park map; at a restaurant, your coding sheet can be designed to look like the other papers you have sitting next to you on the table (Doering & Wansink, 2017). With smart phones and an online coding sheet, this can sometimes be even easier and less conspicuous. The number of consumers that a person observes and codes depends on how long they are observed, how much they move, and on many actions they take (Te et al., 2017). When using paper coding sheets, you can often track up to three people at one time. With electronic coding, accurately observing more than one person at a time can involve switching between screens, which can lead to miscoding mistakes, record write-overs, and lost files. If people are being observed during slow periods of the day, a researcher may be able to code every consumer they see. During busy times, however, it will only be possible to code every nth person. In grocery stores at 5:00 or on weekends, only every 5th person might be tracked; whereas every 2nd one can be tracked on a Monday morning in a small store. One key is to choose the locations and types of locations that will be most relevant for your questions.

5.3.4 Analyzing data Observational data is not clean data (Ross et al., 2015). There are wrong entries and transcription errors. There are sometimes explanatory notes, and unpredicted behaviors to account for (somebody spilling their food or returning a food or meal they just purchased). These explanatory notes can also be used to determine which cases to include in the analysis, and which to note as ones that will not be included (see example at the bottom of Fig. 5.3). Still, some of these decisions are judgment calls on which reasonable people can disagree (Stokes, 2014). To help clean the data, it is useful to examine the response frequencies and means for each variable. Knowing what percentage of people answer a 1 versus a 5 on a 5-point scale can be as useful as knowing how many people check “Don’t know”

102

Context

on another question. This can help sort out egregious data entry errors, and determine what will constitute an outlier (Suen & Ary, 2014). The first analyses also include correlations (or regressions) between the key outcome (or dependent) variables and all of those hypothesized factors you thought would influence this outcome. The goal is to check basic face validity. That is, do people who shop longer in the grocery store seem to buy more food, or people who come in the library early in the morning spend more total time studying than someone who comes in within an hour of closing time. If longer shoppers do not spend more, or if longer studiers do not study more, it could be an indication that there may be a coding problem. Another set of analyses would be the confirmatory ones that compare differences between one group of individuals and another. This could be how slim people act differently than heavy people at buffets, or on whether young boys at a Disney exhibit are more active, or follow the instructions differently compared with young girls. The third sets of analyses are those that are more exploratory (Bernard, Wutich, & Ryan, 2016). Whereas the hypothesized results might seem to be straightforward, these involve more analysis in order to be more certain that significant findings are not just false positive results. As noted earlier with the Chinese buffet study, one way this can be done is to see if an unexpectedly significant finding is also consistent with other results. For example, one study directed at increased female ticket sales at university sporting events unexpectedly showed that females who hold expensive seats at university sports games also appeared to spend less time watching the game than females in less expensive seats or complimentary seats. This and other insights were confirmed through follow-up surveys. That is, the females who attended games most frequently where also more involved in the activities and spectacles surrounding the game rather than the play-by-play action. As a result of these findings, pregame and half-time activities were adjusted to increase involvement in the events surrounding the game (instead of trying to educate them about the rules, players, and history of the game). In the case of theory, these analyses can be useful in knowing whether additional research is merited to better understand behavior. In the case of actual practice, these analyses can be useful, in that they can suggest whether the magnitude is worth seeking a solution.

5.4

Using observational insights to improve the consumer experience

The job is not finished when the insight is discovered. Some observational research projects invest hundreds of hours to discover findings, but then the researchers only spend a couple of hours trying to determine how these insights can be used. The result is a discovery that goes unused because the “So what?” conclusion has not been given even 1% of the thought and effort that went into discovering it. Discoveries are best when they are useful. Unless the interest in the finding is purely academic, unactionable insights are of little practical value. They can be politely dismissed as “interesting,” which contributes to the growing disenchantment, distrust, or disinterest in observational research.

Useful observational research

103

Many researchers expect someone else to translate their findings into action. Some think they are not qualified to do it, and others claim it is not worth their time. As a result, they spend little time in trying to develop useful or specific implications. One perspective is that an observational study that vaguely concludes that what is needed is “better education,” “more awareness,” “more policy,” or “different pricing” may not have merited the effort and cost of the study. A second tendency is to develop three or four suggestions and to stop. This is similar to a student who makes two or three “class participation” remarks during a class, and then stops being engaged for the rest of the class session. Generally, those three or four “top of mind” suggestions are usually not the breakthrough suggestions that would come with more thought. What can be useful is to challenge oneself or one’s team to generate different implications for different aspects of the problem. For example, if the problem is how to encourage people to focus more at work, the conclusions of an observational study could be used to generate implications related to the entry area, the break room, the office arrangement, lunchtime programing, ambiance, travel policies, and so forth. There are some useful general ways to break implications into different categories of action. Three of these systems are the CAN Approach (best for services), the 4-P Marketing Approach (best for products), and the 3-S Framework (best for shopping). The CAN Approach (Convenient, Attractive, and Normal—Wansink, 2015; Wansink, 2013; Wansink & Chandon, 2014) is useful to determine how to change behavior in service contexts, ranging from shopping malls, to school cafeterias, to home kitchens. A second category system is the 4Ps Marketing Framework (Product, Promotion, Place, and Price—e.g., Wansink, 2000) is most easily used with designing, redesigning, or repositioning products. A third method is the 3-S Framework (Structure, Signage, and Service—Wansink, 2017a, 2017b), which is useful for retailing situations both in stores and online, but that can also be used in trying to improve the service in other contexts. For example, if trying to improve the services in a library, a research team could challenge themselves to use their research discoveries to develop recommended changes to the library layout, the way signage is used, the convenience of studying there, or the furniture in the study area. A project that starts out, “Wouldn’t it be interesting to know...” is not likely to be as interesting or useful as one that starts out with, “How do we solve the problem...” Start research with a specific question that will answer a specific problem. In the case of the three examples given in Table 5.2, they all started with a specific problem to solve, and the conclusions helped provide answers to solve that problem. The insights about caramel sauce and the insights about french fries were distributed to a wide number of fast food chains. Within about five years, one chain out of many had reduced the size of the fries in their kids’ meals, and no longer gave out caramel sauce in many locations (Wansink & Hanks, 2014). The insights about the distracting priming that front-facing fiction books had on library patron productivity led at least one large library system to change their display policy. Some of the insights have an even broader range of applications. The insights about what slim buffet diners do differently that heavy diners has changed both what many diners and many buffets do (Wansink, 2014). For example, because slim people tend

104

Context

Table 5.2 Disney’s habit heroes example: from observations to hypotheses to solutions Observation No reason to pay attention. Problem doesn’t seem relevant. Parents seem uneasy they picked a “boring attraction.”

Insights or hypothesis to confirm l

Giving a purpose to the attraction and making it more personalized would increase engagement

Possible solutions Disney could have explored l

l

l

Villains aren’t convincing. Leadbottom doesn’t seem threating and seems unrealistic

No connection made between the food the villains and poor eating, and the food they just ate.

l

Kids focused on activity without understanding the meaning. Anxious for each stage to end

Story line and theme isn’t compelling. Individual activities can be more self-contained and game-like.

l

l

l

People leave without talking or without a sense of direction.

Leaving people with one simple goal or follow-up, and giving them something to discuss or compare upon leaving would encourage follow-up

Find other healthy habits they can put in place right away, such as getting more sleep, drinking more water, wearing sunscreen Use real cast members instead of videotaped ones Create missions Eliminated people as villains (such as overweight people) People fought lots of little germ-like creatures that look like junk food Give people colored wristbands that correspond to the healthy habit they want to improve (eat better, hydrate, sleep more, use sunscreen) Give people a specific follow-up activity

l

Either assign or have them choose a mission to focus on at the end of the attraction

l

Have a follow-up activity on the web or a reward system

to use smaller plates, eat with chopsticks, sit far from the buffet, and scout the buffet before serving themselves, they decided to rearrange—to redesign—their buffets so that all the diners would have to act like slim people. Recall that making the related changes was estimated to help the owner save $1.8 million in food expenses in a year. Turning observation research into useful solutions is difficult. This becomes easier with practice.

Useful observational research

5.5

105

Practice example: Reimagineering a Disney World attraction

Observational coding can be useful in solving problems. Unfortunately, this potential is not always realized when researchers make observations that are “interesting,” but not actionable. But developing this skill to make observations that can be coded and useful, takes practice. This section provides an opportunity to develop this skill by presenting a real problem that Disney faced.

5.5.1 Background: Habit heroes Disney World is unmatched for creating magic and fun for families. In Orlando, Florida they take this to another level at Epcot Center, where they offer interactive exhibits that are both fun and educational. Because of an increasing concern with childhood obesity, they opened a new interactive exhibit called Habit Heroes to help families break the unhealthy habits of overeating, eating the wrong foods, not exercising, and so on. Habit Heroes soft opened at Epcot on February 2, 2012 with a goal of helping families tackle three Habit Villains: (1) too much screen time (which takes place in Control Freak’s electronic lair), too much junk food (which features The Snacker and Sweet Tooth), and too much inactivity (featuring Leadbottom). People who visited the Habit Heroes attraction were divided into groups of around 24. After watching a preshow, they were led into a circular room (Control Freak’s electronic lair) with multiple control panels on the wall, where the objective was to turn off the falling televisions by sprinting from one control panel to the next as a timer counted down. Next, they were led into a room where they used cannons to shoot cartoonish junk foods, by using broccoli, carrots, and blueberries for ammunition. Last, they were led into a room with a dancing aerobic grid, where they did a series of directed aerobic movements in a way that caused Leadbottom to approximately mirror the movements until the villain in him is exorcized. The attraction was not enthusiastically received by families, and became a public relations disaster for Disney (Barrett, 2012). Within three weeks, critics blamed it for reinforcing demeaning stereotypes about fat people, and potentially triggering eating disorders and suicide (McInnis, 2013). One blog claimed the attraction was “horrifying” (Freedhoff, 2013), and the news spread to social networking circles for fat shaming and eating disorders, and it made national news headlines (Gray, 2013). Within two days, Disney closed it for renovation ( Jameson, 2013). One key question needs to be answered: How can the exhibit be changed so that it is enjoyed by families and helps lead to positive behavior changes? Focus groups or interviews would not be useful in answering this. What would be useful is to watch how people interacted with the original exhibit, and use the storytelling method to generate a relevant backstory, and to hypothesize how their behaviors suggest what would be most successful in satisfying their unarticulated needs. To do this, a researcher would focus on one person or family through the entire exhibit. They would

106

Context

then actively revise their story based on the actions that a person takes that do not seem to consistently fit the story they designed. The result of this is a consistent story that can help explain why a person did what they did, how they perceive it, and that suggests what could be done to improve the experience.

5.5.2 Using observations to develop a backstory A video clip of the old Habit Heroes can be found at https://www.youtube.com/watch? v¼SoM38R9xfMs (Disney, 2012, 2013). You can watch this clip, and use the storytelling method to speculate the answers to questions such as the following. l

l

l

l

Where are they from and what are they expecting from is this vacation that makes it different from others? What did they do yesterday, what did they think about it, and how are they feeling? What have they done so far today, what are their expectations for the day, how are they viewing each other, what are they looking forward to? What’s going to happen when they get back to their home in terms of what they do those first couple days to adjust, what do they tell their friends, how is their relationship between each other going to be different.

These are impossible questions to answer accurately, and that does not matter. What matters is that doing this will heighten your sense of observation, your understanding of their actions, and the solutions you develop. For example, here’s one backstory for one such family such as those in this video: Here’s a family that takes a large family vacation like this once every four or five years, but really goes all out to make it a memory that lasts. They are staying at the Disney hotels with a three-day package. They spent the first day at the Theme Park riding all of the classic rides, such as Space Mountain, and seeing the classic attractions such as Pirates of the Caribbean and the Haunted Mansion. They are still exhausted and sunburned from yesterday’s adventure, and decided to tone things down by going to Epcot and doing something more tame and “educational” to help the parents justify how much they were spending on the trip. They got a late start this morning and ate the buffet at the hotel. So far today they have visited a couple of lessthan-thrilling exhibits, and have stopped for snacks, and twice for drinks because today is hot, humid, and sunny, just like yesterday. Their most recent stop was at a kiosk where they could get build-your-own slushy drinks with long, narrow 18-inch souvenir glasses, one of which is still being carried by the son.

This observer-created story is not 100% true, but it’s also probably not 100% wrong. By building and revisiting it, however, you would have developed a much more engaging and active way to observe why the children in this family might be behaving the way they do. Your next goal as a researcher is to develop observations, and to then form hypotheses from these observations that can be confirmed (using observation and coding, for example).

Useful observational research

107

5.5.3 How Disney changed habit heroes Based on hypotheses from observational research and on follow-up confirmation, new heroes were added, including Director Jin, and three agents (Agent Dynamo, representing physical activity; Agent Quench, representing hydration; and Agent Fuel, representing healthy eating and nutrition). The overweight, unattractive villains were replaced with less anthropomorphic counterparts, including Blocker-Bots (opponents that hide healthy food choices), Sappers (boulders that prevent kids from being active), and Scorchers (flames that can dehydrate unsuspecting kids). The floor plan was similar and had three activities: (1) a dance grid that encourages “power moves” against the villains, (2) a cannon range that encourages guests to defend a city using their “power moves” against similar opponents, and (3) a cylindrical room and multiple control panels, in which participants defend the world targeting the same opponents as before. In response to the (hypothesized) insights in Table 5.2, this new attraction increased its cast-member interaction. Also, after the exhibition walk-through was completed, each guest received a mission that addressed the health habit they had chosen (eating, hydration, physical activity, sleep, and so on) and a way to log onto websites to track their progress and get rewards after they returned home. Since the retooling, the attraction was cited by the Sun Sentinel as a “...kinder, more sensitive attraction” and it remained open for three full years after its retooling.

5.6

Conclusion

Powerful, vivid, insightful, compelling, and useful. All of these can explain good observational research. But often they don’t. The best observational research starts with a problem it hopes to generate insights on and to solve. Without a problem focus, observational research can often appear as… random observations. Having a problem-focus will also focus one toward the relevant insights. These insights will not be obvious, or there would be no real need to do the research. The most useful way we have found to generate insights over the years is through the storytelling method that was briefly described. Great observational research can capture imaginations, sear an image into one’s awareness, and provide the basis for solving an important problem. It is important to start with that problem in mind.

References Abrams, B. (2000). The observational research handbook: Understanding how consumers live with your product (p. 5). Chicago, IL: American Marketing Association. Barrett, S. (2012). Habit heroes – bad habits (and a hidden Mickey). AllEars.Net.http://land. allears.net/blogs/stevebarrett/2012/02/habit_heroes_bad_habits_and_a_1.html. Accessed 18 February 2012.

108

Context

Bernard, H. R. (2017). Research methods in anthropology: Qualitative and quantitative approaches. Rowman & Littlefield. Bernard, H. R., Wutich, A., & Ryan, G. W. (2016). Analyzing qualitative data: Systematic approaches. SAGE Publications. Boschini, M., Falasconi, L., Giordano, C., & Alboni, F. (2018). Food waste in school canteens: A reference methodology for large-scale studies. Journal of Cleaner Production, 182, 1024–1032. Brannen, J. (Ed.), (2017). Mixing methods: Qualitative and quantitative research. Routledge. Chorney, J. M. L., McMurtry, C. M., Chambers, C. T., & Bakeman, R. (2014). Developing and modifying behavioral coding schemes in pediatric psychology: A practical guide. Journal of Pediatric Psychology, 40(1), 154–164. Crosby, R. A., Salazar, L. F., & DiClemente, R. J. (2015). Conducting observational research. In Research methods in health promotion (p. 259). Disney (2012). https://www.youtube.com/watch?v¼SoM38R9xfMs. Disney (2013). https://www.youtube.com/watch?v¼E7WR3QOjwSo. Doering, T., & Wansink, B. (2017). The Waiter’s weight: Does a server’s BMI relate to how much food diners order? Environment and Behavior, 49(2), 215–223. Fortini-Campbell, L. (1991). The consumer insight workbook: how consumer insights can inspire better marketing and advertising. In The Copy Workshop, Chicago, IL. Fowler, A., & Montagnes, B. P. (2015). College football, elections, and false-positive results in observational research. Proceedings of the National Academy of Sciences, 112(45), 13800–13804. Freedhoff, Y. (2013). Epcot’s habit heroes reopens. Did they remove the shaming? Weighty Matters. 22 January 2013. http://www.weightymatters.ca/2013/01/epcots-habit-heroesreopens-did-they.html. Friard, O., & Gamba, M. (2016). BORIS: A free, versatile open-source event-logging software for video/audio coding and live observations. Methods in Ecology and Evolution, 7(11), 1325–1330. Gerber, A. S., Green, D. P., & Kaplan, E. H. (2014). The illusion of learning from observational research. In Field experiments and their critics: Essays on the uses and abuses of experimentation in the social sciences (pp. 9–32). Getts, K. (2016). Measuring plate waste: Validity and inter-rater reliability of the quarter-waste method. PhD dissertation, Seattle, WA: University of Washington. Getts, K. M., Quinn, E. L., Johnson, D. B., & Otten, J. J. (2017). Validity and interrater reliability of the visual quarter-waste method for assessing food waste in middle school and high school cafeteria settings. Journal of the Academy of Nutrition and Dietetics, 117(11), 1816–1821. Girard, J. M., & Cohn, J. F. (2016). A primer on observational measurement. Assessment, 23(4), 404–413. Goodman, S. N., Schneeweiss, S., & Baiocchi, M. (2017). Using design thinking to differentiate useful from misleading evidence in observational research. JAMA, 317(7), 705–707. Gray, E. (2013). Disney’s anti-obesity ‘Habit Heroes’ exhibit at Epcot causes controversy. Huffington Post, Retrieved February 11. Grimes, D. A., & Schulz, K. F. (2002). Bias and causal associations in observational research. The Lancet, 359(9302), 248–252. Hanks, A. S., Just, D. R., & Wansink, B. (2013). Reliability and accuracy of real-time visualization techniques for measuring school cafeteria tray waste: Validating the quarter-waste method. Journal of the Academy of Nutrition and Dietetics, 114, 470–474.

Useful observational research

109

Jameson, M. (2013). Epcot’s habit heroes reopens as a kinder, more sensitive attraction. Sun Sentinel, 18 January 2013, http://articles.orlandosentinel.com/2013-01-18/the-daily-dis ney/os-epcot-habit-heroes-reopens-20130117_1_power-and-callie-stenics-attraction-fatbodies. McInnis, F. (2013). Disney World blunders with anti-obesity exhibit. Shine On, Yahoo Canada, https://ca.style.yahoo.com/blogs/shine-on/disney-world-blunders-anti-obesity-exhibit194902369.html. Merriam, S. B., & Tisdell, E. J. (2015). Qualitative research: A guide to design and implementation. John Wiley & Sons. Morgan, S. J., Pullon, S. R. H., Macdonald, L. M., McKinlay, E. M., & Gray, B. V. (2017). Case study observational research: A framework for conducting case study research where observation data are the focus. Qualitative Health Research, 27(7), 1060–1068. Ross, M. E., Kreider, A. R., Huang, Y.-S., Matone, M., Rubin, D. M., & Localio, A. R. (2015). Propensity score methods for analyzing observational data like randomized experiments: Challenges and solutions for rare outcomes and exposures. American Journal of Epidemiology, 181(12), 989–995. Stokes, S. C. (2014). A defense of observational research. In Field experiments and their critics: Essays on the uses and abuses of experimentation in the social sciences (pp. 33–57). Suen, H. K., & Ary, D. (2014). Analyzing quantitative behavioral observation data. Psychology Press. Te, G., Manfred, B. P., Eisinga, R., Nieuwenhuis, R., Schmidt-Catran, A., & Konig, R. (2017). When size matters: Advantages of weighted effect coding in observational studies. International Journal of Public Health, 62(1), 163–167. Wansink, B. (2000). New techniques to generate key marketing insights. Marketing Research, Summer, 28–36. Wansink, B. (2013). Convenient, attractive, and normative: The CAN approach to making children slim by design. Childhood Obesity, 9(4), 277–278. Wansink, B. (2014). Slim by design – Mindless eating solutions for everyday life. New York, NY: William Morrow. Wansink, B. (2015). Change their choice! Changing behavior using the CAN approach and activism research. Psychology & Marketing, 32(5), 486–500. Wansink, B. (2017a). Corrigendum: Change their choice! Changing behavior using the CAN approach and activism research. Psychology & Marketing, 32(5), 486–500. Wansink, B. (2017b). Healthy profits: An interdisciplinary retail framework to increase the sales of healthy food. Journal of Retailing, 93(March), 65–78. Wansink, B., & Chandon, P. (2014). Slim by design: Redirecting the accidental drivers of mindless overeating. Journal of Consumer Psychology, 24, 413–431. Wansink, B., Hanks, A. S., & Stein, K. (2014). Observations from how Kid’s meals and happy meals are eaten: Fast food opportunities to help children eat healthier. Journal of Nutrition Education and Behavior, 46(4S), 161. Wansink, B., & Hanks, A. S. (2014). Calorie reductions and within-meal calorie compensation in Children’s meal combos. Obesity, 22(3), 630–632. Wansink, B., Just, D. J., & Payne, C. R. (2009). Mindless eating and healthy heuristics for the irrational. American Economic Review, 99(2), 165–169. Wansink, B., & Payne, C. R. (2008). Eating behavior and obesity at Chinese buffets. Obesity, 16(8), 1957–1960.

110

Context

Further reading Duhigg, C. (2008). Warning: Habits may be good for you. 13 July. The New York Times. Ejima, K., Li, P., Smith, D. L., Nagy, T. R., Kadish, I., Groen, T., et al. (2016). Observational research rigour alone does not justify causal inference. European Journal of Clinical Investigation, 46(12), 985–993. Jameson, M. (2012). Disney closes new habit heroes exhibit after criticism for stigmatizing fat kids. Orlando Sentinel, 29 February 2012, http://www.orlandosentinel.com/travel/ attractions/the-daily-disney/os-disney-habit-heroes-closes-20120229-story.html. McKenzie, T. L., & Van Der Mars, H. (2015). Top 10 research questions related to assessing physical activity and its contexts using systematic observation. Research Quarterly for Exercise and Sport, 86(1), 13–29. O’Keefe, G. S. (2012). Disney’s habit heroes: A review from a pediatrician mom who experienced it. Huffington Post, 3 May, https://www.huffingtonpost.com/gwenn-okeeffe/disneyhabit-heroes_b_1314401.html. Omerod, P. (2006). Why most things fail: Evolution, extinction, and economics. New York: Pantheon. Payne, C. R., & Wansink, B. (2011). Quantitative approaches to consumer field research. Journal of Marketing Theory and Practice, 19(4), 377–389. Rushkoff, D. (2005). Get back in the box: Innovation from the inside out. New York: HarperCollins. Tal, A., Wansink, B., & Miquel-Kergoat, S. (2014). Groceries and gum: Chewing gum influences grocery store shopping. Journal of Nutrition Education and Behavior, 46(4S), 176–177. Wansink, B. (2006). Mindless eating – Why we eat more than we think. New York: Bantam-Dell. Wikipedia Contributors (2018). Habit Heroes. Wikipedia, The Free Encyclopedia. 11 Mar 2018, https://en.wikipedia.org/wiki/Habit_Heroes#cite_note-8.

Situational appropriateness in food-oriented consumer research: Concept, method, and applications

6

Davide Giacalone SDU Innovation & Design Engineering, Department of Technology and Innovation, Faculty of Engineering, University of Southern Denmark, Odense, Denmark

6.1

Appropriateness as a basic context construct

In the food and beverage industry, hedonic responses (expressed as degree of liking or preference for a set of test products) have traditionally been an important product performance indicator. With both everyday experience and empirical research telling us that the sensory acceptability of foods strongly predicts consumption, there is little doubt that this is a meaningful dimension to study in central location tests (CLT). Indeed, sensory and consumer scientists have long been interested in understanding the relationship between specific sensory properties and food acceptability, to identify those that maximize liking within a specific product category, as well as for specific consumer segments within that category. Unfortunately, strong sensory and consumer performance does not necessarily predict product success: for example, it does not guarantee that a product will be purchased often or that it is appropriate for the usage situation that the product developers had in mind ( Jaeger & Porcherot, 2017; Rosas-Nexticapa, Angulo, & O’Mahony, 2005). As Marshall (1993) reminds us, the final decision to buy or consume a particular food depends as much on the anticipated usage context as it does on the intrinsic qualities of the product (or the consumer). Highly liked and even preferred products may not be chosen simply because they are inappropriate. For example, a consumer may highly appreciate a very complex wine when fine dining, but the same individual would choose a less expensive one for a routine meal or a picnic. Therefore, understanding consumer food choices requires us to think about how consumers use products, as well as how they select them. The concept of appropriateness focuses on the usage contexts of food products (as well as consumer products generally). Appropriateness is defined as “the quality of being especially suitable or fitting” (Merriam-Webster’s Dictionary, 2018, online at merriam-webster.com). In consumer research, situational appropriateness is generally defined as the perceived fit between a product and a target usage context (Ratneshwar & Shocker, 1991; Schutz, 1988). The term “perceived” reflects the fact that situational appropriateness is a cognitive phenomenon that requires a subjective evaluation on the part of the consumer. However, unlike evaluations of preferences Context. https://doi.org/10.1016/B978-0-12-814495-4.00006-4 Copyright © 2019 Elsevier Inc. All rights reserved.

112

Context

and acceptability, where the emphasis is squarely on subjective experiences, the concept of appropriateness includes a plethora of rules that are culturally determined and learned through experience and socialization (e.g., what ingredients go together, what is a typical portion size, what foods one usually eats for breakfast, etc.). Indeed, the term “appropriateness” in and of itself suggests a normative aspect of how well a food fits the situation in which it is supposed to be consumed. How relevant is perceived situational appropriateness for explaining consumers’ food-related behavior? As a starting point, one can certainly expect that appropriateness interacts with aspects related to the product (both intrinsic—such as sensory properties or caloric content, and extrinsic—such as a packaging and brand) and to the consumer (personal preferences, physical states such as hunger, thirst, etc., and mental processes, such as specific goals and motives) to determine consumers’ choices. However, while the latter two aspects are always in focus in sensory and consumer research, situational influences have received comparatively little attention. If we look at the general marketing and consumer psychology literature, the relevance of the appropriateness construct for explaining consumer behavior is supported by a substantial body of research showing that consumers differentiate products on the basis of the anticipated usage situations (Lai, 1991). The underlying theoretical premise for this line of work is that consumers’ perceptions of products rarely occur in isolation, but, rather, relative to some frame of reference. The usage situation of a product is an ecological factor that can help define consumers’ goals, and thus orient their choices toward “situationally appropriate” solutions (Giacalone & Jaeger, 2016; Ratneshwar & Shocker, 1991). In consumer psychology, situational effects on consumer choices are typically explained on the basis of the compatibility principle (Tversky, Sattah, & Slovic, 1988; Tversky & Simonson, 1993), according to which when individuals make choices they tend to select the options that are superior on a salient dimension. In this sense, the anticipated usage situation can orient consumers’ attention to product attributes relevant to fulfilling goals associated with it. To illustrate the point, consider the example of a consumer who wants to purchase a bottle of wine out of n alternatives. In theory, the consumer could evaluate all relevant attributes (e.g., price, origin, vintage, grape variety, alcohol by volume, etc.), assign a subjective value to each, and choose the product with the best overall value. This hypothetical decision-making process is essentially the familiar theory of rational choice based on individual preferences. Let us now assume that the same person is now choosing between the same set of alternative products, keeping in mind, say, that they have just decided to go on a diet. In such a situation, the compatibility principle suggests that they will be more likely to focus on the dimension that is made salient by the anticipated usage situation, which then provides a cognitively efficient “metric” for comparing the wine alternatives. In this instance, this consumer may equate alcohol content with caloric content, and choose the wine with the lowest alcohol by volume as the most appropriate given their current health goal. This second model of decision-making process takes its point of departure into bounded rationality (the notion that consumers’ information processing capabilities are limited and flawed) and assumes that consumer choices are often context-dependent (Tversky & Simonson, 1993; Warlop & Ratneshwar, 1993).

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 113

The important implication of this viewpoint is that consumers in many cases choose products to fulfill the goals associated with a particular consumption situation, rather than solely on the basis of individual preferences and product characteristics (Belk, 1975, Ratneshwar & Shocker, 1991; Giacalone, 2018). Informed by this understanding, a number of methodological approaches based on situational segmentation, that is, on the identification of perceived product benefits across different usage situations, have been developed. Looking at the history of situational research in sensory and consumer science, we can distinguish between two macro approaches. The first one involves the use of natural or physically manipulated environments to vary contextual elements of interest, and then to record consumer responses in a target context (e.g., Bell, Meiselman, Pierson, & Reeve, 1994; Di Monaco, Giacalone, Pepe, Masi, & Cavella, 2014; Meiselman, Johnson, Reeve, & Crouch, 2000; Sester et al., 2013). Most studies adopting this line of work have focused on food acceptability or intake as primary outcome measures, but some have also provided indications that situational appropriateness is relevant to consumer choices. For example, a well cited paper by Bell and colleagues demonstrated that changing the d
ecor and the names of items of a restaurant menu increased the frequency of choice for items that patrons perceived as more congruent with the environment, such as pasta with an Italian ethnic theme (Bell et al., 1994). While natural or naturalistic approaches can provide a high degree of ecological validity, the practical planning and implementation of these studies may be burdensome and/or present pitfalls due to factors unforeseen by the researchers (Cardello & Meiselman, 2018; Jaeger & Porcherot, 2017). This has limited the number of such studies, though efforts to measure consumer responses in natural environments have been called upon by several authors ( Jaeger & Porcherot, 2017; Meiselman, 2013), and may be expected to increase siginificantly in the near future, thanks to the increased availability of immersive virtual reality (VR) technologies ( Jaeger, Hort, Porcherot, et al., 2017; see also Hehn et al., and Hartmann & Siegrist, this volume). The second major stream of research has focused on situational effects in central location tests (CLT), where the majority of sensory and consumer studies takes place. This has consisted in the modification of standard CLT protocols to include a usage context of interest, for example, by presenting consumers with a scenario as a means to evoke a focal situation to consumers, and ask them to imagine this situation while carrying out the product evaluations (Hein, Hamid, Jaeger, & Delahunty, 2010, 2012; Jaeger & Porcherot, 2017). Other CLT-based approaches have involved explicit evaluation of situational appropriateness of products, using a wide variety of methods, ranging from qualitative methods, such as focus groups (Elzerman, van Boekel, & Luning, 2013), personal interviews (Hartwell, Shepherd, Edwards, & Johns, 2016), and word associations (de Andrade, de Aguiar Sobral, Ares, & Deliza, 2016), to quantitative methods such as Free Choice Profiling ( Jack & Piggott, 1991; Piggott, Sheen, & Apostolidou, 1990) and repertory grid interviews ( Jack, Piggott, & Patterson, 1994; McEwan & Thomson, 1988; Jaeger, Rossiter, & Lau, 2005; Scriven, Gains, Green, & Thomson, 1989; Raats & Shepherd, 1991/1992).

114

Context

Among CLT-based methods, one of the earliest and best-known is the “item-byuse” (IBU) method, developed and popularized by Schutz (1988, 1994). In this approach, which is sometimes referred to as “substitution in use” in the marketing literature (e.g., Ratneshwar & Shocker, 1991), consumers are presented with a product set and a list of possible usage contexts, and are asked to indicate how well a product fits each of them (Schutz, 1988, 1994). The IBU approach enables a comprehensive analysis of the appropriateness of products with reference to a specific usage context, thus explicitly considering products as means to reach an end defined by a particular usage situation. Situational appropriateness ratings may have important behavioral correlates. In particular, recent research has shown that appropriateness evaluations (1) are predictive of consumer food choices, willingness to use, and reported intake (Elzerman, Hoek, van Boekel, & Luning, 2015; Giacalone & Jaeger, 2016, 2017; Sosa, Martı´nez, Arruiz, Hough, & Mucci, 2005) and (2) underpin other productrelated variables, such as emotional responses and perceived product uniqueness ( Jaeger, Cardello, Chheang, et al., 2017; Jaeger, Cardello, Jin, et al., 2017; Piqueras-Fiszman & Jaeger, 2014a, 2014b, 2015a, 2015b). Taken collectively, these studies indicated that situational appropriateness should be considered as an important product performance criterion in CLT tests. A thorough discussion of the IBU approach constitutes the main focus for the rest of the chapter.

6.2

The “Item-by-use” (IBU) approach to measuring product appropriateness

6.2.1 Origin and historical use in sensory and consumer science The introduction of the IBU approach in sensory and consumer research is owed to the pioneering work of Howard G. Schutz, who first became exposed to the situational appropriateness construct while working with anthropologist Volney Stefflre, and later adapted a basic anthropological technique (Stefflre, 1971) for application in consumer product testing in the 1970s, during his tenure at the University of California at Davis. Schutz’s reasoning in developing the IBU appropriateness method was that the common food classification systems based on nominal categories (e.g., dairy, bakery, etc.) or nutritional content did not necessarily reflect the way consumers would classify different food items (Schutz, Rucker, & Russell, 1975), and may therefore be of limited value for developing and marketing new products. As originally proposed, the IBU method consists of presenting a consumer with a list of possible consumption situations, and asking him/her to rate how well a product fit each of them, on a 7-point scale ranging from “‘Never appropriate” to “Always appropriate.” In principle, the IBU method can be applied to any product category, including non-food products, such as personal and household care products, and even durable products, as long as they have more than one possible usage context. Table 6.1 provides a comprehensive (yet possibly non-exhaustive) list of journal publications featuring the IBU approach, starting from the early work of Schutz

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 115

Table 6.1 List of research publications employing the IBU approach, sorted chronologically Reference

Year

Schutz & Ortega

1974

Schutz, Fridgen, & Damrell Schutz, Rucker, & Russell Schutz & Rucker

1975

1975

Baird & Schutz

1976

Schutz & Phillips Bruhn & Schutz

1976

Resurreccion Martens et al. White, Resurreccion, & Lillard Raats & Shepherd Nantachai, Petty, & Scriven L€ahteenm€aki & Tuorila Cardello & Schutz

1986 1987 1988

1975

1986

Product category

Subjects (N)

Contexts (N)

Stimuli (N)

Type of stimulus

Type of scale

Wine and other alcoholic drinks Rice and related products Misc. food and beverages Misc. food and beverages Misc. food and beverages Textiles

52

48

56

Names

7-pt scale

200

48

52

Names

7-pt scale

200

48

56

Names

7-pt scale

60

10

10

Names

2-,3-,6-,7pt scales

135

25

20

Names

2-pt scale

50

48

46

Names

7-pt scale

51

45

46

Names

7-pt scale

67 135 203

10 34 12

43 30 10

Names Names Names

7-pt scale 7-pt scale 7-pt scale

Dairy products Vegetables Vegetables Meat products

1991

Milk

40

49

7

Images

7-pt scale

1991

Meat products Ice-cream

30/30

21/19

18/18

Names

40

10

3

Misc. food and beverages (29 studies) Drinks

27 to 38

10

1 to 4

Samples (Blind) Samples (Blind)

100 mm line scale 9-pt scale

243

18

8

115

10

124

1995 1996

L€ahteenm€aki & Tuorila Cardello et al.

1997

Jaeger

2000

Sosa et al.

2005

Schutz, Cardello & Winterhalter Mejlholm & Martens

2005

Misc. food and beverages Apple products Seasoning sauces Textiles

2006

Beer

2000

7-pt scale

22

Samples (Blind) Names

7-pt scale 7-pt scale

15

4

Images

7-pt scale

240

16

5

Names

7-pt scale

100

30

16

Names

7-pt scale

38

23

10

Samples (Blind)

7-pt scale

Continued

116

Context

Table 6.1 Continued Elzermann et al.

2011

Hersleth et al.

2011

Bach et al.

2013

Elzermann et al.

2015

PiquerasFiszman & Jaeger PiquerasFiszman & Jaeger

2014a

2014b

Meat substitutes in dishes Ham

93

4

5

Dishes

100 mm line scale

81

2

8

9-pt scale

Jerusalem artichoke Meat substitutes in dishes (online survey) Apple/ Brownie

49

2

5

251

6

5

Samples (Informed) Samples (Blind) Images

76/81

3

1/1

Images

9-pt scale

115/302/ 188

3/3/3

3/3/1

Images

9-pt scale

1336 (96 to 417)

4 to 16

1

Names

7-pt scale

76/97/ 93/145 49

15/15/ 15/9 3

9/9/9/9

Images

CATA

5

Samples (Blind) Names or images

5-pt scale

5-pt scale 100 mm line scale

PiquerasFiszman & Jaeger

2015

Giacalone et al.

2015

Brownie/ Kiwifruit/ Brownie or crisps Misc. food and beverages (7 studies) Beer

Bach et al.

2015

Beetroot

Giacalone & Jaeger

2016

246/112/ 192/302

16/16/ 15/15

19/12/ 13/12

Geertsen, Allesen-Holm, & Giacalone Stolzenbach et al. Cardello et al.

2016

200

8

7

Samples (Blind)

CATA

2016

Fruit/Wine/ chocolate/ kiwifruit Seabuckthorn beverages Apple juice

196

6

4

7-pt scale

2016

Beer

203

15

8

Jaeger et al.

2017a

Beer

128

14

9

Jaeger et al.

2017b

Dark chocolate/ white chocolate

139/128

12/13

8/8

Samples (Blind) Samples (Blind) Samples (Blind) Samples (Blind)

CATA

CATA CATA CATA

and collaborators in the 1970s, up to the time of writing. Most of the publications involve food and beverage products, but applications in non-food products have also been proposed (Schutz, Cardello, & Winterhalter, 2005; Schutz & Phillips, 1976). Early applications (1970s–1990s) of the IBU approach involved evaluations of food names, and are characterized by a very large number of both stimuli and usage

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 117

contexts, sometimes resulting in >2500 food use combinations, and a completion time of more than two hours (Schutz & Ortega, 1974). These studies were conducted as self-administered questionnaires, and respondents reportedly spread the task over different time intervals. A key takeaway of these early studies on appropriateness is that they consistently show, through the application of factor analysis, that consumers grouped foods according to common usage contexts such as “high-calorie foods,” “specialty meal items,” “common meal items,” “inexpensive filling foods,” “healthy foods,” and so forth (Marshall, 1993; Schutz, 1988, 1994), but that these factors were not synonymous with any objectively defined food category. Schutz (1988) has reviewed most of these early studies utilizing the IBU method, and argued that appropriateness is an important cognitive-contextual aspect for both theoretical and practical purposes. The IBU approach gained momentum in the mid-90s, following some influential papers that extended its application range significantly. Raats and Shepherd (1991) were the first to use the IBU approach to differentiate products within the same category, in a study comparing different types of milk (fresh vs. UHT and with different fat content levels), demonstrating the usefulness of this technique with a product range close to that of a typical CLT. This study was also the first to use product images and accompanying information as test stimuli, as opposed to food names. Shortly afterward, studies by L€ahteenm€aki and Tuorila (1995, 1997) and by Cardello and Schutz (1996) extended the IBU method to measuring responses to actual products in blind tests, demonstrating that (1) consumers can differentiate both within and between products on the basis of sensory differences, and, (2) that products that do not differ in terms of acceptability can nonetheless be widely different in terms of appropriateness for specific usage contexts. Fig. 6.1 shows an example of this incidence, both when considering (a) widely different products, as well as (b) products in the same category. Cardello and Schutz (1996) also reported that the presence of appropriateness ratings does not significantly alter hedonic ratings, suggesting that collecting appropriateness data in conjunction with standard sensory and consumer tests can be done without hampering the validity of the test results. Taken overall, these studies provided a strong rationale for using the IBU method as an adjunct to hedonic testing of food products in CLT. Cardello and Schutz (1996) called for the routine collection of IBU data in CLT studies, in order to evaluate not only whether the test products have high acceptability, but also high appropriateness for the usage context(s) they are intended for. Let us note here that this call is as relevant as ever some 20 years later, and that the IBU method fits quite naturally with common new product development practices, where products are generally developed with an end goal in mind, for example, to fit a specific consumer segment (e.g., women, elderly, etc.), or to provide a specific benefit (e.g., health, convenience, etc.). Later studies have indicated that the situational appropriateness construct may have very important behavioral correlates for predicting consumer choice and consumption. Cardello, Schutz, Snow, and Lesher (2000) found significant associations between IBU appropriateness ratings and expected liking for food (with correlations

118

Context 7

Mean appropriateness

6 5 Stuffed cabbage rolls 4

Mean acceptance 7.1

Baked ham

6.9

Zesty spaghetti

6.7

3 2

* Differ

Cheer

* Little

Nutrit

Snack

Tired

Alone

* Breakf

(A)

* Coldday

Hungry

1

Situation

7

Mean appropriateness

6 5

Roast veal

4

Mean acceptance 7.1

Baked ham

7.0

Pork adobo

6.7

3 2

(B)

* Differ

Little

Cheer

Nutrit

Snack

Tired

Alone

Coldday

* Breakf

Hungry

1

Situation

Fig. 6.1 Products having equal acceptability can nonetheless differ greatly in appropriateness for use, as exemplified by this plot showing mean liking and mean appropriateness ratings for three widely different products (A), and for three products in the same category (B) (*, P < .05; “cheer”, when I want cheering up; “little”, when I have little time to eat; “differ”, when I want something different). Reproduced with permission from Schutz, H. G. (1995). Eating situations, food appropriateness, and consumption. In Marriott, B. M. (Ed.), Not eating enough: Overcoming underconsumption of military operational rations (pp. 341–359), Washington, DC: National Academies Press.

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 119

7

Mean appropriateness

6

5

4

3

2 y = 0.857x + 0.037 r 2 = 0.957 1 1

2

3 4 5 6 7 Mean expected liking/dislikng

8

9

Fig. 6.2 Correlation between mean (N ¼ 115) appropriateness ratings (7-pt scale) and expected liking (9-pt scale) for 220 food/uses combination, showing a high degree of associations between these two measures. Reprinted with permission from Elsevier from Cardello, A. V., Schutz, H., Snow, C., Lesher, L. (2000). Predictors of food acceptance, consumption and satisfaction in specific eating situations. Food Quality and Preference, 11, 201–216.

ranging between 0.53 and 0.83, depending on context), and a very high degree of linear dependence when considering mean ratings (Fig. 6.2). In a paper that has probably flown under the radar, Sosa and colleagues (Sosa et al., 2005) studied the correlation between appropriateness and self-reported frequency of intake in a series of usage contexts, reporting a near perfect correlation (all correlations
0.96) between these two variables (Fig. 6.3). Albeit based on a single product category (seasoning sauces), these results are indicative of appropriateness being a strong predictor of food consumption, especially when compared with hedonic ratings, which, at best, have been found to account for only about 50% of the variance in consumption (Cardello et al., 2000; Lau, Hanada, Kaminskyj, & Krondl, 1979; Tuorila, Hyv€ onen, & Vainio, 1994). In recent years, the IBU method has been adopted for a wide range of applications, such as relating appropriateness for use to sensory attributes (e.g., Hersleth, Berggren, Westad, & Martens, 2005; Mejlholm & Martens, 2006), product packaging and information (e.g., Hersleth, Lengard, Verbeke, Guerrero, & Næs, 2011; Jaeger, 2000), and has been increasingly used in conjunction with hedonic and other perceptual product

120

Context 100

80 Mayonnaise Ketchup

60

Mustard Golf sauce

40 Mostanesa

20

0 1

2

3

4

5

6

7

Fig. 6.3 Correlation between mean (N ¼ 240) appropriateness ratings (7-pt scale) and selfreported frequency of consumption (100 mm scale from “never used” to “always used”) for five seasoning sauces (Sosa et al., 2005). The scatterplot only shows the points for one of the sauces (mayonnaise), but according to the authors all products had a similar dispersion and all correlation coefficients were
0.96. Reprinted with permission from Elsevier from Sosa et al. (2005).

responses, such as expectations, emotions, and well-being (Cardello et al., 2016; Jaeger, Cardello, Chheang, et al., 2017; Jaeger, Cardello, Jin, et al., 2017; Jaeger, Hort, Porcherot, et al., 2017). The value of IBU data in new product development has also been highlighted (e.g., Geertsen, Allesen-Holm, & Giacalone, 2016; Stolzenbach, Bredie, Christensen, & Byrne, 2016). An interesting application niche has focused on appropriateness in the context of culinary preparations, that is, by using meals and dishes as usage contexts, to drive gastronomic food development (Bach, Kidmose, Thybo, & Edelenbos, 2013; Bach, Mikkelsen, Kidmose, & Edelenbos, 2015), or to evaluate the possibility of ingredient substitutions (Elzerman et al., 2015; Elzerman, Hoek, van Boekel, & Luning, 2011). Summarizing, the IBU approach is a flexible instrument for evaluating appropriateness in CLT tests. As a natural supplement to hedonic testing, it can help ensure that the test products not only have high acceptability, but also a high fit with the usage context(s) they are intended for. Although the total number of publications listed (N ¼ 34) indicates a certain popularity of this method, the bulk of the work on appropriateness comes from few authors. Additionally, one may argue that this number is very low if compared with the number of studies that have focused on liking or preference in that same period. There may be different reasons for this. Low awareness may be one: for example, in my experience, many colleagues and reviewers (wrongly) assume that the IBU method is only applicable to differentiate between widely different products, but that it is not useful for products within the same category. This is likely a reminiscence of the early appropriateness papers; however, there are several examples of successful applications of discriminations within a range of product

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 121

differences that is very typical for CLTs (see Section 6.2.2.2). Perhaps there is a need for developing standard tools that can facilitate the adoption of the IBU method, in a similar way as methodological development in emotional measurement has paved the way for research in this area (Meiselman, 2015). Or maybe Schutz was simply way ahead of his time, and the field is still catching up. If that is the case, in light of the growing interest for context in consumer research, we can expect an increase in the utilization of situational appropriateness in sensory and consumer research. With this in mind, the reminder of the chapter covers the main methodological aspects to consider when using the IBU method, and discusses recent advances and venues for future research in the area.

6.2.2 Methodological considerations for appropriateness evaluations using the “Item-by-use” (IBU) approach 6.2.2.1 Ballot format Table 6.1 shows that the IBU method is quite flexible with respect to key methodological choices. In early applications, the basic questionnaire format has consisted of a matrix crossing foods and uses, such as the one shown in Fig. 6.4. The task for the respondent is to fill in the matrix with values corresponding to perceived appropriateness in each use situation (typically on a 7-pt scale). With respect to measurement, the 7-pt category scale originally proposed by Schutz has been the most common option, but other types of scales (e.g., 5-pt, 9-pt, line scales), as well as the check-all-that-apply (CATA) format, have also been used to collect appropriateness responses. The question of whether scale length and format affect the data has received little attention in the literature, with the exception of an old paper by Schutz and Rucker (1975) comparing data from 2-, 3-, 6-, and 7-point rating scales. This research concluded that the number of scale points did not affect the cognitive structures derived from appropriateness responses, indicating a high degree of robustness for this type of measurement (Schutz & Martens, 2001; Schutz & Rucker, 1975). Interestingly, this was the case also for a 2-point scale version (1 ¼ Appropriate, 0 ¼ Inappropriate) whose binary nature is equivalent to what one would today call a CATA question in a forced choice format, some 35 years before this method became popular in sensory and consumer science. The same binary scale was also employed in a later study comparing cognitive food structures in different ethnic groups (Baird & Schutz, 1976). Generally, shorter scales (and especially the CATA format) are easier to process for consumers ( Jaeger et al., 2013; Schutz and Rucker, 1975), so the choice of scale should ultimately take into consideration practical aspects such as the total testing time, the number of stimuli to evaluate, the co-presence of an acceptability and/or other questions in the same ballot, etc. Fig. 6.5 shows an example of a CATA ballot using for evaluation of visual stimuli. The food-use matrix is more commonly associated with evaluation of food names in a self-administered questionnaires. When visual and taste stimuli are used—which

For Sunday dinner

When you are really hungry

Late at night

When riding in a car

With coffee

In the summer

When you are sick

When eating out

When you are depressed

Context

When watching TV

122

Jello Potato chips Chicken Orange juice Celery Soup Pizza Cereal Pie Grape

Fig. 6.4 Example of a IBU matrix for appropriateness evaluations of food names across different usage situations. The respondents are instructed to fill each cell with a value ranging from 1 (Never appropriate) to 7 (Always appropriate). Modified from Schutz, H. G., & Rucker, M. H. (1975). A comparison of variable configurations across scale lengths: An empirical study. Educational and Psychological Assessment, 35, 319–324.

is the common case when appropriateness evaluations are conducted in the context of standard CLT tests—stimuli are normally evaluated monadically in separate ballots. It is customary to ask about overall acceptance first, and then have them rate the product for use appropriateness after, in order to minimize the risk of potential biases.

6.2.2.2 Selection of test stimuli As already mentioned, the IBU is a flexible approach that can be used with a variety of stimuli, including names, actual products, and images (Table 6.1). Depending on the aims of the study, stimuli can either include items from different product categories, or be restricted to a specific product category. Including a broad number of stimuli spanning across product categories is typically the case when the goal is to understand how consumers classify foods, as in earlier applications of the IBU approach (Schutz, 1988). The number of items for this type of study has ranged from 10 to as many as 56 (Table 6.1). There are no strict rules for

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 123 Please look at the picture on the right, and then answer the questions below.

DB Export Dry (DB Breweries, NZ)

Which of the following situation(s) do you think this beer would be appropriate for? Please tick all that apply.

When I want something different

At a pub

To drink alone

Anytime

When I want something refreshing

As an alternative to wine

With dinner

As a drink for women

For a special occasion

At a sport event

At a casual dining restaurant

To impress someone

At home

At parties

When I want to relax

I don’t know

How familiar are you with the beer shown in the picture?

1 Not at all familiar

2

3

4

5 Extremely familiar

How often do you drink the beer shown in the picture?

Never

Rarely (1–2 times a year, or less frequently)

Sometimes (3–12 times a year)

Often (1–3 times a month)

Regularly (once a week or more)

How familiar are you with the DB EXPORT brand in general?

1 Not at all familiar

2

3

4

5 Extremely familiar

Fig. 6.5 Example of a ballot for evaluation of IBU appropriateness in a CATA format using visual stimuli from a study on beer (Giacalone et al., 2015).

the choice of items, other than they should, of course, be representative of the foods available in that particular market, and cover different sensory intensity levels and nutritional groups (for examples of such lists, see Schutz, 1988, 1994; Schutz & Martens, 2001). An important thing to consider is the level of specificity to provide

124

Context

to the consumers, which depends entirely on the research aims. For example, whether to provide a generic or a branded food name (e.g., “cola” or “Coca-Cola”), or whether to present food items separately or in combination (e.g., eggs vs. eggs and bacon), with or without additional qualifiers (e.g., “eggs” vs “fried eggs”), etc. These are very important choices, as they will almost certainly affect perceived appropriateness. In other situations, stimuli may be restricted to a specific product category, where physical products or pictorial representations are used to elicit appropriateness evaluation from consumers. This would often be the case when appropriateness data is collected in CLT tests. There is still somewhat of a misconception to this day that the IBU appropriateness context has relevance only for differentiating between different food products, but that it cannot differentiate between products within the same category. However, this is not the case. For example, L€ahteenm€aki and Tuorila showed that appropriateness ratings obtained from blind taste tests differentiated between brands of vanilla ice-creams (1995), and between blueberry-raspberry juices varying in sweetener content (1997). Several other examples can be found in the literature (see e.g., Geertsen et al., 2016; Hersleth et al., 2005; Mejlholm & Martens, 2006; Stolzenbach et al., 2016), showing that appropriateness as a concept is relevant to differentiating product variants within the same category. Naturally, the magnitude of difference in perceived appropriateness is related to the heterogeneity in the stimuli set, and will generally increase manifolds as soon as extrinsic factors are included (Giacalone et al., 2015; Giacalone & Jaeger, 2016). The number and choice of stimuli for studies restricted to one specific product category is generally a lot lower, with the majority of studies including 6–8 stimuli (Table 6.1). This is because the majority of these studies include other types of evaluations in the same test ballot, such as acceptability, sensory attributes, and productrelated emotions (e.g., Cardello et al., 2016; Cardello & Schutz, 1996), so the number of stimuli to evaluate needs to be kept low, as to not interfere with the normal progress of the test. In this case, selecting an appropriate number of stimuli is generally based on the usual considerations for consumer tests (ballot length, the type of stimuli and the type of evaluation requested, avoiding fatigue, etc.). Regarding the type of stimuli, blind samples are a common choice in CLT tests (Table 6.1), which makes the most sense when one wants to link appropriateness to sensory variation and acceptability. The use of pictorial images as stimuli is also increasingly seen in the literature, and with good reason: because vision is the most important sensory modality at the point of purchase, product appearance is often a very important cue for assessing the perceived usage appropriateness of products. Accordingly, this stimulus format has been effectively employed in studies investigating the influence of extrinsic product aspects, such as the packaging, in categories that depend strongly on visual inspection (Giacalone & Jaeger, 2016; Jaeger, Hedderley, & MacFie, 2001).

6.2.2.3 Usage contexts Selection and number of usage contexts, as for the stimuli, will depend on the study objectives. The generation of usage contexts is a crucial aspect of the IBU method, although as noted by Jaeger and Porcherot (2017), the process is seldom described in much detail in the literature.

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 125

As a general rule, the usage contexts should cover all those that are representative and frequent for that product category, as well as any additional target usage that may be important to testers. A theoretical and practical challenge is obviously in defining and operationalizing usage contexts. Thus, a good way to develop a list of usage contexts is to take as one’s point of departure the existing classifications of situational variables (Belk, 1975; Bisogni et al., 2007; Blake, Bisogni, Sobal, Devine, & Jastran, 2007; Meiselman, 2006, Meiselman, 2008; Wansink, 2004), and start developing relevant options for each class. For example, in his landmark paper on situational influences on consumer behavior, Belk (1975) proposed a broad taxonomy of situational variables that include (1) physical and (2) social surroundings, (3) temporal perspectives, (4) task definition, and (5) antecedent states. Table 6.2 shows an example of usage classes from Belk’s taxonomy with examples relevant for the product category “beer” (from Giacalone et al., 2015). Several lists of usage contexts for specific product categories can be found in the papers listed in Table 6.1. Although existing literature is a good starting point, attribute generation could also be done with a bottom-up approach, for example, by using a small group of consumers to generate a list of usage situations. In general, some degree of consumer involvement in developing the list of items is highly recommended to ensure that the items included in the ballot are understood, and that no attributes that may be important to consumers are left out. An important decision concerns the number of usage contexts to include. The key thing to keep in mind is that the number of food-usage context combinations will grow geometrically, so that a 10 by 10 matrix, such as the one in Fig. 6.4, will entail 100 evaluations, whereas a 50 by 50 matrix similar to those used in early applications will require 2500 evaluations on the part of each consumer. Obviously, the test situation (home, CLT, online), the stimuli type, and the possible presence of other tasks need to be considered, as it is crucial not to overburden the cognitive and affective system of Table 6.2 Belk’s (1975) taxonomy of usage situations Physical surroundings On a camping trip Watching TV at home At a pub

Task definition To celebrate an achievement As a gift for someone For a special occasion

l

l

l

l

l

l

Social surroundings To serve to guests To impress someone For women

Antecedent states As a thirst quencher When I want to relax When I want something different

l

l

l

l

l

l

Temporal perspective At a BBQ in the summer l

l

l

For lunch As an alternative to wine for dinner

Examples of contextual attributes within each class are given for beer (from Giacalone et al., 2015).

126

Context

the consumers. Thorough pilot testing will serve as the best indication of the expected elapsed time for completion, and therefore what is reasonable to include the ballot. One final crucial consideration is the way the usage contexts are to be presented to the consumers. In the vast majority of studies, IBU contexts are presented as very short phrases (see examples in Table 6.2 and Fig. 6.4). While this way is efficient and easy to implement, the downside is that the corresponding mental image may not be very compelling and vivid in the mind of the respondents ( Jaeger & Porcherot, 2017). Additionally, because consumers are asked to imagine a usage context in response to a written depiction, there is a risk that specific aspects of the context may vary quite substantially across consumers (Hein et al., 2010, 2012), and/or not match the intention of the researcher. This is in fact not necessarily a negative feature. For example, K€oster and Mojet (2015) suggest that subjectively interpreting usage contexts may often be more relevant for the consumers than evaluating a more objective definition of context, because it enables them to draw on their own previous personal experiences. However, from the researcher’s perspective, a more objective definition of the usage context is often desirable, as it allows for a clearer interpretation of the results. If that is the case, more detailed descriptions and pictorial representations may be used to better convey the target usage context to consumers (Elzerman et al., 2015; Giacalone et al., 2015). For example, Fig. 6.6 shows two usage contexts from a study on situational appropriateness of beer (Giacalone et al., 2015) in the form of specific eating situations with an accompanying caption. Formal comparisons of different stimuli formats in the context of IBU studies have not been carried out; however, visual stimuli are assumed to enhance the ecological validity of CLT-based studies of contexts (Cardello & Meiselman, 2018; Jaeger & Porcherot, 2017; K€ oster, 2003), and have the added benefit of reducing the ambiguity regarding what consumers have imagined during their appropriateness evaluations.

6.2.2.4 Respondents Regarding how many consumers to include, Table 6.1 shows a considerable variation in the literature, with sample sizes ranging from as few as 30 consumers (Nantachai et al., 1991), to >400 (Piqueras-Fiszman & Jaeger, 2015a, 2015b). The N of respondents seem to have progressively increased in recent years, following a general trend in sensory and consumer toward larger and more representative sample sizes, as well as more attention to diversity and consumer segmentation (Meiselman, 2013). It has been noted that IBU data present a low amount of inter-individual variability relative to other types of consumer evaluations. For example, Schutz (1994) reports that it is possible to achieve stable results, that is, reliable mean ratings and component structures, with as few as 25 consumers (with the 7-pt scale, but in my experience, that it is the case even with CATA). This is remarkable if compared with for example, hedonic data, where 100+ consumers are typically considered as the minimum sample size (Hough et al., 2006). This discrepancy can be attributed to the different nature of affective and cognitive evaluations; while hedonic data are affected by a multitude of factors that provide a large amount of individual variation, appropriateness data are

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 127

Fig. 6.6 Example of visual depictions of usage contexts in a IBU appropriateness task (in this case with a CATA format). Reprinted with permission from Elsevier from Giacalone, D., Frøst, M. B., Bredie, W. L., Pineau, B., Hunter, D. C., Paisley, A. G., Beresford, M. K., & Jaeger, S. R. (2015). Situational appropriateness of beer is influenced by product familiarity. Food Quality and Preference, 39, 16–27.

more likely to reflect the culturally agreed place that a food occupies within a certain consumer group. For example, it seems safe to assume that asking consumers how much they like beer would result in a higher variability in response than if the same consumers were asked to rate how appropriate they thought beer was “for breakfast.” Accordingly, Giacalone and colleagues report that IBU data have very high

128

Context

reproducibility also when replicating studies within different consumer groups, confirming the primarily coenotropic nature of appropriateness evaluations (Giacalone et al., 2015).

6.2.2.5 Analysis of appropriateness data The analysis of IBU appropriateness data involves both univariate and multivariate techniques. The first step is generally to compute descriptive statistics for each product for each individual usage context, and to test for statistical differences between the products by an appropriate statistical test (e.g., ANOVA on product means when rating scales are used, Cochran’s Q test when the CATA format is used). Univariate analyses such as these can be presented visually, for example, using bar or line charts, or numerically in a table where products are ranked by appropriateness supplemented by post-hoc testing comparing the means or proportions presented. These simple results are already of great practical value in a product development context, as they give a ranking of products enabling a quick evaluation of how each product performs in each usage situation. Cardello and Schutz (1996) also proposed to supplement product means analyses with an overall distance metric (D) computed for each product pair. This can be done using the standardized euclidean distance formula: n  X

Dij ¼ k¼1

1

h

2 2 i  Xik  Xjk =XMax n

where Dij is the distance between product i and product j, Xik and Xjk are the mean appropriateness ratings for usage context k, XMax is the largest possible appropriateness value (corresponding to the highest point on the scale or to 1 when proportions are used), and n is the number of usage contexts. The advantage of using this scaled distance measure is that it provides a global assessment of absolute similarity between products, not just a similarity of pattern (as, e.g., a correlation coefficient would). As observed by Schutz (1994), this may be particularly relevant to determine the viability of substituting one product for another; for example, when introducing a novel food from one culture to another. In practice, one should always supplement the overall analysis with an analysis at the individual usage context. Especially if the list of usage contexts is long, Dij may often tend toward high value that may overestimate the similarity between two items. Multivariate data analyses of IBU data are very useful to derive perceptual maps showing how consumers cognitively organize different food items and usages. Principal Component Analysis (PCA) on a table containing mean ratings across usage contexts has been the most common approach for rating data (e.g., Schutz, 1988, 1994; Schutz & Martens, 2001); whereas for the CATA format, the same is typically accomplished using correspondence analyses on a contingency table crossing products and usage contexts (e.g., Giacalone et al., 2015). Regardless of the analytical approach used, it is interesting to notice that multivariate analyses of appropriateness data

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 129

generally result in efficient representations and interpretable component structures. The proportion of variance explained in the first two components is generally high (70%–90%)—especially compared with what one usually gets from hedonic data—suggesting that patterns of product-usage associations are robust and widely held by consumers. In addition to PCA and CA, multiblock approaches such as Partial Least Squares Regression (PLSR) can be used to relate appropriateness data to other types of data. For example, Hersleth et al. (2005) used a PLSR model to predict appropriateness of bread from sensory data, and Mejlholm and Martens (2006) used a three-block PLSR approach to predict consumer preferences for beer from appropriateness data, sensory profiles, and consumer background data simultaneously.

6.2.3 Other approaches to evaluating product appropriateness As previously mentioned, the IBU approach is not the only CLT-based approach to evaluating situational appropriateness. For example, the Repertory Grid Method (RGM) (Kelly, 1955) has been used by several authors to investigate contextual aspects of food choice and acceptance ( Jack et al., 1994; Jaeger et al., 2005; McEwan & Thomson, 1988; Monteleone, Raats, & Mela, 1997; Raats & Shepherd, 1991/1992; Scriven et al., 1989). Briefly, the RGM is an interview in which a consumer is presented with groups of three products (triads), two of which are arbitrarily associated with each other and dissociated from the third. The consumer is then asked to describe how (s) he thinks that the two associated products are similar, and how they are different from the third. The specific question may be tweaked to meet the specific intention of the researcher. For example, if situational appropriateness is in focus, consumers may be asked to describe in what usage contexts they would use the two associated products, but not the third (Gains, 1994). Once the consumer has exhausted all possible constructs, the task is repeated for the other two combinations in the triad, as well for any other triads, resulting in a large number of usage contexts that discriminate between the product under test. Then, consumers typically provide ratings for each product for each of the constructs they had previously individually generated. The data are then submitted to Generalized Procrustes Analysis (Gower, 1975) to derive a common perceptual map of the products from the individual ratings (or “repertory grids”). In a sense, RGM can be considered a more structured version of the Free Choice Profile ( Jack & Piggott, 1991), which also has been used for eliciting situational appropriateness constructs from consumers (e.g., Piggott et al., 1990). A lengthy comparison of all possible approaches is beyond the scope of this chapter. However, one point that is important to recognize is that approaches such as RGM and FCP are ideographic in nature, that is, they let each consumer develop their own individual set of usage contexts, whereas the IBU approach is nomothetic, where one set of usage contexts is evaluated by all consumers (Schutz, 1994). The IBU approach thus assumes that all usage contexts are relevant to all consumers, which may not necessarily be the case. To what degree this difference is important in practice is unclear, as studies that have compared the two approaches (Nantachai et al., 1991; Raats & Shepherd, 1991/1992) have generally concluded that they provide similar results.

130

Context

Based on practical consideration, the IBU approach is certainly faster, more flexible, and overall better suited for large-scale CLT tests in a product development context, where the aim is to generalize to a population of interest. By contrast, RGM is an excellent method when one wants to really elucidate the cognitive structures consumers hold within a certain product category, but can quickly become impractical as soon as the number of consumers (and of products) increases. It might be best to think in terms of complementarities. Since the RGM is a process of contrast and similarities, it seems especially suited as a method for generating usage contexts to be further studied with the IBU approach, thus ensuring that usage contexts are as relevant as possible to the consumer population. For example, Nantachai et al. (1991) used repertory grid interviews to develop usage contexts for testing in a larger IBU study. Geertsen et al. (2016) used a version of projective mapping to the same effect. Another methodological approach that has been proposed to measuring appropriateness is conjoint analysis, where usage contexts can be specified as design factors in a conjoint design, together with other factors of interest, such as sensory, packaging, information, and so forth (Almli et al., 2011; Jaeger, 2000; Jager & Rose, 2008). The significant advantage of the conjoint approach is that it allows one to estimate interaction effects between usage contexts and other experimental factors, which is of course of great interest. The downside is that, compared with the IBU approach, a much lower number of usage contexts can be included in a single study. The application context should be considered in this case. For example, in a product development context, the IBU approach may be more suited to early stage product development, when the focus is often interested in testing different product formulations. Conjoint methods appear advantageous when closer to market launch, where a few target usage contexts have been identified, and where blending intrinsic and extrinsic product elements is most important.

6.3

Current directions in appropriateness research

This section reviews recent advances in appropriateness research, as well as open venues for future research. Attention is given to three areas that have substantive interest for sensory and consumer researchers: (1) behavioral and attitudinal correlates of situational appropriateness, (2) consumer segmentation based on appropriateness data, and (3) the potential of immersive technology in appropriateness research.

6.3.1 Attitudinal and behavioral correlates of appropriateness 6.3.1.1 Choice, consumption, willingness to pay Having already established that appropriateness is not identical to liking (Cardello et al., 2000; Cardello & Schutz, 1996), one important research question is the relevance of appropriateness for other behavioral and attitudinal outcomes of interest. Can appropriateness be used to predict consumer choice, frequency of consumption, and willingness to pay?

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 131

Regarding frequency of purchase, the already mentioned study by Sosa and collaborators (Sosa et al., 2005) suggested a strong link between IBU ratings of situational appropriateness and self-reported frequency of consumption of different seasoning sauces. This is in agreement with an earlier study by Shepherd, Schutz, and Sparks (1993), in which appropriateness evaluations of a 50  50 item-by-use matrix was found to account for 44% of the variance in frequency of use. It is interesting to note that in the same dataset, the authors report that preference (operationalized as the attribute “when I want something I really like” in the IBU matrix) could explain 12% of variance in frequency of use. Although both studies are based only on self-reported estimates of consumption frequency, these data are very indicative that appropriateness evaluations are predictive of repeat purchase and frequency of consumption. Studies on different product categories, and possibly using actual consumption frequency (e.g., using food recall diaries) would be very useful to confirm and extend these results. A few studies have investigated the relationships between appropriateness and willingness to try and purchase. Elzerman and colleagues (Elzerman et al., 2015) reported that willingness to try meat substitutes were highly related to the appropriateness of the usage context (in that study, meal combinations). Another similarly motivated study was published by Lai (1991), who investigated consumers’ willingness to adopt different variants of a canned beverage (Wulong tea) across different usage contexts. This study also found that consumers were more willingly to adopt products that could be seen as more situationally appropriate. What about choice itself? As laid out in the introduction, the general marketing and consumer behavior literature supports the notion that consumers’ anticipated usage is a strong determinant of consumer choice (Belk, 1975; Lai, 1991; Jager & Rose, 2008). Indeed, even if one took the view that consumers’ choices are primarily hedonicoriented, the near perfect correlation observed by appropriateness ratings and expected liking (Fig. 6.2) should point to a strong relationship between appropriateness and choice. To this day, surprisingly little attention has been devoted to empirically verifying and quantifying this relationship. These effects were recently studied by Giacalone and Jaeger (2017) in a series of 14 studies spanning a wide range of product categories and usage contexts. In all studies, participants evaluated a set of stimuli chosen to represent fixed levels of IBU appropriateness (low, medium, high) in a target context, and completed a choice task using a 7-point choice likelihood scale, or a discrete choice using the best-worst scaling method. Linear regression results from all studies consistently indicated that the IBU level significantly predicted choice response. The results were robust with respect to variation in product category and experimental protocol. Effect sizes varied substantially between studies, with appropriateness level explaining from a minimum of 3% to >60% of variance in consumers’ choice responses, and this variation was linearly related to the range in appropriateness in the product sets. Overall, the results strongly support the notion that IBU appropriateness is a significant predictor of food choice, and hence a meaningful product performance measure.

132

Context

6.3.1.2 Appropriateness and emotions Some authors have emphasized the important effect that situational appropriateness may have for the measurement of food-related emotions in consumer research, a topic that has enjoyed increased attention in recent years (K€oster & Mojet, 2015; Meiselman, 2015). Such effects were elegantly elucidated in a series of papers by Piqueras-Fiszman and Jaeger (2014a, 2014b, 2014c, 2015a, 2015b): using a variety of methods and products. These authors demonstrated that the usage context affected consumer emotional responses toward food products, and specifically that the frequency of use of positive (vs. negative) emotion terms strongly depended on whether a product was considered appropriate (vs. inappropriate) for the target usage context. This research has obviously important implications for emotion measurement. Although sometimes one may want to separate the product effect from the context effect, neglecting the situational component of emotional responses hinders the ecological validity of the results. Therefore, a future research direction with both theoretical and practical interest would then be to study how individual aspects of the usage context (e.g., time of the day, presence of others, etc.) affect perceived appropriateness and emotional responses. The afore-mentioned papers (in particular PiquerasFiszman & Jaeger, 2015a) already showcase examples of how the impact of variations in usage context dimensions on appropriateness can be quantified.

6.3.2 Situation-based consumer segmentation As observed earlier, appropriateness data tend to have high reproducibility and a lower level of variability than liking or other attitudinal data. However, inter-individual variation occurs, and there might be situations in which one may want to study it. Extending the reasoning that appropriateness of food is socially determined, culture or ethnic groups can be expected to be a major source of variation. This is supported by extant data (e.g., Baird & Schutz, 1976; Jaeger, 2000; Nantachai et al., 1991). For example, Jaeger (2000) demonstrated differential patterns of situational appropriateness for apples between New Zealand and Samoan consumers related to the cultural orientation (individualism vs. collectivism) between these two countries. Within a specific culture, Rucker and Schutz (1982) have presented examples of consumer segmentation on the basis of a factor analysis of appropriateness data, concluding that this data can be used to segment consumers in a meaningful way. It is not known where these segments correspond to any predefined data in terms of demographic, psychographic, or behavioral variables. In the already mentioned paper by Sosa and collaborators (Sosa et al., 2005) focusing on the relationships between degree of appropriateness and (self-reported) frequency of consumption, the authors found appropriateness to be highly correlated to consumption when considering averaged data, but also a significant spread in individual responses. However, segmenting consumers by demographics did not improve prediction, leading the authors to conclude that the effect of appropriateness on consumption frequencies must be mediated

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 133

by non-demographic factors. Evidently, understanding the specific composition of an appropriateness-based consumer segment would be interesting in the context of both product development and marketing, as well in other areas in which understanding product-usage patterns for specific consumer groups is of interest (e.g., in a public health context). Nevertheless, this aspect has not yet been explicitly addressed by extant research. In general, constructs relating to general attitudes toward food (e.g., food neophobia), and to previous experience with a product category (e.g., product knowledge, involvement, etc.) are more likely to affect appropriateness evaluation. For example, Giacalone and Jaeger (2016) suggest that consumers who are knowledgeable about a product may process product elements more confidently, and therefore infer appropriate usage contexts more easily compared with less experienced consumers (e.g., a wine connoisseur may quickly envision food pairing of a wine based on grape variety or vintage, whereas a novice would not). The next section reviews recent advances in this area that may explain why this is the case.

6.3.2.1 Familiarity as a moderator of appropriateness Previous studies have suggested that consumers’ familiarity with a product may be an important factor in shaping appropriateness of use evaluations (e.g., Jaeger et al., 2005; Tuorila et al., 1994). The influence of product familiarity in explaining consumers’ evaluations of appropriateness of use has been recently investigated in two papers by Giacalone, Jaeger, and collaborators (Giacalone et al., 2015; Giacalone & Jaeger, 2016), which demonstrated that product familiarity is positively related to product versatility, defined as the total number of situations a product is deemed appropriate for. In other words, as product familiarity increases, products are perceived as appropriate for a larger number of uses. Conversely, consumers experienced difficulty in identifying appropriate uses for unfamiliar products. This finding may explain the relationship between appropriateness and frequency of purchase (Sosa et al., 2005; Shepherd et al., 1993). Generally, consumers have little incentive to sample unfamiliar foods and beverages, and therefore familiar products may be routinely chosen because they are more readily available, and more likely to enter the consideration set because of, for example, previous satisfactory consumption experiences. Thus, it makes sense that product familiarity is a strong predictor of repeat purchase, at least when consumers make generic food provisioning decisions (“stocking up” the household pantry with no particular end usage in mind) (Giacalone et al., 2015). These results also suggest that consumers may find it challenging to envisage how unfamiliar food products can be incorporated into their existing dietary practices, resulting in food-related usage patterns that are resistant to change. This is, of course, a major barrier to consumer adoption of new products. In the future, it would be interesting to elucidate the mechanisms by which unfamiliar products are perceived as less appropriate. One possibility is that familiarity breeds comfort, making existing usage patterns and routines more desirable.

134

Context

Additionally, consumers might worry that an unfamiliar product might not taste as expected, and be a potential waste of money. The cognitive effort associated with evaluating unfamiliar products or product features may itself be a source of negative bias against new products. For food and beverage specifically, risk perception may be a factor. For example, consumers may think that familiar products are healthier and safer. Additionally, the potential social embarrassment coming from poor choices (e.g., serving a product that is contaminated, or simply does not taste as expected) might also lead consumers to rate unfamiliar products as generally less appropriate. Uncovering these motives should be a focus in future appropriateness research, and it would be especially needed in understanding how this negative bias toward novel products can be overridden. Giacalone and Jaeger (2016) also suggest devoting greater attention to retail-level strategies that can suggest product usage, such as free trials (useful if the low appropriateness is rooted in risk aversion), and a goal-based shelf display (i.e., where products are organized around intended uses or benefits), which may be a good strategy to improve appropriateness in a usage context not currently associated with that product (e.g., reduced alcohol wine in the health food section).

6.3.3 Immersive technologies in appropriateness research Immersive technologies, where consumers experience and interact with a computer generated virtual environment, are currently a major trend in sensory and consumer science. The possibility of conducting product evaluations in a fully immersive virtual environment is clearly a hotbed for context research, and is seen as a major opportunity to enhance the external validity of CLT tests ( Jaeger & Porcherot, 2017). The application of VR is dealt with extensively in this book, and the reader is referred to Chapters 16 and 23 for in-depth discussions. A relevant question to ask here is what does the increasing availability of immersive technologies mean for appropriateness evaluations? This is difficult to answer sensibly, because at the time of writing, the VR technology is still in its infancy, and we probably need more experience to fully understand its value in sensory and consumer science. In general, VR technologies should offer interesting possibilities for addressing two of the perceived drawbacks of appropriateness research related to the elicitation of usage contexts in response to written descriptions (see Section 6.2.2.3): the low level of immersiveness, and the potential misalignment of these mental images between consumers (as well as between consumers and researchers). In principle, the use of VR technology can remedy both. That aside, in this author’s opinion, IBU and VR do rather different things, and are likely to have different uses. The VR approach seems most useful when the goal is go in depth into one (or a few) target contexts that are of interest to the researcher or product developer, or when that context is actually the only one relevant to that product category, for example, a car cabin for evaluation of a car air freshener. Here, the VR approach can be very useful to provide results (e.g., liking, willingness to pay) with a higher external validity than a context-neutral CLT-test. From this perspective, the VR approach is more in line with the tradition of conducting situational research in

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 135

natural or physically manipulated environments (e.g., Bell et al., 1994; Di Monaco et al., 2014; Petit & Sieffermann, 2007; Sester et al., 2013). While the IBU approach cannot replace real contexts to the same degree as (in principle) VR can, it affords the possibility of evaluating a larger number of products and usage contexts in the same session. This makes it especially useful for evaluating the performance of different products across different usage contexts (do consumers understand how to use the products? Which benefits do each product deliver, which products may be replaced or repositioned, etc.), and their perceived versatility. This information is, of course, most useful for products that have a variety of applicable contexts (typically the case for food and beverages), and when one is interested in understanding the cognitive structures pertaining to product usage within a specific product category. From a practical perspective, the IBU method overall seems much better suited for the context of current CLT tests, which, for the most part, involve actual samples, because when it comes to eating while immersed in a VR environment, reaching a high enough level of realism and usability still presents a significant challenge.

6.4

Conclusions

This chapter has discussed situational appropriateness in the context of sensory and consumer research, with particular focus on the IBU method proposed by Schutz (1988, 1994) as an adjunct to CLT evaluations. Appropriateness, intended as perceived fit with a specific usage situation, is a basic product-related construct distinct from other affective and attitudinal measures. The end-goal of CLT tests should be to ensure that the test products not only score high on absolute performance indicators (acceptability, willingness to pay, etc.), but also high appropriateness for the usage situation envisioned by the product developer. As recently emphasized by Jaeger and Porcherot (2017), product understanding cannot be complete without appropriateness for use evaluations. While the IBU approach cannot fully replace real contexts in terms of external validity, it does provide an easy and flexible way to gather contextual information in a CLT setting. Like any method, the IBU approach has its own pros and cons, which should be weighted based on the specific goals against alternative approaches (including CLT, non-CLT, and immersive technologies) to investigate situational influences in product-related research. Recent research has indicated that IBU appropriateness ratings (1) may be a strong predictor of consumer food choices, willingness to use, and reported intake, and (2) underpin other product-related variables, such as emotional responses and perceived product uniqueness. Taken collectively, these studies indicate that appropriateness should be considered as an important product performance criterion in CLT testing. Given the complexity of consumer behavior, the ability to predict product success must ultimately encompass a combination of product performance indicators. Situational appropriateness is, however, an important part of this mix that should not be overlooked.

136

Context

References Almli, V. L., Næs, T., Enderli, G., Sulmont-Ross
e, C., Issanchou, S., & Hersleth, M. (2011). Consumers’ acceptance of innovations in traditional cheese. A comparative study in France and Norway. Appetite, 57, 110–120. Bach, V., Kidmose, U., Thybo, A. K., & Edelenbos, M. (2013). Sensory quality and appropriateness of raw and boiled Jerusalem artichoke tubers (Helianthus tuberosus L.). Journal of the Science of Food and Agriculture, 93, 1211–1218. Bach, V., Mikkelsen, L., Kidmose, U., & Edelenbos, M. (2015). Culinary preparation of beetroot (Beta vulgaris L.): The impact on sensory quality and appropriateness. Journal of the Science of Food and Agriculture, 95, 1852–1859. Baird, P. C., & Schutz, H. G. (1976). The marketing concept applied to “selling” good nutrition. Journal of Nutrition Education, 8, 13–17. Belk, R. W. (1975). Situational variables and consumer behavior. Journal of Consumer Research, 2, 157–164. Bell, R., Meiselman, H. L., Pierson, B. J., & Reeve, W. G. (1994). Effects of adding an Italian theme to a restaurant on the perceived ethnicity, acceptability, and selection of foods. Appetite, 22, 11–24. Bisogni, C. A., Winter Falk, L., Madore, E., Blake, C. E., Jastran, M., Sobal, J., et al. (2007). Dimensions of everyday eating and drinking episodes. Appetite, 48, 218–231. Blake, C. E., Bisogni, C. A., Sobal, J., Devine, C. M., & Jastran, M. (2007). Classifying foods in contexts: How adults categorize foods for different eating settings. Appetite, 49, 500–510. Bruhn, C. M., & Schutz, H. G. (1986). Consumer perceptions of dairy and related-use foods. Food Technology, 40, 79–85. Cardello, A. V., & Meiselman, H. L. (2018). Contextual influences on consumer responses to food products. In: G. Ares & P. Varela (Eds.), vol. 2 Methods in consumer research (pp. 3– 54): Elsevier. Cardello, A. V., Pineau, B., Paisley, A. G., Roigard, C. M., Chheang, S. L., & Guo, L. F. (2016). Cognitive and emotional differentiators for beer: An exploratory study focusing on “uniqueness” Food Quality and Preference, 54, 23–38. Cardello, A. V., Schutz, H., Snow, C., & Lesher, L. (2000). Predictors of food acceptance, consumption and satisfaction in specific eating situations. Food Quality and Preference, 11, 201–216. Cardello, A. V., & Schutz, H. G. (1996). Food appropriateness measures as an adjunct to consumer preference/acceptability evaluation. Food Quality and Preference, 7, 239–249. de Andrade, J. C., de Aguiar Sobral, L., Ares, G., & Deliza, R. (2016). Understanding consumers’ perception of lamb meat using free word association. Meat Science, 117, 68–74. Di Monaco, R., Giacalone, D., Pepe, O., Masi, P., & Cavella, S. (2014). Effect of social interaction and meal accompaniments on acceptability of sourdough prepared croissants: An exploratory study. Food Research International, 66, 325–331. Elzerman, J. E., Hoek, A. C., van Boekel, M. A., & Luning, P. A. (2011). Consumer acceptance and appropriateness of meat substitutes in a meal context. Food Quality and Preference, 22, 233–240. Elzerman, J. E., Hoek, A. C., van Boekel, M. J., & Luning, P. A. (2015). Appropriateness, acceptance and sensory preferences based on visual information: A web-based survey on meat substitutes in a meal context. Food Quality and Preference, 42, 56–65. Elzerman, J. E., van Boekel, M. A., & Luning, P. A. (2013). Exploring meat substitutes: Consumer experiences and contextual factors. British Food Journal, 115, 700–710.

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 137

Gains, N. (1994). The repertory grid approach. In H. J. H. MacFie & D. M. H. Thomson (Eds.), Measurement of food preferences (pp. 51–76). Boston, MA: Springer. Geertsen, J., Allesen-Holm, B. H., & Giacalone, D. (2016). Consumer-led development of novel sea-buckthorn based beverages. Journal of Sensory Studies, 31, 245–255. Giacalone, D. (2018). Product performance optimization. In: G. Ares & P. Varela (Eds.), vol. 1 Methods in consumer research (pp. 159–185): Elsevier. Giacalone, D., Frøst, M. B., Bredie, W. L., Pineau, B., Hunter, D. C., Paisley, A. G., et al. (2015). Situational appropriateness of beer is influenced by product familiarity. Food Quality and Preference, 39, 16–27. Giacalone, D., & Jaeger, S. R. (2016). Better the devil you know? How product familiarity affects usage versatility of foods and beverages. Journal of Economic Psychology, 55, 120–138. Giacalone, D., & Jaeger, S. R. (2017). Situational appropriateness (item-by-use): an overlooked predictor of food choice. 12th Pangborn Sensory Science Symposium, 20–24 August 2017, Providence, RI, USA. Gower, J. C. (1975). Generalized procrustes analysis. Psychometrika, 40, 33–51. Hartwell, H. J., Shepherd, P. A., Edwards, J. S., & Johns, N. (2016). What do patients value in the hospital meal experience? Appetite, 96, 293–298. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2010). Application of a written scenario to evoke a consumption context in a laboratory setting: Effects on hedonic ratings. Food Quality and Preference, 21, 410–416. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2012). Effects of evoked consumption contexts on hedonic ratings: A case study with two fruit beverages. Food Quality and Preference, 26, 35–44. Hersleth, M., Berggren, R., Westad, F., & Martens, M. (2005). Perception of bread: A comparison of consumers and trained assessors. Journal of Food Science, 70, 95–101. Hersleth, M., Lengard, V., Verbeke, W., Guerrero, L., & Næs, T. (2011). Consumers’ acceptance of innovations in dry-cured ham: Impact of reduced salt content, prolonged aging time and new origin. Food Quality and Preference, 22, 31–41. Hough, G., Wakeling, I., Mucci, A., Chambers, E., IV, Gallardo, I. M., & Alves, L. R. (2006). Number of consumers necessary for sensory acceptability tests. Food Quality and Preference, 17, 522–526. Jack, F. R., & Piggott, J. R. (1991). Free choice profiling in consumer research. Food Quality and Preference, 3, 129–134. Jack, F. R., Piggott, J. R., & Patterson, A. (1994). Use and appropriateness in cheese choice, and an evaluation of attributes influencing appropriateness. Food Quality and Preference, 5, 281–290. Jaeger, S. R. (2000). Uncovering cultural differences in choice behaviour between Samoan and New Zealand consumers: A case study with apples. Food Quality and Preference, 11, 405–417. Jaeger, S. R., Cardello, A. V., Chheang, S. L., Beresford, M. K., Hedderley, D. I., & Pineau, B. (2017a). Holistic and consumer-centric assessment of beer: A multi-measurement approach. Food Research International, 99, 287–297. Jaeger, S. R., Cardello, A. V., Jin, D., Hunter, D. C., Roigard, C. M., & Hedderley, D. I. (2017b). Product uniqueness: Further exploration and application of a consumer-based methodology. Food Quality and Preference, 60, 59–71. Jaeger, S. R., Giacalone, D., Roigard, C. M., Pineau, B., Vidal, L., Gim
enez, A., et al. (2013). Investigation of bias of hedonic scores when co-eliciting product attribute information using CATA questions. Food Quality and Preference, 30, 242–249.

138

Context

Jaeger, S. R., Hedderley, D., & MacFie, H. J. (2001). Methodological issues in conjoint analysis: A case study. European Journal of Marketing, 35, 1217–1239. Jaeger, S. R., Hort, J., Porcherot, C., Ares, G., Pecore, S., & MacFie, H. J. H. (2017). Future directions in sensory and consumer science: Four perspectives and audience voting. Food Quality and Preference, 56, 301–309. Jaeger, S. R., & Porcherot, C. (2017). Consumption context in consumer research: Methodological perspectives. Current Opinion in Food Science, 15, 30–37. Jaeger, S. R., Rossiter, K. L., & Lau, K. (2005). Consumer perception of novel fruit and familiar fruit: A repertory grid application. Journal of the Science of Food and Agriculture, 85, 480–488. Jager, S. R., & Rose, J. M. (2008). Stated choice experimentation, contextual influences and food choice: A case study. Food Quality and Preference, 19, 539–564. Kelly, G. A. (1955). The psychology of personal construct. New York, NY: WW & Norton Co. K€ oster, E. P. (2003). The psychology of food choice: Some often encountered fallacies. Food Quality and Preference, 14, 359–373. K€ oster, E. P., & Mojet, J. (2015). From mood to food and from food to mood: A psychologycal perspective on the measurement of food related emotions in consumer research. Food Research International, 76, 180–191. L€ahteenm€aki, L., & Tuorila, H. (1995). Consistency of liking and appropriateness ratings and their relation to consumption in a product test of ice cream. Appetite, 25, 189–198. L€ahteenm€aki, L., & Tuorila, H. (1997). Item-by-use appropriateness of drinks varying in sweetener and fat content. Food Quality and Preference, 8, 85–90. Lai, A. (1991). Consumption situation and product knowledge in the adoption of a new product. European Journal of Marketing, 25, 55–67. Lau, D., Hanada, L., Kaminskyj, O., & Krondl, M. (1979). Predicting food use by measuring attitudes and preference. Food Product Development, 13, 66–72. Marshall, D. (1993). Appropriate meal occasions: Understanding conventions and exploring situational influences on food choice. International Review of Retail, Distribution and Consumer Research, 3, 279–301. Martens, M., Schutz, H. G., Risvik, E., Rodbotten, M., & Raundhardt, O. (1987). Consumer perception of nutritional value related to other quality attributes and to chemical components. In J. Solms, D. R. Booth, & R. M. Pangborn (Eds.), Food acceptance and nutrition (pp. 321–355). London, UK: Academic Press. McEwan, J. A., & Thomson, D. M. H. (1988). An investigation of the factors influencing consumer acceptance of chocolate confectionery using the repertory grid method. In D. M. H. Thomson (Ed.), Food acceptability (pp. 347–362). London, UK: Elsevier. Meiselman, H. L. (2006). The role of context in food choice, food acceptance, and food consumption. R. Shepherd & M. Raats (Eds.), The Psychology of Food Choice: vol. 3. Frontiers in nutritional science (pp. 179–200). Wallingford, UK: CABI. Meiselman, H. L. (2008). Experiencing food products within a phisycal and social context. In H. N. J. Schifferstein & P. Hekkert (Eds.), Product Experience. Amsterdam, NL: Elsevier. Meiselman, H. L. (2013). The future in sensory/consumer research: Evolving to a better science. Food Quality and Preference, 27, 208–214. Meiselman, H. L. (2015). A review of the current state of emotion research in product development. Food Research International, 76, 192–199. Meiselman, H. L., Johnson, J. L., Reeve, W., & Crouch, J. E. (2000). Demonstration of the influence of the eating environment on food acceptance. Appetite, 35, 231–237. Mejlholm, O., & Martens, M. (2006). Beer identity in Denmark. Food Quality and Preference, 17, 108–115.

Situational appropriateness in food-oriented consumer research: Concept, method, and applications 139

Monteleone, E., Raats, M. M., & Mela, D. J. (1997). Perceptions of starchy food dishes: Application of the repertory grid method. Appetite, 28, 255–265. Nantachai, K., Petty, M. F., & Scriven, F. M. (1991). An application of contextual evaluation to allow simultaneous food product development for domestic and export markets. Food Quality and Preference, 3, 13–22. Petit, C., & Sieffermann, J. M. (2007). Testing consumer preferences for iced-coffee: Does the drinking environment have any influence? Food Quality and Preference, 18, 161–172. Piggott, J. R., Sheen, M. R., & Apostolidou, S. G. (1990). Consumer perceptions of whiskies and other alcoholic products. Food Quality and Preference, 2, 177–185. Piqueras-Fiszman, B., & Jaeger, S. R. (2014a). The impact of evoked consumption contexts and appropriateness on emotion responses. Food Quality and Preference, 32, 277–288. Piqueras-Fiszman, B., & Jaeger, S. R. (2014b). Emotion responses under evoked consumption contexts: A focus on the consumers’ frequency of product consumption and the stability of responses. Food Quality and Preference, 35, 24–31. Piqueras-Fiszman, B., & Jaeger, S. R. (2014c). The impact of the means of context evocation on consumers’ emotional associations towards eating occasions. Food Quality and Preference, 37, 61–70. Piqueras-Fiszman, B., & Jaeger, S. R. (2015a). The effect of product–context appropriateness on emotion associations in evoked eating occasions. Food Quality and Preference, 40, 49–60. Piqueras-Fiszman, B., & Jaeger, S. R. (2015b). Emotions associated with mealtimes: Memorable meals and typical evening meals. Food Research International, 76, 243–252. Raats, M. M., & Shepherd, R. (1991/1992). An evaluation of the use and perceived appropriateness of milk using the repertory grid method and the ‘item-by-use’ appropriateness method. Food Quality and Preference, 3, 89–100. Ratneshwar, S., & Shocker, A. D. (1991). Substitution in use and the role of usage context in product category structures. Journal of Marketing Research, 28, 281–295. Resurreccion, A. V. (1986). Consumer use patterns for fresh and processed vegetable products. International Journal of Consumer Studies, 10, 317–332. Rosas-Nexticapa, M., Angulo, O., & O’Mahony, M. (2005). How well does the 9-point hedonic scale predict purchase frequency? Journal of Sensory Studies, 20, 313–331. Rucker, M. H., & Schutz, H. G. (1982). Development of consumer typologies from appropriateness ratings. Proceedings of the 1982 Academy of Marketing Science (AMS) Annual Conference, Cham: Springer587. Schutz, H. G., Fridgen, J. D., & Damrell, J. D. (1975). Consumer perceptions of rice and related products. Journal of Food Science, 40, 277–281. Schutz, H. G. (1988). Beyond preference: Appropriateness as a measure of contextual acceptance of food. In D. M. H. Thomson (Ed.), Food acceptability (pp. 115–134). Amsterdam, NL: Elsevier. Schutz, H. G. (1994). Appropriateness as a measure of the cognitive-contextual aspects of food acceptance. In D. M. H. Thomson (Ed.), Measurement of food preferences (pp. 25–50). London: Blackie Academic. Schutz, H. G. (1995). Eating situations, food appropriateness, and consumption. In B. M. Marriott (Ed.), Not eating enough: Overcoming underconsumption of military operational rations (pp. 341–359). Washington, DC: National Academies Press. Schutz, H. G. (1999). Consumer data – Sense and nonsense. Food Quality and Preference, 10, 245–251. Schutz, H. G., Cardello, A. V., & Winterhalter, C. (2005). Perceptions of fiber and fabric uses and the factors contributing to military clothing comfort and satisfaction. Textile Research Journal, 75, 223–232.

140

Context

Schutz, H. G., & Jaeger, S. R. (2010). Innovation in sensory practices and education: An interview with Howard Schutz. In S. R. Jaeger & H. MacFie (Eds.), Consumer-driven innovation in food and personal care products (pp. 102–131). Oxford, UK: Woodhead Publishing. Schutz, H. G., & Martens, M. (2001). Appropriateness as a cognitive-contextual measure of food attitudes. In L. Frewer, E. Risvik, & H. Schifferstein (Eds.), Food, people and society (pp. 247–266). Berlin: Springer. Schutz, H. G., & Ortega, J. H. (1974). Consumer attidues towards wine. American Journal of Enology and Viticulture, 25, 33–38. Schutz, H. G., & Phillips, B. A. (1976). Consumer perceptions of textiles. Family and Consumer Sciences Research Journal, 5, 2–14. Schutz, H. G., & Rucker, M. H. (1975). A comparison of variable configurations across scale lengths: An empirical study. Educational and Psychological Assessment, 35, 319–324. Schutz, H. G., Rucker, M. H., & Russell, G. F. (1975). Food and food-use classification systems. Food Technology, 29, 50–64. Scriven, F. M., Gains, N., Green, S. R., & Thomson, D. M. H. (1989). A contextual evaluation of alcoholic beverages using the repertory grid method. International Journal of Food Science & Technology, 24, 173–182. Sester, C., Deroy, O., Sutan, A., Galia, F., Desmarchelier, J.-F., Valentin, D., et al. (2013). “Having a drink in a bar”: An immersive approach to explore the effects of context on food choice. Food Quality and Preference, 28, 23–31. Shepherd, R., Schutz, H. G., & Sparks, P. (1993). Prediction of frequency of food consumption from use attitude and demographic variables. Food Quality and Preference, 4, 89. Sosa, M., Martı´nez, C., Arruiz, F., Hough, G., & Mucci, A. (2005). Degree of appropriateness and frequency of consumption of mayonnaise, ketchup, mustard and similar sauces in Argentina. Food Quality and Preference, 16, 667–674. Stefflre, V. (1971). Some eliciting and computational procedures for descriptive semantics. In P. Kay, P. Reich, & M. McLaran (Eds.), Explorations in mathematical anthropology. Cambridge, MA: MIT Press. Stolzenbach, S., Bredie, W. L., Christensen, R. H., & Byrne, D. V. (2016). Understanding liking in relation to sensory characteristics, consumer concept associations, arousal potential and “appropriateness for use” using apple juice as an application. Journal of Sensory Studies, 31, 135–142. Thomson, D. M. H., & McEwan, J. M. (1988). An application of the repertory grid method to investigate consumer perceptions of food. Appetite, 10, 181–193. Tuorila, H., Hyv€onen, L., & Vainio, L. (1994). Pleasantness of cookies, juice and their combinations rated in brief taste tests and following ad libitum consumption. Journal of Sensory Studies, 9, 205–216. Tversky, A., Sattah, S., & Slovic, P. (1988). Contingent weighting in judgment and choice. Psychological Review, 95, 371–384. Tversky, A., & Simonson, I. (1993). Context-dependent preferences. Management Science, 39, 1179–1189. Wansink, B. (2004). Environmental factors that increase the food intake and consumption volume of unknowing consumers. Annual Review of Nutrition, 24, 455–479. Warlop, L., & Ratneshwar, S. (1993). The role of usage context in consumer choice: A problem solving perspective. Advances in Consumer Research, 20, 377–382. White, F. D., Resurreccion, A. V. A., & Lillard, D. A. (1988). Consumer characteristics, food related attitudes and consumption of meat and meat products. Nutrition Research, 8, 1333–1344.

Section B Meals in context

This page intentionally left blank

Food choices in context Maartje P. Poelman*, Ingrid H.M. Steenhuis† *Utrecht University, Utrecht, The Netherlands, †VU University Amsterdam, Amsterdam, The Netherlands

7

Every day people have to make food choices. People have to decide what to eat, when to eat, and where to eat. People also have to decide where to buy their food. In the supermarket, at the farmers market, in a convenience store, or in one of the other many stores that sell food nowadays. Will people prepare their own meal, or will they go out to a restaurant, to a fast-food outlet, or get a take-away? When on the go, do people purchase snacks, meals, or drinks at the petrol or train station, or do they take snacks from home? Day in, and day out, people have to make these decisions and choose what to eat. Although these decisions may seem to be properly thought-out and rational, many food choices are habitual, automatic, and steered by social and physical environmental cues. Moreover, food choices are driven by bigger societal forces such as globalization, welfare, and urbanization. Therefore, peoples’ contexts cannot be ignored in research into food choices. This chapter discusses the context of food choices. First, a theoretical underpinning of the importance of context in food choice research is outlined. Thereafter, different contextual levels are discussed. We provide an overview of the macro context, local context, and the social context. Subsequently, we provide insight into several contextual settings in which food consumption or food purchases are taking place. These are the supermarket, workplace cafeteria/restaurant, home, and online/digital world. For each context, sample studies are provided to illustrate influences on food choices. The majority of these studies are in the field of public health nutrition, social psychology, and consumer marketing.

7.1

Why is the food choice context important?—A theoretical perspective

Food choices are complex. The reason why we eat what we eat is multifactorial, and changes from person to person. Not only do individual factors (such as neurobiological, physiological, psychological determinants) contribute to eating decisions, but also wider sociological, ecological, environmental, and cultural factors (e.g., culture and economics of food production) steer food choices. Yet, the conventional economic viewpoint is that humans are rational creatures who make deliberate decisions in their own best interest. However, over the past decades, evidence has revealed that people are often irrational and food choices are not always deliberated or in people’s best interest. Food choices are often the result of automatic and undeliberated processes. In peoples’ food choices, senses such as visual, olfactory, and auditory factors Context. https://doi.org/10.1016/B978-0-12-814495-4.00007-6 Copyright © 2019 Elsevier Inc. All rights reserved.

144

Context

also play a role, but often occur without full awareness, and food choices are, in large part, the result of undeliberated responses to contextual food cues, many of which lead to increased caloric consumption and poor dietary choices (Cohen & Babey, 2012). The fact that people do not always make food choices in their best interest is shown by the rise in obesity over the past few decades. Persistently consuming more calories from foods than are expended leads to a positive energy balance that in turn, results in weight gain. The energy intake side of the balance, which is the result of food choices, has been recognized as a dominant driver of the rise in obesity (Slater et al., 2009; Swinburn, Sacks, Lo, et al., 2009). In the short term, food intake can serve the best interest of good taste, satiation, and pleasure that food can bring. However, in the long term, overconsumption does not serve the best interest of individuals. Some individuals with obesity are stigmatized, bullied, experience chronic co-morbidity (e.g., type 2 diabetes, cardiovascular disease and depression), and their life expectancy drops considerably. It is argued that being overweight and obesity are the result of a normal response to an abnormal environment. In terms of economics, obesity (and overconsumption) can be seen as the result of market failure, as the market fails to deliver the best output for society, and puts a long-term burden on healthcare costs (Anand & Gray, 2009; Moodie, Swinburn, Richardson, & Somaini, 2006). Several theories underpin the influence of peoples’ contexts in behaviors such as food choices. These theories go beyond the more traditional health behavior conceptual models that explain individual factors of behaviors such as food choices. For example, according to the theory of planned behavior, peoples’ intention to engage in a certain behavior is central. Intentions are clear decisions to act in a certain way, and seen as an indication of how hard people are willing to try, and how much effort they want to put in to a certain behavior. The stronger the intention, the more likely it is that the behavior will be engaged in (Ajzen, 1991). Yet, intention cannot fully explain behavioral decisions. Moreover, people’s intentions to act in a certain way (e.g., I want to eat vegetables every day) do not always translate into the actual behavior (e.g., actual daily vegetable intake), also called the (Sheeran and Webb, 2016) which also shows that other—personal and environmental—factors steer behavior. Many theories and conceptual models with respect to the influence of contextual influences on food choices have been developed over the past few decades. For example, the social cognitive theory (Bandura) describes how individuals develop and maintain certain behavioral patterns, such as food consumption. According to this theory, a dynamic interaction between the person, the persons’ behavior, and the environment in which the behavior is performed steers people’s behavior, and these factors constantly influence each other (Bandura, 1989). Individual factors include cognitive, affective, and biological events. The factor behavior is rather broad, and includes, for example, people’s capacity to perform a behavior (e.g., skills to prepare a healthy meal), expectations of the behavior, and people’s self-efficacy. The environment includes social and physical influences. Social influences are behaviors of family members, peers, or role models. By observing the behaviors of others, people learn to behave in a certain way. The physical environment includes factors that steer

Food choices in context

145

behavior, such as the availability and prices of foods. Besides that, the actual environment, and people’s perception of the environment, also need to be taken into account. Also, the dual process theory reflects the importance of the context in which food choices are made, and provides an understanding of two different psychological processes steering people’s behavior and choices. On one hand, behavior and choices are made unconsciously or automatically, and may result in system 1, which includes behavioral responses to environmental factors of which individuals may even not be aware. These responses are intuitive, fast, unintentional, and emotional. On the other hand, behavior can be the result of system 2, including well-deliberated, controlled decisions that are rational, reasoned, and logical (Kahneman, 2011). A dual process view is used to gain an understanding of the circumstances in which food choices are made. According to the Environmental Research Framework for Weight Gain prevention (EnRG framework), food choices may be conscious decisions in response to the environment, or a result of undeliberated responses to contextual cues (Kremers et al., 2006). For example, a supermarket in which fruits and vegetables are nicely presented and advertised may positively steer individual attitudes and beliefs toward eating fruit and vegetables, and might improve people’s intention to buy these products (system 2). On the other hand, the sight and the nice smell of fruits and vegetables may prompt individuals to purchase these products (system 1) (Brug, Kremers, van Lenthe, Ball, & Crawford, 2008). Ecological models of behavior also highlight the influence of multiple environmental contexts (e.g., physical and policy circumstances) of behaviors, while taking individual, social, and psychological influences into account. Many ecological models have been developed over the past decades, all including this multi-level structure, and underpinning the interacting effect of factors across these levels on food choice behaviors (Sallis, Owen, & Fisher, 2015). Therefore, food contexts may have a different impact on different individuals. For example, a food truck in the neighborhood (local context) selling ice cream on a hot summer day may have a different impact on an individual with a strong intention to eat healthily versus an individual who does not, or on a single person versus a family with children walking by (Brug et al., 2008). Food policies or regulations may also have a different impact on different people. For example, the recently introduced soft-drink tax is having a different impact on people with a low socio-economic position, compared with those with a high socioeconomic position (Batis, Rivera, Popkin, & Taillie, 2016). To summarize this overview, several theories and scientific viewpoints underpin the importance and interaction between personal, intrapersonal, and environmental determinants, indicating that several circumstances influence food choices. One should be aware that there are several other theories and models that address the importance of contextual influences on food choices, but this paragraph underpins some of the theories that shaped the field of contextual food choices. Besides, in line with existing ecological models, we compose a food choice context framework (Fig. 7.1), indicating different contexts that will be outlined in this chapter that should be considered when conducting food choice research (Fig. 7.1).

146

Context

Fig. 7.1 Contexts of food choices.

7.2

Macro context

The macro food context is an intangible and invisible environment. It is not a welldefined place or setting where people come together, but is rather an anonymous context that steers the characteristics of micro-level food contexts (Swinburn, Egger, & Raza, 1999). For example, food production, safety, and price; and trade, agriculture policies, regulations, and agreements; which impact all aspects of the food supply chain (from production to sales), fall under the macro-context, and may steer national and local environments in which food choices take place. For centuries, people have attempted to overcome food scarcity and hunger, as they were dependent on the success of crops that were influenced by seasonal variations (Kapsiotis, 1975). The onset of the industrial revolution was crucial for stable food production, and led to an increase in the availability of dietary energy (Caballero, 2007). However, after World War II, evident changes in agricultural policies (e.g., subsidies), farming practices (e.g., factory farming), and technical innovations (e.g., irrigation) (Bleich, Cutler, Murray, & Adams, 2008; Tillotson, 2004; Wallinga, 2010) influenced the macro food context. These technical, political, and organizational developments in the past resulted in a farm system that was less labor-intensive and more efficient, that accelerated growth, and eventually resulted in the mass production of food (Swinburn, Sacks, Hall, et al., 2011). More societal developments, such as globalization, urbanization, and increases in economical welfare are also macro contextual factors that steer food choices. The so-called nutrition-transition shows a shift in dietary patterns due to several macro contextual changes over the past several decades (Popkin, 1993). For example, as a result of globalization, Western food companies spread to low-income and middle-income countries, as a result of which Western, often highly processed food, became available and more easily accessible. For example, the number of McDonald’s outlets increased from 5 in 1985, to 100 in 1992, to 214 in 1996, to 568 in 2001 in Brazil, reflecting an increase of 11.280% in a 16 year period (Gheza´n, Mateos, & Viteri, 2002). Also, in the Asia Pacific region, the number of these outlets increased, from 951 outlets in 1987 to 7135 in 2002, reflecting an increase of 650% in a 15 year period (Trail, 2017).

Food choices in context

147

In the United States, the number increased from 7000 outlets in 1985 to 13,099 outlets in 2001, reflecting an increase of approximately 87% (Gheza´n et al., 2002). Although the number of these fast food outlets is still higher in the USA, the relative growth has been higher in the lower-middle income countries. Also, other food sources, such as supermarkets (Reardon & Berdegu
e, 2002; Weatherspoon & Reardon, 2003), entered and expanded in developing countries. Such macro contextual influences have caused changes in food consumption patterns (Drewnowski & Popkin, 2009). For example, in China, animal-based foods tripled from 30 g to 103 g per day, and the energy intake from fat increased from 7.6% to 22.5% between 1949 and 1992 (Du, Wang, Zhang, Zhai, & Popkin, 2014). These changes are also reflected by the increasing number of residents who are overweight and obese in these countries (Popkin, Adair, & Ng, 2012). By way of an illustration, the number of children (2–18 years) considered obese has risen substantially in the past two decades. In 1991, 6.2% of Chinese children were overweight. In 2000, this had risen to 8.5%, and in 2011, the percentage was 15.4% (Gordon-Larsen, Wang, & Popkin, 2014). In the years ahead, ongoing challenges will define and reshape the macro context of food choices. For example, technological and digital developments, the growing world population, and sustainable development goals will drive the macro-contextual playing field of food choices. Macro contextual factors that become visible in the local contexts are, for example, those of pricing strategies (e.g., the price of food, as affected by subsidies and taxes) and regulations shaping product characteristics (e.g., information presented on food labels or the size of food packages). Pricing strategies affect food choices, and are used to improve food choices. A review including pooled analyses indicated that a price decrease of 10% on healthy foods increased the consumption of healthy foods by 12% (95%CI ¼ 10%–15%) and, more specifically, increased the consumption of fruits and vegetables by 14% (95%CI ¼ 11%–17%). A 10% price increase on unhealthy foods decreases the consumption of unhealthy products by 6% (95%CI ¼ 4%–8%). Each 10% price increase reduced sugar-sweetened beverage intake by 7% (95% CI ¼ 3%–10%) (Afshin, Pen˜alvo, Del Gobbo, et al., 2017). Worldwide, several countries have already implemented a sugar sweetened beverage tax. For example, Mexico implemented a tax on nonessential foods and sugar sweetened beverages in 2014. Observational data indicates a decline in purchases of these products after one year (5.1%), with larger effects among low SES households (10,2%) (Batis et al., 2016). Another example is that of food labeling regulations. In many high income countries, it a requirement to provide nutritional information on all packed food and beverage items, including a recommended serving size that reflects “the recommended amount to consume in one sitting.” However, different international regulations exist. In Australia and New Zealand, it is mandatory to display a recommended serving size, although standard serving sizes for food and beverages are not provided. In contrast, regulated serving sizes are available for the USA and Canada, including acceptable ranges in milliliters, within which beverage serving sizes must fall. What is more, package sizes differ substantially between countries. For example, the mean package size (bottle) of Dutch SSBs ((1313 (323) mL) is significantly smaller compared with the bottle sizes in New Zealand ¼1481 (595) mL,

148

Context

Australia ¼ 1542 (595) mL or Canada ¼ 1550 (434) mL) (Poelman, Eyles, Dunford, et al., 2016). Nevertheless, lacking national regulations also influence serving size availability. For example in the Netherlands, coffee, milkshakes, and ice cream are often for sale in different sizes, denoted with “small,” “medium,” or “large.” Yet, which amounts (e.g., milliliters) reflect these annotations are variable. For example, the sizes of milkshakes varied for different estimates; small was 210–490 mL; medium was 340–570 mL; and large was 440–660 mL (Poelman & Steenhuis, unpublished data).

7.3

Local context

With respect to the local context of food choices, we refer in this chapter to certain areas (e.g., neighborhood, city, or municipality) that include a range of facilities and settings in which food choices take place (Swinburn et al., 1999). There are several ways to conceptualize the local context in research with respect to food choices (Thornton, Pearce, & Kavanagh, 2011). Often used are the availability and the accessibility of food outlets (e.g., supermarkets, convenience stores, restaurants) in people’s living environment, or the so-called community nutrition environment (Glanz, Sallis, Saelens, & Frank, 2005). Availability refers to the presence and number of certain food outlets available in a pre-defined area. Public buildings (train stations), schools, recreation and sports facilities offering foods may also be included in the local contexts. Accessibility refers to the easiness of access to a particular food outlet, with more accessible destinations having lower travel costs in terms of distance, time, or monetary resources (Handy & Niemeier, 1997). Both the accessibility and availability of food outlets have been studied in heterogeneous ways; for example, studies define the food environment as the number of food outlets in a certain area (census tract, neighborhood, street network buffer around the home; distance to the nearest store, travel time to the store), or perceived availability or accessibility of certain products (e.g., fruits or vegetables) (Caspi, Sorensen, Subramanian, & Kawachi, 2012a). Later in this chapter we discuss the characteristics and impact on food choices of individual settings in more detail (also called the consumer nutrition environment) (Glanz et al., 2005), but here we focus on the overall local context in which all these settings are established. Several studies have been conducted on the local food context over the past few decades. On one hand, studies have been conducted to measure and define the local food context, and to detect changes over time. On the other hand, multiple studies have been conducted to examine associations between the local food context and food choices (e.g., food intake). Many studies with respect to defining the local food context have focused on so-called “food deserts,” defined as disadvantaged urban areas with poor availability and limited access to healthy foods (i.e., Osorio, Corradini, & Williams, 2013). In high-income countries, often the presence or absence of supermarkets in a certain area is taken into account to define food deserts as selling fresh and healthy foods such as fruits and vegetables. There is some evidence of the presence of food deserts in the United States, although evidence of food deserts is mixed for Canada (Larsen &

Food choices in context

149

Gilliland, 2008; Lu & Qiu, 2015; Smoyer-Tomic, Spence, & Amrhein, 2006), or is lacking in European countries (e.g., in the Netherlands) (Helbich, Schadenberg, Hagenauer, & Poelman, 2017). Nevertheless, most studies have been conducted in smaller areas (e.g., in cities), most often in high-income countries, and for many low- and mid-income countries, insight in the presence or absence of food deserts is lacking. Nevertheless, the concept “food desert” has been disputed over the past few years, as only a small number of residents shop for groceries within their own census area, and can use a car to access a supermarket that is not in walking distance. For example, a study showed that the vast majority of households (93%) living in food deserts have access to a car (Ploeg, Breneman, Farrigan, et al., 2009). Another point of critique is that food deserts may have been used as strategy to open more supermarkets. Yet, it is debatable whether simply enlarging food outlets that offer healthy food options, without restrictions on unhealthy options, has a beneficial impact on healthy food choices (Cohen, Sturm, Scott, Farley, & Bluthenthal, 2010; Cummins, Petticrew, Higgins, Findlay, & Sparks, 2005; Wrigley, Warm, & Margetts, 2003). By way of an illustration, a quasi-natural experiment that evaluated the impact of the opening of a large supermarket in a deprived Scottish community showed no population impact on daily fruit and vegetable consumption (Cummins et al., 2005). In response to food deserts, the term “food swamps” was introduced more recently to define local food contexts in which healthy fresh and whole foods are available, but where there is an overabundance of ultra-processed (unhealthier) foods sold in excessive numbers of food stores, such as fast food outlets, convenience stores, or petrol stations (Luan, Law, & Quick, 2015; Osorio et al., 2013). To illustrate, a Dutch study has been conducted recently, measuring the local food context within a 400 m walking distance of secondary schools in a large city in the Netherlands. The authors conclude that unhealthy food and drink products were predominantly for sale and promoted. Fruit was for sale in less than a quarter of the outlets around the schools (23.5%), and this was even less for vegetable snacks (12.2%); whereas sugar sweetened beverages (84.3%) were more often available than lite drinks (77.4%) or bottled water (76.0%) (Timmermans et al., 2018). Another recent study from the United States even indicates that the presence of food swamps is a stronger predictor of obesity rates than food deserts (Cooksey-Stowers, Schwartz, & Brownell, 2017). Theoretically, neighborhoods offering low access to healthy foods are expected to steer individual food choices disadvantageously with respect to health, and vice versa for neighborhoods offering high access to healthy food. Studying the influence of local food contexts on food choices is empirically complex, and evidence is inconsistent. A review study including 38 studies that explored the local food environment and diet found moderate evidence in support of the causal hypothesis that neighborhood food environments influence dietary health (Caspi, Sorensen, Subramanian, & Kawachi, 2012b). A large issue was that the overall reproducibility was lacking because of the absence of a standard measure of local food access, and assessment measures varied considerably across the included studies (Caspi et al., 2012b). Similar limitations were mentioned by a more recent review, including 51 studies on the community food environment-diet relationship. Moreover, only 32% of the associations between the food environment and obesity-related outcomes were in the expected

150

Context

direction, 58% showed no association, and 10% were in the unexpected direction (Gamba, Schuchter, Rutt, & Seto, 2015). The most recent review, including 432 studies that were conducted between 2007 and 2015, emphasize that work is needed to understand construct validity as it relates to measures used to assess the food environment in studies determining the association with dietary behaviors. The authors conclude that it remains important to understand how the food environment influences individual and population-level health. Additionally, they stress that it is important to see the food environment as one construct in a larger, ecologically conceptual model, and that we need to enhance the level of interdisciplinary work between different disciplines working in the field (Lytle & Sokol, 2017). There are also several limitations in the literature to understanding the impact of the local food context, such as the majority of cross-sectional studies, and the wide variety of measures of food-outlet availability and accessibility (e.g., reliance on commercial business listings, ignoring changes in availability over time) (Gordon-Larsen, 2014). Longitudinal studies are needed to discern causality, as approximately 14% of the studies included had a longitudinal design (Lytle & Sokol, 2017). What is more, most studies conducted focused on achieving a better understanding of people’s interaction with a static local food context. From a this static viewpoint, the local food context is a predefined area, often related to people’s living environment; for example, the availability and accessibility of food outlets in residential neighborhood and census tracks, or “buffers” around the exact addresses of the individual (e.g., all outlets that can be accessed within 500 m). Yet, during the day, people not only undertake daily activities in close proximity to their home, but also travel to diverse places (e.g., work, shopping center) where they also face facilities and settings in which they make food choices. From this dynamic perspective, not the residential area, but the individual is the starting point in defining his or her personal local food context, or also people’s activity space (Crawford, Jilcott Pitts, McGuirt, Keyserling, & Ammerman, 2014). Although the latter may define exposure more specifically, one should be aware of selective daily mobility bias. This bias refers to the tendency of people to choose routes based on their personal needs and preferences, and may bias the association between the local food context and food choices (Chaix, M
eline, Duncan, et al., 2013). Therefore, insights about the decision-making process underlying people’s food choices in the local contexts are a necessity. Yet, most studies conducted so far lack this information (Gordon-Larsen, 2014). Although we currently lack a full understanding of the interplay between the local food context and food choices, including underlying mechanisms, local policy efforts have been introduced to improve the accessibility of healthier food outlets in the local food context. For example, the London (UK) mayor recently announced (2017) a plan to ban fast-food outlets from opening within a quarter mile radius of primary and secondary schools in London. In Amsterdam (the Netherlands), food companies are no longer allowed to advertise unhealthy products aimed at citizens younger than 18 in almost 60 metro stations. These examples show efforts to shape the local context to steer food choices in a positive direction with respect to health. The effectiveness of these efforts needs to be evaluated in the years ahead.

Food choices in context

7.4

151

Social context

Food and eating are entangled with our social lives, and often people eat together. The social context encompasses social relationships and cultural milieus within which defined groups of people function and interact. Individuals can often have multiple social environments simultaneously (e.g., family, colleagues, friends) that are dynamic and change over time (Barnett & Casper, 2001). People around us—our social context—may regulate, influence, or constrain our eating behavior. Several social influences on food intake are described in the literature, and also addressed more extensively in Chapter 2. Social facilitation of eating refers to the phenomenon of increased food consumption when people eat together instead of eating alone. In 1989, the first study into the particular influence of the number of other people and food intake was explored by means of an uncontrolled food diary study. Over a one-week period, participants filled out a food diary, including the social conditions of the meal. De Castro and de Castro found that individuals eating together ate significantly more (on average 44%) than individuals eating alone (de Castro & de Castro, 1989). Various experimental studies followed in subsequent years, showing comparable results: intake during group dinners is higher than intake during solo dinners (Berry, Beatty, & Klesges, 1985; Edelman, Engell, Bronstein, & Hirsch, 1986). In particular, social facilitation is present when participants eat together with others (rather than in the presence of a noneating audience) (Hetherington, Anderson, Norton, & Newson, 2006; Salvy & Pliner, 2010), and when eating together with friends (rather than strangers) (de Castro, 1994). In practice, social facilitation has been implemented and tested as a strategy in retirement homes to encourage residents to increase their meal intake. In a Dutch intervention study in a nursing home, the effect of family-style meals (e.g., eating together with other residents) versus individual preplating services on food intake was tested over a sixmonth period. The study showed that the intervention group (family-style meal) significantly increased their intake (992 kJ; 95%CI ¼ 504–1479) in comparison with the control group (preplated solo meal) (Nijs, de Graaf, Siebelink, et al., 2006). Several underlying explanations have been suggested for the social facilitation effect, although there is no definitive explanation. In a recent review, Herman (2015) outlines potential underlying mechanisms in detail, and proposes future directions for research to enhance our understanding of the social facilitation of eating (Herman, 2015). Via social modeling of food choices, people adapt their food choice to that of the food choice of their companion. In modeling, people use others’ food choices as a guide for their own behavior. A recent study reviewing 69 modeling studies over 40 years, including >5800 participants, indicated that the majority of the studies (92.8%) found significant modeling effects on food intake, irrespective of the social context, methodology, food consumed, or demographics of the studied group (Cruwys, Bevelander, & Hermans, 2015). For example, a study of social modeling effects on food purchase behavior in supermarkets examined whether food purchase

152

Context

behavior of teenage girls would adapt to the same-sex confederate peer that purchased either low-caloric, mid-caloric, or high-caloric food products. Results suggested that teenage girls who shopped with a peer who chose high-caloric food products purchased higher caloric food products compared with the girls who shopped with a teenage peer buying low-caloric foods (Bevelander, Ansch€u, & Engels, 2011). First, people “model” because they search for information of the appropriate or “correct” behavior. Second, individuals model because they want to affiliate with others and be liked (Cruwys et al., 2015). Social norms are “implicit codes of conduct” or “perceived standards” that exist in a social group and provide a guide to appropriate action (Higgs, 2015). There are two types of social norms. Descriptive norms refer to individuals’ perceptions about how others around them behave (e.g., make food choices), so “what do others do?”. Injunctive norms refer to the perceived approval of food choice behavior and represent perceived moral rules of the peer group, so “what do other people accept?” (Cialdini, Reno, & Kallgren, 1990; Reno, Cialdini, & Kallgren, 1993). Previous studies have outlined the relation between social norms and food choices (Robinson, Thomas, Aveyard, & Higgs, 2014; Stok, de Vet, de Ridder, & de Wit, 2016). For example, in a lab and real-life study, information about how others behaved was controlled by using empty chocolate wrappers, indicating that others ate chocolates on a previous occasion. In the real-life study, customers of bakery lunchrooms participated. A transparent bowl with individually wrapped chocolates was placed on the food counter. In the experiment, a bowl with empty wrappers was placed beside the bowl of chocolates. In the control condition, this bowl was empty. Outcome measures were the number of chocolates consumed. The results indicated that the number of consumed chocolates was higher when it was indicated that previous customers had taken a chocolate (RR 2.10, 95%CI ¼ 1.08–4.09). The follow-up study in a laboratory setting resulted in similar outcomes (Odds ¼ 3.07, 95%CI ¼ 1.09–8.60) (Prinsen, De Ridder, & De Vet, 2013). These prior examples of social influences on food choices illustrate that individuals are not always fully aware of social influence. Yet, social influences do not always occur unnoticed, and social influences are also embedded in family or group structures and dynamics. This is nicely illustrated by a qualitative study examining the role wives played in shaping the eating behaviors of middle-aged and older urban AfricanAmerican men. The men indicated that women played a dominant role in household food provision and decision-making, and agreed that their wives influenced what they ate at home more than their own preferences. Quoting from this study: “When we first got married at 21 and 20, my wife decided that we weren’t going to eat like our parents. She made that decision for me. I didn’t think about it.... She made the decision that we’re not going to do this, and I didn’t argue with her.” Or “All my life I’ve been influenced by either my mother or my wife as far as food choices. I really didn’t have any choice other than what she put in front of me at the table” (Allen, Griffith, & Gaines, 2013). Although the interviewed men perceived themselves to have little control over what they ate at home, they appreciated the care and concern of their wives. This example illustrates that people experience and recognize social control others have over their food choices.

Food choices in context

7.5

153

Setting context

Whereas macro and local food contexts shape the overall scene in which populationlevel food choices take place, people interact with the food environment in specific settings. In this paragraph, we discuss five settings that are important for food choices, and objectives for food choice-related research: home, supermarket, workplace cafeteria/restaurants, and the digital/online setting. For each setting, we predominantly focus on the physical or policy factors that influence food choices. Yet, one should be conscious of other contextual factors (e.g., social influences) that also shape food choices being made.

7.5.1 Home The home food environment is fundamental in the development of food preferences and consumption habits, steering food choices. In fact, the home food environment is the place where the retail food environment interacts with actual food intake. The home food environment can be conceptualized as coinciding interactive domains made up of different environments, namely the socio-cultural, political, economic, and the built environments (at both micro and macro levels) that each contributes in a unique way to the determinants of the home food context (Rosenkranz & Dzewaltowski, 2008). Home contextual factors that shape food consumption include the availability of products, the salience of foods, and the size of dinnerware (Robinson, Nolan, Tudur-Smith, et al., 2014). Previous studies have illustrated the interdependence of home food availability and food intake (Campbell et al., 2007; Ding, Sallis, Norman, et al., 2012). Moreover, storing foods in visible places increased individuals’ consumption rates, consumption frequencies, and the amount of food consumed, especially for high-convenience foods and large packages (Chandon & Wansink, 2002). The salience of snack foods evokes individuals’ desire to eat, prompts their desire to consume larger amounts, and increases the actual amount they consume (Fedoroff, Polivy, & Herman, 1997; Ferriday & Brunstrom, 2008). For both high-and low-convenience foods, large amounts of stockpiled foods also induce increased usage and the intake of larger amounts (Raynor & Wing, 2007). Another factor in the home food environment that may influence food intake is dinnerware size. It has been implied that consuming from larger dishware affects the amount people consume, because people serve themselves a larger portion (Sobal & Wansink, 2007). Using mathematical modeling to estimate the influence of dinnerware size on energy intake, a study indicated that small increases in dishware (plates and bowls) could lead to a substantial increase in energy intake (Pratt, Croager, & Rosenberg, 2012). A systematic review and meta-analysis from 2014 that included eight heterogeneous studies concluded that the majority found no significant difference if people consumed food from small or large dishware (Robinson, Nolan, et al., 2014). However, contrary to this finding, a meta-analysis from 2015, which included 13 independent comparisons from 10 studies, found a small to moderate effect of

154

Context

portion or tableware size on selection of food among adults. They also found a very large effect of exposure to differently shaped tableware on the selection of nonalcoholic beverages in adults and children (Hollands, Shemilt, Marteau, et al., 2014). In a study in the Netherlands among 278 households, it was found that the majority (70%) had a large amount of ultra-processed snack foods in stock (8 packages). In 33% of the households, processed snack foods were visible in the kitchen, and in 15% of the households, processed snack foods were visible in the living room. Of the dinnerware items, 14% (plates), 57% (glasses), 78% (dessert bowls), 67% (soup bowls), and 58% (mugs) were larger than the reference of the Netherlands Nutrition Centre (Poelman, De Vet, Velema, Seidell, & Steenhuis, 2015).

7.5.2 Supermarkets Supermarkets are a major source for food purchases, and subsequently, food intake (Stern, Ng, & Popkin, 2016). Many contextual factors within the supermarket influence grocery shopping, as many purchase decisions are made in-store (Cohen & Babey, 2012). Factors that play a role in these decisions are variety, place, promotion, and price. These in-store influences go mostly unnoticed by shoppers, as they target unconscious decisions (e.g., via system 1 according to the dual process theory). Generally, in-store contextual influences often lead to unhealthier dietary choices, and an increased caloric consumption (Cohen & Babey, 2012). The number of products available in supermarkets has increased immensely in the past few decades, to >40,000 nowadays (Cohen & Babey, 2012). The sales of processed and ultra-processed foods have particularly increased. For example, expressed as a percentage of total purchased calories, the sales of processed foods have increased over the past few decades. In Canada, the increase was from 24.4% in 1938 to 54.9% in 2001. In Brazil, it was from 18.7% in 1987 to 26.1% in 2003 (Monteiro, Moubarac, Cannon, Ng, & Popkin, 2013). A study in New Zealand supermarkets showed that >80% of products available could be classified as “ultraprocessed” (Luiten, Steenhuis, Eyles, Ni Mhurchu, & Waterlander, 2016). The increased availability may be overwhelming for shoppers, and may lead to “choice stress.” Iyengar & Lepper demonstrated that shoppers were more inclined to buy jam if they could choose between 6 flavors, compared with choosing among 24 flavors (Iyengar & Lepper, 2000). On the other hand, studies have shown that a wider variety of products leads to an increase in sales (Sela, Berger, & Liu, 2009). In terms of placement, the routing in supermarkets is designed to expose consumers to as many products as possible. High exposure locations are ends of aisles, shelves at eye level, free-standing product displays, and products displayed at checkouts. The number of facings of a product on the shelf has an influence (Chandon, Hutchinson, Bradlow, & Young, 2009). The in-store promotion of products can be done in different ways; for example, via product sampling, which is known to be an effective method to increase product sales. Research has demonstrated that, when offered, 70% of the shoppers consumed a free sample, of whom 40% bought the sampled food (Heilman, Lakishyk, & Radas, 2011).

Food choices in context

155

Advertisements on shopping carts, or by in-store audio are other examples of product promotion. Price discounts and promotions in supermarkets (such as “buy 1 get 1 free”) are often used to steer consumer behavior (Cohen & Babey, 2012). This is no surprise, because price is—next to other determinants of food choice, such as sensory appeal or convenience—a major determinant of food choice, and applies even more for consumers with a low socio-economic background (Steenhuis, Waterlander, & de Mul, 2011). Larger volumes offer consumers more “value for money,” that is, a relatively lower price per unit of the product (Vermeer, Alting, Steenhuis, & Seidell, 2010). Moreover, portion size is an important product feature that impacts purchase or intake. A development that has taken place over the past few decades with respect to product characteristics is the enlargement of portion and package sizes, and the introduction of multi-packages (Steenhuis, Leeuwis, & Vermeer, 2017; Steenhuis & Poelman, 2017). Larger portions and packages increase used volume (Hollands, Shemilt, Marteau, et al., 2015). Besides price and portion, the shape and color of the packaging, as well as the images used on the packaging, also play a role in attracting the attention of consumers, and that, in turn, determines purchases of the product, in addition to the perceived quality of the product (Cohen & Babey, 2012). The majority of price promotions in supermarket flyers are in the category of unhealthy food products (Ravensbergen, Waterlander, Kroeze, & Steenhuis, 2015). Health professionals have developed interventions to promote the purchase of healthier groceries in the supermarket by using price interventions. Waterlander et al. tested the effects of a 50% price discount on selected fruits and vegetables among Dutch consumers with a low socio-economic position (Waterlander, de Boer, Schuit, Seidell, & Steenhuis, 2013). The results of the randomized controlled trial indicated that the intervention resulted in substantially higher fruit and vegetable purchases during a 6 month period, and this effect was even stronger if combined with an educational program. A randomized controlled trial conducted in Australia showed that a 20% price reduction in fruits and vegetables was also effective, and led to a 35% increase in the purchases of fruit, and a 15% increase in the purchase of vegetables (Ball, McNaughton, Le, et al., 2015). Adam and Jensen conclude in their review that, of all interventions conducted in the supermarket setting to encourage people to eat more healthily, consumers respond most to economic incentives (Adam & Jensen, 2016). It is very challenging to study pricing interventions in real life. Therefore, virtual supermarkets have been developed (Waterlander, Jiang, Steenhuis, & Ni, 2015). These 3D software programs simulate a shopping experience, and make it feasible to study price interventions that are not yet feasible, or are difficult to implement and study in real life, such as food taxation. A randomized trial with a virtual supermarket showed, for example, the potential of a higher VAT-rate on sugar sweetened beverages (Waterlander, Ni Mhurchu, & Steenhuis, 2014). It is important to continue investing in IT developments in virtual technology and make sure it is up-to-date with the most recent computer systems (Poelman, Kroeze, Waterlander, de Boer, & Steenhuis, 2017).

156

Context

7.5.3 Cafeterias, restaurants, and all-you-can-eat Workplace cafeterias are considered a relevant setting to study and intervene in food choices. Employees visit workplace cafeterias on a regular, or even daily, basis during their career. A qualitative study among employees into motives and drivers for visiting the workplace cafeteria showed that employees wanted to have a break from their work, and valued the convenience of purchasing their meal at the cafeteria. It also transpired that healthiness played a less important role in selecting foods at the workplace cafeteria. This was opposite of the key drivers mentioned for food selection in general, in which health was considered important. Reasons for unhealthy choices at the workplace cafeteria were feeling tempted by the offerings, and people feeling they deserved it after hard or stressful work (Velema, Vyth, & Steenhuis, 2019). Several factors in the workplace cafeteria setting are considered important when determining food choices. Characteristics of the products offered, the placement in the cafeteria, the price of the product, and the way products are promoted are all relevant (Velema, Vyth, & Steenhuis, 2017). Relevant product characteristics are the availability, the portion size, and convenience (for example, peeled fruit snacks). With respect to placement, products that are placed at the beginning of the route, and highly visible products, are purchased more. Visibility relates to placement at eye level, the visible share of products, and placement near the checkout. In addition, price promotions also have an effect, as well as free offerings of, for example, water. Promotional tactics that enhance sales numbers of products and meals include using attractive names on the menu, and promotional displays (Velema et al., 2017). Whether food labeling in the workplace cafeteria influences food choice is still the subject of debate (Afshin, Penalvo, Del Gobbo, et al., 2015). We found no meaningful effects of labeling with a nutrition logo on sales of sandwiches, soups, snacks, fruit, and salads (Vyth et al., 2011). In recent years, nudging has also become a strategy of interest that can be applied in workplace cafeterias in order to promote healthy eating. A nudge is defined by Thaler and Sunstein as “any aspect of the choice architecture that alters people’s behavior in a predictable way without forbidding any options or significantly changing their economic incentives” (Thaler & Sunstein, 2008). Another term used is choice architecture. By altering the eating environment in workplace cafeterias, people can be encouraged to choose the healthy option. This is in line with the recommendation of the WHO that defines the workplace as a priority setting for health promotion (WHO, 2010). A healthy workforce is considered essential for future success. Nudging in the workplace cafeteria targets the aforementioned contextual factors of the workplace cafeteria. A study by Vermeer et al. (2011) showed that simply offering an option to choose a smaller size of a hot meal resulted in 10.2% of consumers replacing the large meal with the small meal (Vermeer, Steenhuis, Leeuwis, Heymans, & Seidell, 2011). However, possible compensation behavior (i.e., buying more products than usual due to having chosen the small meal) should also be taken into account. Van Kleef et al. (2012) demonstrated a positive effect of an increased availability of healthy snacks at the cash desk on sales in a hospital cafeteria (van Kleef, Otten, & van Trijp, 2012). Velema et al. (2018) applied a combination of several nudging strategies in the workplace cafeteria, together with some pricing

Food choices in context

157

strategies. A randomized controlled trial showed that some strategies were effective in promoting healthier choices, and that these effects remained over a 12-week period (Velema, Vyth, Hoekstra, & Steenhuis, 2018). Workplace cafeterias are not the only relevant location within the work setting. Vending machines and food and beverages offered at meetings and conferences are also important (Gardner, Whitsel, Thorndike, et al., 2014). Beyond the workplace cafeteria, other places where people go out are, for example, coffee shops, lunch rooms, and fast food and take-a-way outlets. The number of these food outlets increased hand in hand with visits to these settings. In the United States, 82% of adults eat out at least once per week (US Department of Agriculture ERS, 2010). In all-you-can-eat restaurants, people pay a fixed amount, and are responsible for the type and amount of food they serve themselves. Both the variety at these restaurants and the fixed-pricing strategy steer people’s food choices in these places ( Just & Wansink, 2008; Rolls et al., 1981). The considerable meal variety means people might experience more hedonic hunger, and are prompted to serve themselves with larger quantities (Rolls et al., 1981). Because most buffet-style restaurants have fixedprice offers, most visitors are motivated by the desire to get their money’s worth, and consume as much as possible. Consequently, the more people pay for their all-youcan-eat deal, the more they consume ( Just & Wansink, 2008). In restaurants, the impact of menu labeling on food choices is a frequently studied factor globally. In the United States, the US Congress passed Section 4205 of the Patient Protection and Affordable Care Act, “Nutrition labeling of standard menu items at chain restaurants,” which requires restaurants with 20 or more locations to provide nutritional information for standard items on menus (Public Law United States, 2010). The US Food and Drug Administration began enforcing the final rule in May 2018 (US Food and Drug Administration, 2018). To date, many global chain restaurants have implemented menu labeling, including fast-food restaurants, bakeries, coffee shops, ice cream shops, and movie theaters. Several mechanisms of the influence of menu labeling on food choices have been hypothetically suggested. First, menu labeling could motivate consumers to select healthier food options, and ultimately lower their overall energy intake. Second, it could teach visitors over time that the majority of restaurant meals have a high energy content, and subsequently steer consumers to visit restaurants less often; or third, to choose the restaurant that serves healthier meals more often. Fourth, it may stimulate consumers to compensate for their restaurant intake (e.g., by reducing the energy intake of other meals during the day), and finally, menu labels may motivate the restaurant industry to reformulate meals so that they contain less energy (VanEpps, Roberto, Park, Economos, & Bleich, 2016). Currently, a number of meta-analyses and review studies have been conducted to examine the effectiveness of menu labeling. The evidence to date is, however, mixed (Littlewood, Lourenc¸o, Iversen, & Hansen, 2016; Long, Tobias, Cradock, Batchelder, & Gortmaker, 2015; Sacco, Lillico, Chen, & Hobin, 2017; VanEpps et al., 2016). For example, a meta-analysis including 19 intervention studies found an overall small, but significant, reduction in calories (18.1 kcal, 95%CI ¼  33.56 to 2.70) ordered/purchased per meal associated with menu labeling. There was, however, considerable heterogeneity in settings across the 19 studies. Subsequent

158

Context

analyses that only included studies conducted in restaurants under control conditions showed a nonsignificant effect of menu-labeling on calories purchased (7.6 kcal, 95%CI ¼  21.02 to 5.76) (Long et al., 2015). Evidence to date suggests little effect of menu labeling on traditional fast-food purchases, but it may be encouraging lowercalorie purchases for some people in some contexts (VanEpps et al., 2016). Highquality real life studies are needed in the future.

7.5.4 Digital and online context An important context that has arisen in the past decade, and which will grow even more in the upcoming years, is the digital and online context. The presence and use of televisions, computers, tablets, and smartphones has increased over the past few decades. On one hand, the digital context facilitates the marketing of foods and, in this way, steers food preferences and food choices. Online and digital channels facilitate digital marketing, and the possibility to purchase online groceries or order meals. Digital marketing includes promotional activities via websites, social networking channels (e.g., Facebook), e-mail, online games, smartphone applications, or mobile phone text messaging (Kelly, Vandevijvere, Freeman, & Jenkin, 2015). Digital marketing is available 24/7, is interactive, and ubiquitous (Spero & Stone, 2004). Adolescents and young adults have a strong online presence, and growing purchase power (Freeman, Kelly, Vandevijvere, & Baur, 2015). A recent review, including a metaanalysis, showed that children exposed to digital marketing of unhealthy products had a slightly, but significantly higher, risk of selecting the online promoted foods or beverages (relative risk ¼ 1.1, 95%CI ¼ 1.0–1.2). Moreover, they significantly increased their intake of these products during or shortly after exposure to the online advertisements (mean differences 30.4 kcal (95%-CI ¼ 2.9–57.9) (Sadeghirad, Duhaney, Motaghipisheh, Campbell, & Johnston, 2016). On the other hand, the upcoming market of online food shopping and meal ordering is creating an intangible virtual setting in which food choices can take place. Meal kit delivery services with recipes and preportioned ingredients (such as Blue Apron, Plated, or Hello Fresh) or farm-to-table boxes (FarmFreshToYou) have also been introduced. Online supermarkets and the restaurant and takeaway delivery sector are becoming more and more important for individual food choices. Although still quite small, these are on the rise around the world, with an annual growth rate of 14% over the past 5 years (Anesbury, Nenycz-Thiel, Dawes, & Kennedy, 2016; Halzack, 2015). In the United States, 31% of consumers were likely to purchase groceries online in 2017. Moreover, the total of US online grocery sales was about $14.2 billion in 2017, and is expected to increase to nearly $30 billion by 2021 (Statista, 2017). Prior research has identified differences in shopping behavior between the online and offline environments. Studies have indicated that brand names were more important in the online than in the offline grocery context, that online consumers have higher loyalty to brands than in-store purchasers, are less sensitive to price, and preferred buying larger pack sizes (Andrews & Currim, 2004; Chu, ArceUrriza, Cebollada-Calvo, & Chintagunta, 2010; Degeratu, Rangaswamy, & Wu, 2000).

Food choices in context

159

In addition to grocery shopping, the meal delivery sector is also increasing in importance. In the past, food delivery services were limited to some local take-away stores, or to pizza or Asian outlets, and were obtained via phone orders. There has been an increase in instant food delivery initiatives that extend consumers’ options to a larger range of products, including premium restaurants. Nowadays, the online ordering of take-aways and meals has become more and more popular in Europe and the United States. In 2010, approximately 1.39 billion phone delivery orders were placed in the US. By May 2015, that number had dropped to about 1.02 billion. In the same period, online orders more than doubled, from approximately 403 million to nearly 904 million. This was facilitated by online food ordering platforms (“digital marketplaces”) that provide online ordering opportunities to restaurants (e.g., takeaway. com). A more recent development is the instant delivery service platform that enables restaurants and take-outs to offer delivery without employing their own drivers, and includes various channels for food orders (e.g., UberEats, Deliveroo) (Dablanc et al., 2017). In the years ahead, research should assess how and what features of these online settings have an impact on food choices.

7.6

Closing paragraph and future directions

In this chapter on food choice in context, we provided an insight into factors in different food contextual layers that steer food choices, varying from macro contexts to everyday settings in which people make food choices. At each contextual level (Fig. 7.1), we provided an overview of contextual factors that affect people’s food choices. However, one should still be aware that the list of examples provided in this chapter is not definitive, and more factors within each contextual layer may impact food choices. In addition, all contexts are connected and interrelated, illustrating the extensiveness and complexity of contextual influences on individual food choices. It is a challenge for researchers to understand the sum and interactive effect of contextual factors of all these levels on eating behavior (Swinburn et al., 2011), or account for each contextual influence during the day, or over the life course. Currently, researchers in the field are challenged to understand the combined effect of several contexts on food choices. System thinking in the field of food choices reinforces the influences of both interrelated and interdependent factors in a system, consisting of multiple contexts and factors (Glass & McAtee, 2006; Mabry, Olster, Morgan, & Abrams, 2008). From this perspective, food choices are more than the sum of contextual affects, given that changes in one context affect other contexts, and the whole system as well.

Acknowledgments The contribution by the first author (M.P.) is supported by the Innovational Research Incentives Scheme (#451-16-029), financed by the Netherlands Organization for Scientific Research (NWO).

160

Context

References Adam, A., & Jensen, J. D. (2016). What is the effectiveness of obesity related interventions at retail grocery stores and supermarkets? A systematic review. BMC Public Health, 16(1), 1247. https://doi.org/10.1186/s12889-016-3985-x. Afshin, A., Penalvo, J., Del Gobbo, L., et al. (2015). CVD prevention through policy: A review of mass media, food/menu labeling, taxation/subsidies, built environment, school procurement, worksite wellness, and marketing standards to improve diet. Current Cardiology Reports, 17(11), 98. https://doi.org/10.1007/s11886-015-0658-9. Afshin, A., Pen˜alvo, J. L., Del Gobbo, L., et al. (2017). The prospective impact of food pricing on improving dietary consumption: A systematic review and meta-analysis. PLoS One, 12 (3), e0172277. https://doi.org/10.1371/journal.pone.0172277. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T. Allen, J. O., Griffith, D. M., & Gaines, H. C. (2013). “She looks out for the meals, period”: African American men’s perceptions of how their wives influence their eating behavior and dietary health. Health Psychology, 32(4), 447–455. https://doi.org/10.1037/a0028361. Anand, P., & Gray, A. (2009). Obesity as market failure: Could a “deliberative economy” overcome the problems of paternalism? Kyklos, 62(2), 182–190. https://doi.org/10.1111/ j.1467-6435.2009.00430.x. Andrews, R. L., & Currim, I. S. (2004). Behavioural differences between consumers attracted to shopping online versus traditional supermarkets: Implications for enterprise design and marketing strategy. International Journal of Internet Marketing and Advertising, 1(1), 38. https://doi.org/10.1504/IJIMA.2004.003689. Anesbury, Z., Nenycz-Thiel, M., Dawes, J., & Kennedy, R. (2016). How do shoppers behave online? An observational study of online grocery shopping. Journal of Consumer Behaviour, 15, 261–270. https://doi.org/10.1002/cb.1566. Ball, K., McNaughton, S. A., Le, H. N., et al. (2015). Influence of price discounts and skillbuilding strategies on purchase and consumption of healthy food and beverages: Outcomes of the supermarket healthy eating for life randomized controlled trial. The American Journal of Clinical Nutrition, 101(5), 1055–1064. https://doi.org/10.3945/ajcn.114.096735. Bandura, A. (1989). Human agency in social cognitive theory. The American Psychologist, 44 (9), 1175–1184. https://doi.org/10.1037/0003-066x.44.9.1175. Barnett, E., & Casper, M. (2001). A definition of “social environment” American Journal of Public Health, 91(3), 465. http://www.ncbi.nlm.nih.gov/pubmed/11249033. Accessed 5 March 2018. Batis, C., Rivera, J. A., Popkin, B. M., & Taillie, L. S. (2016). First-year evaluation of Mexico’s tax on nonessential energy-dense foods: An observational study. PLOS Medicine, 13(7), e1002057. https://doi.org/10.1371/journal.pmed.1002057. Berry, S. L., Beatty, W. W., & Klesges, R. C. (1985). Sensory and social influences on ice cream consumption by males and females in a laboratory setting. Appetite, 6(1), 41–45. https:// doi.org/10.1016/S0195-6663(85)80049-0. Bevelander, K. E., Ansch€u, D. J., & Engels, R. C. M. E. (2011). Social modeling of food purchases at supermarkets in teenage girls. Appetite, 57, 99–104. https://doi.org/10.1016/j. appet.2011.04.005. Bleich, S. N., Cutler, D., Murray, C., & Adams, A. (2008). Why is the developed world obese? Annual Review of Public Health, 29(1), 273–295. https://doi.org/10.1146/annurev. publhealth.29.020907.090954.

Food choices in context

161

Brug, J., Kremers, S. P., van Lenthe, F., Ball, K., & Crawford, D. (2008). Environmental determinants of healthy eating: In need of theory and evidence. The Proceedings of the Nutrition Society, 67(3), 307–316. https://doi.org/10.1017/S0029665108008616. Caballero, B. (2007). The global epidemic of obesity: An overview. Epidemiologic Reviews, 29 (1), 1–5. https://doi.org/10.1093/epirev/mxm012. Campbell, K. J., Crawford, D. A., Salmon, J., Carver, A., Garnett, S. P., & Baur, L. A. (2007). Associations between the home food environment and obesity-promoting eating behaviors in adolescence*. Obesity, 15(3), 719–730. https://doi.org/10.1038/oby.2007.553. Caspi, C. E., Sorensen, G., Subramanian, S. V., & Kawachi, I. (2012a). The local food environment and diet: A systematic review. Health & Place, 18(5), 1172–1187. https://doi.org/ 10.1016/J.HEALTHPLACE.2012.05.006. Caspi, C. E., Sorensen, G., Subramanian, S. V., & Kawachi, I. (2012b). The local food environment and diet: A systematic review. Health & Place, 18(5), 1172–1187. https://doi.org/ 10.1016/J.HEALTHPLACE.2012.05.006. Chaix, B., M
eline, J., Duncan, S., et al. (2013). GPS tracking in neighborhood and health studies: A step forward for environmental exposure assessment, a step backward for causal inference? Health & Place, 21, 46–51. https://doi.org/10.1016/J.HEALTHPLACE. 2013.01.003. Chandon, P., Hutchinson, J. W., Bradlow, E. T., & Young, S. H. (2009). Does in-store marketing work? Effects of the number and position of shelf facings on brand attention and evaluation at the point of purchase. Journal of Marketing, 73(6), 1–17. https://doi.org/10.1509/ jmkg.73.6.1. Chandon, P., & Wansink, B. (2002). When are stockpiled products consumed faster? A convenience–salience framework of postpurchase consumption incidence and quantity. Journal of Marketing Research, 39(3), 321–335. https://doi.org/10.1509/jmkr.39.3.321.19111. Chu, J., Arce-Urriza, M., Cebollada-Calvo, J.-J., & Chintagunta, P. K. (2010). An empirical analysis of shopping behavior across online and offline channels for grocery products: The moderating effects of household and product characteristics. Journal of Interactive Marketing, 24(4), 251–268. https://doi.org/10.1016/J.INTMAR.2010.07.004. Cialdini, R. B., Reno, R. R., & Kallgren, C. A. (1990). A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology, 58(6), 1015–1026. http://psycnet.apa.org/buy/1990-30919-001. Accessed 8 March 2018. Cohen, D. A., & Babey, S. H. (2012). Contextual influences on eating behaviours: Heuristic processing and dietary choices. Obesity Reviews, 13(9), 766–779. https://doi.org/ 10.1111/j.1467-789X.2012.01001.x. Cohen, D. A., Sturm, R., Scott, M., Farley, T. A., & Bluthenthal, R. (2010). Not enough fruit and vegetables or too many cookies, candies, salty snacks, and soft drinks? Public Health Reports, 125(1), 88–95. https://doi.org/10.1177/003335491012500112. Cooksey-Stowers, K., Schwartz, M. B., & Brownell, K. D. (2017). Food swamps predict obesity rates better than food deserts in the United States. International Journal of Environmental Research and Public Health. 14(11)https://doi.org/10.3390/ijerph14111366. Crawford, T. W., Jilcott Pitts, S. B., McGuirt, J. T., Keyserling, T. C., & Ammerman, A. S. (2014). Conceptualizing and comparing neighborhood and activity space measures for food environment research. Health & Place, 30, 215–225. https://doi.org/10.1016/J. HEALTHPLACE.2014.09.007. Cruwys, T., Bevelander, K. E., & Hermans, R. C. J. (2015). Social modeling of eating: A review of when and why social influence affects food intake and choice. Appetite, 86, 3–18. https://doi.org/10.1016/J.APPET.2014.08.035.

162

Context

Cummins, S., Petticrew, M., Higgins, C., Findlay, A., & Sparks, L. (2005). Large scale food retailing as an intervention for diet and health: Quasi-experimental evaluation of a natural experiment. Journal of Epidemiology and Community Health, 59(12), 1035–1040. https:// doi.org/10.1136/jech.2004.029843. Dablanc, L., Morganti, E., Arvidsson, N., Woxenius, J., Browne, M., & Saidi, N. (2017). The rise of on-demand “instant deliveries” in European cities. Supply Chain Forum: An International Journal, 18(4), 203–217. https://doi.org/10.1080/16258312.2017.1375375. de Castro, J. M. (1994). Family and friends produce greater social facilitation of food intake than other companions. Physiology & Behavior, 56(3), 445–455. http://www.ncbi.nlm.nih.gov/ pubmed/7972393. Accessed 8 March 2018. de Castro, J. M., & de Castro, E. S. (1989). Spontaneous meal patterns of humans: Influence of the presence of other people. The American Journal of Clinical Nutrition, 50(2), 237–247. http://www.ncbi.nlm.nih.gov/pubmed/2756911. Accessed 8 March 2018. Degeratu, A. M., Rangaswamy, A., & Wu, J. (2000). Consumer choice behavior in online and traditional supermarkets: The effects of brand name, price, and other search attributes. International Journal of Research in Marketing, 17(1), 55–78. https://doi.org/10.1016/ S0167-8116(00)00005-7. Ding, D., Sallis, J. F., Norman, G. J., et al. (2012). Community food environment, home food environment, and fruit and vegetable intake of children and adolescents. Journal of Nutrition Education and Behavior, 44(6), 634–638. https://doi.org/10.1016/j. jneb.2010.07.003. Drewnowski, A., & Popkin, B. M. (2009). The nutrition transition: New trends in the global diet. Nutrition Reviews, 55(2), 31–43. https://doi.org/10.1111/j.1753-4887.1997.tb01593.x. Du, S. F., Wang, H. J., Zhang, B., Zhai, F. Y., & Popkin, B. M. (2014). China in the period of transition from scarcity and extensive undernutrition to emerging nutrition-related noncommunicable diseases, 1949-1992. Obesity Reviews, 15(S1), 8–15. https://doi.org/ 10.1111/obr.12122. Edelman, B., Engell, D., Bronstein, P., & Hirsch, E. (1986). Environmental effects on the intake of overweight and normal-weight men. Appetite, 7(1), 71–83. https://doi.org/10.1016/ S0195-6663(86)80043-5. Fedoroff, I. C., Polivy, J., & Herman, C. P. (1997). The effect of pre-exposure to food cues on the eating behavior of restrained and unrestrained eaters. Appetite, 28(1), 33–47. https:// doi.org/10.1006/APPE.1996.0057. Ferriday, D., & Brunstrom, J. M. (2008). How does food-cue exposure lead to larger meal sizes? The British Journal of Nutrition, 100(6), 1325. https://doi.org/10.1017/S00071145 08978296. Freeman, B., Kelly, B., Vandevijvere, S., & Baur, L. (2015). Young adults: beloved by food and drink marketers and forgotten by public health?: Fig. 1. Health Promotion International. 31(4)dav081https://doi.org/10.1093/heapro/dav081. Gamba, R. J., Schuchter, J., Rutt, C., & Seto, E. Y. W. (2015). Measuring the Food environment and its effects on obesity in the United States: A systematic review of methods and results. Journal of Community Health, 40(3), 464–475. https://doi.org/10.1007/s10900-014-9958-z. Gardner, C. D., Whitsel, L. P., Thorndike, A. N., et al. (2014). Food-and-beverage environment and procurement policies for healthier work environments. Nutrition Reviews, 72(6), 390–410. https://doi.org/10.1111/nure.12116. Gheza´n, G., Mateos, M., & Viteri, L. (2002). Impact of supermarkets and fast-food chains on horticulture supply chains in Argentina. Development and Policy Review, 20(4), 389–408. https://doi.org/10.1111/1467-7679.00179.

Food choices in context

163

Glanz, K., Sallis, J. F., Saelens, B. E., & Frank, L. D. (2005). Healthy nutrition environments: Concepts and measures. American Journal of Health Promotion, 19(5), 330–333. https:// doi.org/10.4278/0890-1171-19.5.330. Glass, T. A., & McAtee, M. J. (2006). Behavioral science at the crossroads in public health: Extending horizons, envisioning the future. Social Science & Medicine, 62(7), 1650–1671. https://doi.org/10.1016/J.SOCSCIMED.2005.08.044. Gordon-Larsen, P. (2014). Food availability/convenience and obesity. Advances in Nutrition, 5 (6), 809–817. https://doi.org/10.3945/an.114.007070. Gordon-Larsen, P., Wang, H., & Popkin, B. M. (2014). Overweight dynamics in Chinese children and adults. Obesity Reviews, 15(S1), 37–48. https://doi.org/10.1111/obr.12121. Halzack, S. (2015). The staggering challenges of the online grocery business. The Washington Post.https://www.washingtonpost.com/news/the-switch/wp/2015/01/20/the-staggeringchallenges-of-the-online-grocery-business/?utm_term¼.4cdd467a1f79. Accessed 11 February 2018. Handy, S. L., & Niemeier, D. A. (1997). Measuring accessibility: An exploration of issues and alternatives. Environment & Planning A, 29(7), 1175–1194. https://doi.org/10.1068/ a291175. Heilman, C., Lakishyk, K., & Radas, S. (2011). An empirical investigation of in-store sampling promotions. British Food Journal, 113(10), 1252–1266. https://doi.org/10.1108/000707 01111177674. Helbich, M., Schadenberg, B., Hagenauer, J., & Poelman, M. (2017). Food deserts? Healthy food access in Amsterdam. Applied Geography, 83, 1–12. https://doi.org/10.1016/J. APGEOG.2017.02.015. Herman, C. P. (2015). The social facilitation of eating. A review. Appetite, 86, 61–73. https:// doi.org/10.1016/J.APPET.2014.09.016. Hetherington, M. M., Anderson, A. S., Norton, G. N. M., & Newson, L. (2006). Situational effects on meal intake: A comparison of eating alone and eating with others. Physiology & Behavior, 88(4–5), 498–505. https://doi.org/10.1016/J.PHYSBEH.2006.04.025. Higgs, S. (2015). Social norms and their influence on eating behaviours. Appetite, 86, 38–44. https://doi.org/10.1016/J.APPET.2014.10.021. Hollands, G. J., Shemilt, I., Marteau, T. M., et al. (2014). Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco. In G. J. Hollands (Ed.), Cochrane database of systematic reviews. John Wiley & Sons, Ltd: Chichester, UKhttps:// doi.org/10.1002/14651858.CD011045. Hollands, G. J., Shemilt, I., Marteau, T. M., et al. (2015). Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco. In G. J. Hollands (Ed.), Cochrane Database of Systematic Reviews (p. CD011045). Chichester, UK: John Wiley & Sons, Ltd. https://doi.org/10.1002/14651858.CD011045.pub2. Iyengar, S. S., & Lepper, M. R. (2000). When choice is demotivating: Can one desire too much of a good thing? Journal of Personality and Social Psychology, 79(6), 995–1006. https:// doi.org/10.1037//0022-3514.79.6.995. Just, D. R., & Wansink, B. (2008). The Fixed Price Paradox: Conflicting Effects of “‘All-You-CanEat’” Pricing. Cornell University.www.agecon.purdue.edu/news/seminarfiles/MS12118.pdf. Kahneman, D. (2011). Thinking, Fast and Slow. https://books.google.nl/books?id¼ZuKTvER uPG8C&printsec¼frontcover&dq¼kahneman+thinking+fast+slow&hl¼nl&sa¼X&ved ¼0ahUKEwjLsJTd9oHYAhVJqKQKHXi1C48Q6AEIOjAD#v¼onepage&q¼kahneman%20 thinking%20fast%20slow&f¼false. Accessed 11 December 2017. Kapsiotis, G. D. (1975). An international programme for famine relief. The Proceedings of the Nutrition Society, 34(3), 195–199. https://doi.org/10.1079/PNS19750038.

164

Context

Kelly, B., Vandevijvere, S., Freeman, B., & Jenkin, G. (2015). New media but same old tricks: Food marketing to children in the digital age. Current Obesity Reports, 4(1), 37–45. https:// doi.org/10.1007/s13679-014-0128-5. Kremers, S. P., De Bruijn, G.-J., Visscher, T. L., Van Mechelen, W., De Vries, N. K., & Brug, J. (2006). Environmental influences on energy balance-related behaviors: A dual-process view. International Journal of Behavioral Nutrition and Physical Activity, 3, 9. https:// doi.org/10.1186/1479-5868-3-9. Larsen, K., & Gilliland, J. (2008). Mapping the evolution of “food deserts” in a Canadian city: Supermarket accessibility in London, Ontario, 1961–2005. International Journal of Health Geographics, 7(1), 16. https://doi.org/10.1186/1476-072X-7-16. Littlewood, J. A., Lourenc¸o, S., Iversen, C. L., & Hansen, G. L. (2016). Menu labelling is effective in reducing energy ordered and consumed: A systematic review and meta-analysis of recent studies. Public Health Nutrition, 19(12), 2106–2121. https://doi.org/10.1017/ S1368980015003468. Long, M. W., Tobias, D. K., Cradock, A. L., Batchelder, H., & Gortmaker, S. L. (2015). Systematic review and meta-analysis of the impact of restaurant menu calorie labeling. American Journal of Public Health, 105(5), e11–e24. https://doi.org/10.2105/AJPH.2015.302570. Lu, W., & Qiu, F. (2015). Do food deserts exist in Calgary, Canada? Can Geogr/Le G
eographe Can, 59(3), 267–282. https://doi.org/10.1111/cag.12176. Luan, H., Law, J., & Quick, M. (2015). Identifying food deserts and swamps based on relative healthy food access: A spatio-temporal Bayesian approach. International Journal of Health Geographics, 14(1), 37. https://doi.org/10.1186/s12942-015-0030-8. Luiten, C. M., Steenhuis, I. H., Eyles, H., Ni Mhurchu, C., & Waterlander, W. E. (2016). Ultraprocessed foods have the worst nutrient profile, yet they are the most available packaged products in a sample of New Zealand supermarkets. Public Health Nutrition, 19(3), 530–538. https://doi.org/10.1017/S1368980015002177. Lytle, L. A., & Sokol, R. L. (2017). Measures of the food environment: A systematic review of the field, 2007–2015. Health & Place, 44, 18–34. https://doi.org/10.1016/j.healthplace .2016.12.007. Mabry, P. L., Olster, D. H., Morgan, G. D., & Abrams, D. B. (2008). Interdisciplinarity and systems science to improve population health: A view from the NIH Office of Behavioral and Social Sciences Research. American Journal of Preventive Medicine, 35(2 Suppl), S211–S224. https://doi.org/10.1016/j.amepre.2008.05.018. Monteiro, C. A., Moubarac, J.-C., Cannon, G., Ng, S. W., & Popkin, B. (2013). Ultra-processed products are becoming dominant in the global food system. Obesity Reviews, 14, 21–28. https://doi.org/10.1111/obr.12107. Moodie, R., Swinburn, B., Richardson, J., & Somaini, B. (2006). Childhood obesity – A sign of commercial success, but a market failure. International Journal of Pediatric Obesity, 1(3), 133–138. https://doi.org/10.1080/17477160600845044. Nijs, K. A. N. D., de Graaf, C., Siebelink, E., et al. (2006). Effect of family-style meals on energy intake and risk of malnutrition in Dutch nursing home residents: A randomized controlled trial. Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 61(9), 935–942. https://doi.org/10.1093/gerona/61.9.935. Osorio, A. E., Corradini, M. G., & Williams, J. D. (2013). Remediating food deserts, food swamps, and food brownfields: Helping the poor access nutritious, safe, and affordable food. AMS Review, 3(4), 217–231. https://doi.org/10.1007/s13162-013-0049-6. Ploeg, M., Breneman, V., Farrigan, T., et al. (2009). Access to affordable and nutritious food: measuring and understanding food deserts and their consequences. Report to Congress. In Access to Afford Nutr food Meas Underst food deserts their consequences Rep to Congr. https://www.cabdirect.org/cabdirect/abstract/20113168727. Accessed 16 March 2018.

Food choices in context

165

Poelman, M. P., De Vet, E., Velema, E., Seidell, J. C., & Steenhuis, I. H. M. (2015). The home food environment of overweight gatekeepers in the Netherlands. Public Health Nutrition. 18(10). https://doi.org/10.1017/S1368980014002298. Poelman, M. P., Eyles, H., Dunford, E., et al. (2016). Package size and manufacturerrecommended serving size of sweet beverages: a cross-sectional study across four highincome countries. Public Health Nutrition, 19(6), 1008–1016. https://doi.org/10.1017/ S1368980015001974. Poelman, M., Kroeze, W., Waterlander, W., de Boer, M., & Steenhuis, I. (2017). Food taxes and calories purchased in the virtual supermarket: A preliminary study. British Food Journal, 119(12), 2559–2570. https://doi.org/10.1108/BFJ-08-2016-0386. Popkin, B. M. (1993). Nutritional patterns and transitions. Population and Development Review, 19(1), 138. https://doi.org/10.2307/2938388. Popkin, B. M., Adair, L. S., & Ng, S. W. (2012). Global nutrition transition and the pandemic of obesity in developing countries. Nutrition Reviews, 70(1), 3–21. https://doi.org/10.1111/ j.1753-4887.2011.00456.x. Pratt, I. S., Croager, E. J., & Rosenberg, M. (2012). The mathematical relationship between dishware size and portion size. Appetite, 58(1), 299–302. https://doi.org/10.1016/j. appet.2011.10.010. Prinsen, S., De Ridder, D. T. D., & De Vet, E. (2013). Eating by example. Effects of environmental cues on dietary decisions. Appetite, 70, 1–5. https://doi.org/10.1016/j. appet.2013.05.023. Public Law United States (2010). The Patient Protection and Affordable Care Act - Sec. 4205: (pp. 759–762). 111(48) http://legcounsel.house.gov/Comps/PatientProtectionAndAfforda bleCareAct.pdf. Ravensbergen, E. A., Waterlander, W. E., Kroeze, W., & Steenhuis, I. H. (2015). Healthy or unhealthy on Sale? A cross-sectional study on the proportion of healthy and unhealthy foods promoted through flyer advertising by supermarkets in the Netherlands. BMC Public Health, 15(1), 470. https://doi.org/10.1186/s12889-015-1748-8. Raynor, H. A., & Wing, R. R. (2007). Package unit size and amount of food: Do both influence intake?*. Obesity, 15(9), 2311–2319. https://doi.org/10.1038/oby.2007.274. Reardon, T., & Berdegu
e, J. A. (2002). The rapid rise of supermarkets in Latin America: Challenges and opportunities for development. Development and Policy Review, 20(4), 371–388. https://doi.org/10.1111/1467-7679.00178. Reno, R. R., Cialdini, R. B., & Kallgren, C. A. (1993). The transsituational influence of social norms. Journal of Personality and Social Psychology, 64(1), 104–112. https://insights.ovid. com/personality-social-psychology/jpspy/1993/01/000/transsituational-influence-social-norm s/10/00005205. Accessed 8 March 2018. Robinson, E., Nolan, S., Tudur-Smith, C., et al. (2014). Will smaller plates lead to smaller waists? A systematic review and meta-analysis of the effect that experimental manipulation of dishware size has on energy consumption. Obesity Reviews, 15(10), 812–821. https://doi.org/10.1111/obr.12200. Robinson, E., Thomas, J., Aveyard, P., & Higgs, S. (2014). What everyone else is eating: A systematic review and meta-analysis of the effect of informational eating norms on eating behavior. Journal of the Academy of Nutrition and Dietetics, 114(3), 414–429. https:// doi.org/10.1016/j.jand.2013.11.009. Rolls, B. J., Rowe, E. A., Rolls, E. T., Kingston, B., Megson, A., & Gunary, R. (1981). Variety in a meal enhances food intake in man. Physiology & Behavior, 26(2), 215–221. https://doi. org/10.1016/0031-9384(81)90014-7. Rosenkranz, R. R., & Dzewaltowski, D. A. (2008). Model of the home food environment pertaining to childhood obesity. Nutrition Reviews, 66(3), 123–140. https://solisvpn.uu.

166

Context

nl/+CSCO+0h75676763663A2F2F76616676747567662E626976712E70627A++/nutrition -reviews/nutrev/2008/03/000/model-home-food-environment-pertaining-childhood/2/01434 664. Accessed 3 February 2018. Sacco, J., Lillico, H. G., Chen, E., & Hobin, E. (2017). The influence of menu labelling on food choices among children and adolescents: A systematic review of the literature. Perspectives in Public Health, 137(3), 173–181. https://doi.org/10.1177/1757913916658498. Sadeghirad, B., Duhaney, T., Motaghipisheh, S., Campbell, N. R. C., & Johnston, B. C. (2016). Influence of unhealthy food and beverage marketing on children’s dietary intake and preference: A systematic review and meta-analysis of randomized trials. Obesity Reviews, 17 (10), 945–959. https://doi.org/10.1111/obr.12445. Sallis, J. F., Owen, N., & Fisher, E. B. (2015). Ecological models of health behavior. In Health behavior: Theory, research, and practice (pp. 43–64). San Francisco, CA, USA: JosseyBass. Salvy, S.-J., & Pliner, P. P. (2010). Social influences on eating in children and adults. In Obesity Prevention (pp. 617–627). https://doi.org/10.1016/B978-0-12-374387-9.00050-7. Sela, A., Berger, J., & Liu, W. (2009). Variety, vice, and virtue: How assortment size influences option choice. Journal of Consumer Research, 35(6), 941–951. https://doi.org/10.1086/593692. Sheeran, P., & Webb, T. L. (2016). “The intention-behavior gap.” Social and personality psychology compass 10.9, 503–518. https://onlinelibrary.wiley.com/doi/full/10.1111/spc3. 12265. Slater, J., Green, C. G., Sevenhuysen, G., Edginton, B., O’Neil, J., & Heasman, M. (2009). The growing Canadian energy gap: More the can than the couch? Public Health Nutrition, 12 (11), 2216. https://doi.org/10.1017/S1368980009990309. Smoyer-Tomic, K. E., Spence, J. C., & Amrhein, C. (2006). Food deserts in the prairies? Supermarket accessibility and neighborhood need in Edmonton, Canada*. The Professional Geographer, 58(3), 307–326. https://doi.org/10.1111/j.1467-9272.2006.00570.x. Sobal, J., & Wansink, B. (2007). Kitchenscapes, tablescapes, platescapes, and foodscapes. Environment and Behavior, 39(1), 124–142. https://doi.org/10.1177/0013916506295574. Spero, I., & Stone, M. (2004). Agents of change: How young consumers are changing the world of marketing. Quality Market Research: An International Journal, 7(2), 153–159. https:// doi.org/10.1108/13522750410530057. Statista (2017). Consumer grocery shopping methods U.S. 2017. Statistic.https://www.statista. com/statistics/292933/us-shoppers–primary-food-shopping-locations-by-type. Accessed 11 February 2018. Steenhuis, I. H., Leeuwis, F. H., & Vermeer, W. M. (2017). Small, medium, large or supersize: Trends in food portion sizes in the Netherlands. Public Health Nutrition, 13(6), 852–857. https://doi.org/10.1017/S1368980009992011. Steenhuis, I., & Poelman, M. (2017). Portion size: Latest developments and interventions. Current Obesity Reports, 6(1), 10–17. https://doi.org/10.1007/s13679-017-0239-x. Steenhuis, I. H., Waterlander, W. E., & de Mul, A. (2011). Consumer food choices: The role of price and pricing strategies. Public Health Nutrition, 14(12), 2220–2226. https://doi.org/ 10.1017/S1368980011001637. Stern, D., Ng, S. W., & Popkin, B. M. (2016). The nutrient content of U.S. household food purchases by store type. American Journal of Preventive Medicine, 50(2), 180–190. https:// doi.org/10.1016/j.amepre.2015.07.025. Stok, F. M., de Vet, E., de Ridder, D. T. D., & de Wit, J. B. F. (2016). The potential of peer social norms to shape food intake in adolescents and young adults: A systematic review of effects and moderators. Health Psychology Review, 10(3), 326–340. https://doi.org/ 10.1080/17437199.2016.1155161.

Food choices in context

167

Swinburn, B., Egger, G., & Raza, F. (1999). Dissecting obesogenic environments: The development and application of a framework for identifying and prioritizing environmental interventions for obesity. Preventive Medicine (Baltimore), 29(6), 563–570. https://doi. org/10.1006/PMED.1999.0585. Swinburn, B. A., Sacks, G., Hall, K. D., et al. (2011). The global obesity pandemic: Shaped by global drivers and local environments. Lancet (London, England), 378(9793), 804–814. https://doi.org/10.1016/S0140-6736(11)60813-1. Swinburn, B. A., Sacks, G., Lo, S. K., et al. (2009). Estimating the changes in energy flux that characterize the rise in obesity prevalence. The American Journal of Clinical Nutrition, 89 (6), 1723–1728. https://doi.org/10.3945/ajcn.2008.27061. Thaler, R. H., & Sunstein, C. (2008). Nudge; improving decisions about health, wealth, and happiness. New Haven & London: Yale University Press. Thornton, L. E., Pearce, J. R., & Kavanagh, A. M. (2011). Using geographic information systems (GIS) to assess the role of the built environment in influencing obesity: A glossary. International Journal of Behavioral Nutrition and Physical Activity, 8(1), 71. https://doi. org/10.1186/1479-5868-8-71. Tillotson, J. E. (2004). AMERICA’S OBESITY: Conflicting public policies, industrial economic development, and unintended human consequences. Annual Review of Nutrition, 24(1), 617–643. https://doi.org/10.1146/annurev.nutr.24.012003.132434. Timmermans, J., Dijkstra, C., Kamphuis, C., Huitink, M., van der Zee, E., & Poelman, M. (2018). “Obesogenic” school food environments? An urban case study in the Netherlands. International Journal of Environmental Research and Public Health, 15(4), 619. https:// doi.org/10.3390/ijerph15040619. Trail, W. B. (2017). TRADE POLICY TECHNICAL NOTES, Trade and Food security – transnational corporations, food systems and their impacts on diets in developing countries. http://www.fao.org/3/a-i8192e.pdf. Accessed 18 May 2018. US Department of Agriculture ERS (2010). Consumer. Diet quality and food consumption: flexible consumer behavior survey. http://www.ers.usda.gov/Briefing/%0ADietQuality/flexi ble.htm%0A. US Food and Drug Administration (2018). Changes to the Nutrition Facts Label. https://www. fda.gov/Food/GuidanceRegulation/GuidanceDocumentsRegulatoryInformation/Labeling Nutrition/ucm385663.htm. Accessed 16 March 2018. van Kleef, E., Otten, K., & van Trijp, H. C. (2012). Healthy snacks at the checkout counter: A lab and field study on the impact of shelf arrangement and assortment structure on consumer choices. BMC Public Health, 12(1), 1072. https://doi.org/10.1186/1471-2458-12-1072. VanEpps, E. M., Roberto, C. A., Park, S., Economos, C. D., & Bleich, S. N. (2016). Restaurant menu labeling policy: Review of evidence and controversies. Current Obesity Reports, 5 (1), 72–80. https://doi.org/10.1007/s13679-016-0193-z. Velema, E., Vyth, E. L., Hoekstra, T., & Steenhuis, I. H. (2018). Nudging and social marketing techniques encourage employees to make healthier food choices: A randomized controlled trial in 30 worksite cafeterias in the Netherlands. The American Journal of Clinical Nutrition, 107(2), 236–246. https://doi.org/10.1093/ajcn/nqx045. Velema, E., Vyth, E. L., & Steenhuis, I. H. M. (2019). “I have worked so hard that I deserve a snack in the worksite cafeteria”: A focus group study. Appetite, 133, 297–304. Velema, E., Vyth, E. L., & Steenhuis, I. H. M. (2017). Using nudging and social marketing techniques to create healthy worksite cafeterias in the Netherlands: Intervention development and study design. BMC Public Health, 17(1), 63. https://doi.org/10.1186/s12889-0163927-7.

168

Context

Vermeer, W. M., Alting, E., Steenhuis, I. H. M., & Seidell, J. C. (2010). Value for money or making the healthy choice: The impact of proportional pricing on consumers’ portion size choices. European Journal of Public Health, 20(1), 65–69. https://doi.org/10.1093/eurpub/ ckp092. Vermeer, W. M., Steenhuis, I. H. M., Leeuwis, F. H., Heymans, M. W., & Seidell, J. C. (2011). Small portion sizes in worksite cafeterias: Do they help consumers to reduce their food intake? International Journal of Obesity, 35(9), 1200–1207. https://doi.org/10.1038/ijo.2010.271. Vyth, E. L., Steenhuis, I. H. M., Heymans, M. W., Roodenburg, A. J. C., Brug, J., & Seidell, J. C. (2011). Influence of placement of a nutrition logo on cafeteria menu items on lunchtime food choices at Dutch work sites. Journal of the American Dietetic Association, 111(1), 131–136. https://doi.org/10.1016/j.jada.2010.10.003. Wallinga, D. (2010). Agricultural policy and childhood obesity: A food systems and public health commentary. Health Affairs, 29(3), 405–410. https://doi.org/10.1377/hlthaff.2010.0102. Waterlander, W. E., de Boer, M. R., Schuit, A. J., Seidell, J. C., & Steenhuis, I. H. (2013). Price discounts significantly enhance fruit and vegetable purchases when combined with nutrition education: A randomized controlled supermarket trial. The American Journal of Clinical Nutrition, 97(4), 886–895. https://doi.org/10.3945/ajcn.112.041632. Waterlander, W. E., Jiang, Y., Steenhuis, I. H. M., & Ni, M. C. (2015). Using a 3D virtual supermarket to measure food purchase behavior: A validation study. Journal of Medical Internet Research. 17(4)e107https://doi.org/10.2196/jmir.3774. Waterlander, W. E., Ni Mhurchu, C., & Steenhuis, I. H. M. (2014). Effects of a price increase on purchases of sugar sweetened beverages. Results from a randomized controlled trial. Appetite, 78, 32–39. https://doi.org/10.1016/J.APPET.2014.03.012. Weatherspoon, D. D., & Reardon, T. (2003). The rise of supermarkets in Africa: Implications for Agrifood systems and the rural poor. Development and Policy Review, 21(3), 333–355. https://doi.org/10.1111/1467-7679.00214. WHO (2010). Workplace health promotion. WHO.http://www.who.int/occupational_health/ topics/workplace/en. Accessed 5 March 2018. Wrigley, N., Warm, D., & Margetts, B. (2003). Deprivation, diet, and food-retail access: Findings from the Leeds “food deserts” study. Environment & Planning A, 35(1), 151–188. https://doi.org/10.1068/a35150.

Further reading StatLine (2016). https://opendata.cbs.nl/statline/#/CBS/nl/dataset/83749NED/table?ts¼15155 03221345. Accessed 9 January 2018.

Meal and snack: Two different contexts for foods and drinks

8

Uyen Thuy Xuan Phan Institute of Biotechnology and Food Technology, Industrial University of Ho Chi Minh City, Ho Chi Minh City, Vietnam

8.1

Defining meals and snacks

The situational nature of eating and drinking has been a topic of considerable interest in the domain of food choice studies (e.g., Bisogni et al., 2007; Blake, Bisogni, Sobal, Devine, & Jastran, 2007) due to foods being often consumed as part of a certain eating event. These events often come in the form of either a meal or a snack. The essence of meals has been the focus of a number of scientific studies that have explored their many different dimensions, including psychological (Pliner & Rozin, 2000), nutritional (de Graaf, 2000), social (de Castro, Brewer, Elmore, & Orozco, 1990; Douglas, 1972; Herman, 2015; Higgs, 2015; Sobal, 2000), historical (Walker, 2002), biological (Lawless, 2000; Prescott & Logan, 2017), and cultural aspects (Chiva, 1997; Fjellstr€ om, 2004; Kjærnes, Holm, Gronow, M€akel€a, & Ekstr€om, 2009; Ochs & Shohet, 2006). However, no matter which dimension of the meal we look at, the necessity of meals in our daily diets is revealed. Snacks, on the other hand, have recently gained much attention due to their significant contribution to people’s energy intake throughout the day, and their increased consumption over the past several decades (Bellisle, 2014). This “snacking” trend has been observed in many European countries (Bellisle, Monneuse, Steptoe, & Wardle, 1995; Kerr et al., 2009; Ovaskainen, Tapanainen, & Pakkala, 2010), as well as in North (Kant & Graubard, 2015) and South America (Duffey, Pereira, & Popkin, 2013). Some argue that a meal observes a structured rule set with certain accepted combinations, while a snack functions without such limitations (Douglas & Nicod, 1974; M€akel€a, 1991; M€akel€a, 2000). Bellisle (2014) presses the idea of a difference in nutrient and energy intake, as well as body energy balance between meals and snacks. Meals and snacks, thus, can be two significantly different forms for foods and drinks. This chapter compares and contrasts these two contextual frameworks of eating experience from a number of perspectives, including: dictionary definition, portion size, energy intake, food pattern, and social and motivational factors. The main findings are summarized in Table 8.1.

8.1.1 Dictionary definition Meiselman (2009), in the book entitled Meals in Science and Practice, has noted Pliner and Rozin’s observation that 17 out of 18 languages studied have a word for meals, and 16 have a word for snacks. Within the scope of this chapter, the dictionary Context. https://doi.org/10.1016/B978-0-12-814495-4.00008-8 Copyright © 2019 Elsevier Inc. All rights reserved.

170

Context

Table 8.1 Main findings from comparing meals and snacks Dimension

Meals

Portion size Energy intake Nutrients

Universally large and heavier Universally small and lighter At least 15% of total daily No specific requirement intake High in fat and animal protein High in sugar Low in carbohydrates High in carbohydrates More likely to be a hot eating More likely to be a cold eating event event Fixed time (breakfast, lunch, In between meals (mid-morning, middinner) afternoon, late-night/evening snack) No clear distinction. Both meals and snacks can happen inside and outside the home. Generally more likely to be More likely to be eaten alone eaten with others Sophisticated with a number Less sophisticated with interplaying of of interacting motivations fewer motivations

Time Place of consumption Social facilitation Associated Motivations

Snacks

definitions for meals and snacks are presented in four languages, including English, French, Chinese, and Vietnamese. The definitions of meals and snacks in English are taken from three online dictionaries: Oxford Dictionaries (https://www. oxforddictionaries.com/), Cambridge Dictionary (https://dictionary.cambridge.org/ us/), and Merriam-Webster’s Dictionary (https://www.merriam-webster.com/). The French definitions are from Larousse Gastronomique (1997, pp. 1215). The definitions of meals and snacks in Vietnamese are from an online Vietnamese dictionary at Vdict.com, and those in Chinese are from a Chinese dictionary at zdic.net. The reported definitions are the common usage. Oxford Dictionaries

Meal: (1) Any of the regular occasions in a day when a reasonably large amount of food is eaten; (2) the food eaten during a meal. Snack: (1) A small amount of food eaten between meals; (2) A light meal that is eaten in a hurry or in a casual manner. Cambridge Dictionary

Meal: an occasion when food is eaten, or the food that is eaten on such an occasion. Snack: a small amount of food that is eaten between meals, or a very small meal. Merriam-Webster

Meal: (1) an act or the time of eating a portion of food to satisfy appetite; (2) the portion of food eaten at a meal. Snack: a light meal: food eaten between regular meals; also: food suitable for snacking.

Meal and snack: Two different contexts for foods and drinks

171

All these three English dictionaries define snacks in relation to meals. The three definitions agree that snacks are not meals, are generally smaller in portion size, and happen in between meals. Definitions of meals provided in these three dictionaries all imply that meals are the main context for food consumption, and the amount of food eaten at a meal is larger than a snack. Meals are also more regular than snacks, which are more spontaneous. Larousse Gastronomique

In French, repas is the word meaning meal. Its definition in French is “nourriture consommee chaque jour a` heure fixe.” The translation into English is “food consumed every day at a fixed time.” So it can be said that meal in French means a food event that is routine and rigid in terms of time of the day. Casse-crouˆte is the French word for snacks. It is defined as “repas rapide et sommaire” in Larousse Gastronomique, meaning “fast and simple meal.” The dictionary also goes further to provide the food stuff that compose French casse-crouˆte in general, which are bread, cheese, and cold cuts of processed meats; it could also be mixed salads with pasta/rice. However, casse-crouˆte is not used as much as snacks in English. To date, other more popular words are en-cas (light meal that can be consumed immediately) or “grignotage” (food eaten in tiny pieces). Thus, snacks, in French, are defined in relation to meals in terms of simplicity/complexity and the time spent eating it. Zdic.net

From zdic.net, the Chinese word for meals is 膳食 (pinyin: sha`nshı´), defined as 日常 吃的饭菜, meaning “daily food consumption.” Deconstructing this definition, we find this “broken” English sentence: 日 (day) 常 (often) 吃 (eat) 的 (of ) 饭 (rice) 菜 (dish). Literarily, meals in Chinese language mean “daily eating events that relate to rice consumption.” Snacks, on the other hand, go by the word 小吃 (pinyin: xiǎochı¯), meaning “simple and inexpensive dishes” or “simple, casual eating.” The two components of this word also have their own meaning: 小 means “small, tiny, few” and 吃 means “eat.” Thus, snacks in Chinese also imply light eating. www.vdict.com

The word meal in Vietnamese language is bữa a˘n with the word bữa roughly translated as “a routine food and drink consumption at a specific time of the day,” and a˘n means eat. Beyond this, the Vietnamese people often use the word a˘n cơm, which literarily means “eat cooked rice” to refer to “a meal.” Therefore, meals can also be called “bữa cơm.” This indicates a strong association between rice and meals (normally lunch and dinner in Vietnamese traditional meal patterns). Vietnamese breakfast, on the other hand, is not described by the word “cơm” (cooked rice). When Vietnamese people mention breakfast, they say a˘n sa´ng, meaning “eating in the morning,” instead of a˘n cơm. By name, Vietnamese breakfast is quite different from the other two main meals. In Vietnam, snacks are widely referred to by a˘n nhẹ (light eating) or a˘n vặt (junk eating). In contrast to meals, Vietnamese people do not use the word “a˘n cơm”

172

Context

(eat cooked rice) to indicate snacks, despite rice products (i.e., congee, sticky rice, rice noodles), and even cooked rice being used for snacking. In this aspect, the definition for snacks in the Vietnamese language is consistent with its English and Chinese counterparts, and is dissociated from meal qualifiers.

8.1.2 Portion size, energy intake, and food categories In line with the definition from established dictionaries, a number of studies have defined meals and snacks in terms of the amount of foods consumed. In an early study, meals were quantitatively confirmed to be larger than snacks, and foods eaten for meals contain more carbohydrates, fat, and protein (de Castro et al., 1990). However, snacks contain a higher proportion of carbohydrates to fat and protein than meals. Younginer et al. (2016), in a U.S. study (Boston, MA and Philadelphia, PA) of low-income caregivers’ definitions of snacks offered to preschool-aged children, has defined snacks as foods that are less filling than meals. The authors summarized these empirically collected definitions as follows: A small portion of food that is given in-between meals, frequently with an intention of reducing or preventing hunger until the next mealtime.

This definition appears to be consistent with the definition from dictionaries. Similar conclusions were found by Jacquier, Gatrell, and Bingley (2017) in a study of attitudes and perceptions about eating in-between meals among Swiss caregivers of young children. The Swiss participants of this study also associated snack foods with small portion sizes. Blake et al. (2015) reported similar distinctions between meals and snacks in a study of American parents’ perception of portion sizes of children’s snacks. According to Ritchie (2012), a meal is any eating episode comprised of at least 15% of the total daily energy intake, regardless of the time of day or composition of foods or beverages consumed. Ovaskainen et al. (2006) compared main meals and snacks of Finnish men and women in terms of energy intake (MJ/day), and energy density (the energy content per unit weight of food, kJ/100 g). The authors found that daytime energy intake was predominantly from main meals for both men and women, with one peak at lunch and one at dinner. At night, it is snacks that provide much of the energy. For the Finnish, meals are seen as the eating occasions that are responsible for energy intake during the day, and snacks are the main source of energy in the evening. However, the energy density (kJ/100 g) of foods and energy-containing beverages was higher in snacks than in main meals. Observations from a number of studies have shown that there is a general difference in the food categories consumed for meals and for snacks. For example, foods consumed at meals are often higher in fat and lower in carbohydrates than foods consumed as snacks for American adults (Ballard-Barbash, Thompson, Graubard, & Krebs-Smith, 1994), French adults (Bellisle et al., 2003), Brazilian people (Duffey et al., 2013), and at least three diferent age groups in a British population sample (Summerbell, Moody, Shanks, Stock, & Geissler, 1995). Meals are also richer in

Meal and snack: Two different contexts for foods and drinks

173

animal protein than snacks in American eating (Phan, 2015; Phan & Chambers, 2016a; Phan & Chambers, 2016b). Wadhera and Capaldi (2012) found that American college students categorize snacks into “milk, cheese, and yogurt” and “fruits and vegetables” groups; and fats, oils, and sweets’ groups, while foods for meals are mainly from the “breads, cereals, rice, and pasta” and “meats, poultry, fish, beans, eggs, and nuts” groups. Bellisle et al. (2003) also found that meats, fish, dairy products, cheese, and fruits are consumed most in the context of meals, while sweets, cereal bars, biscuits, and sodas are reported mostly in the context of snacks for French adults. Kjærnes et al. (2009) found that eating events in four Nordic countries, that is, Sweden, Norway, Finland, and Denmark, are categorized into “breakfast” (the first eating of the day), “cold eating events” or “hot meals.” Snacks were found to belong to the “cold eating event,” not the “hot one” in each of the studied Nordic countries. Meanwhile, despite the difference in the classification of lunch (“hot” in Finland and Sweden, “cold” in Norway and Denmark), dinner was always a hot meal. Nordic foods for dinner were often composed of a variety of meats and other sources of protein such as fish. Kjærnes et al. (2009) also identified “a snack” event by examples such as “an ice cream, an apple, or a chocolate bar with/without a beverage.” The examples from these studies support the higher occurrence of protein in meals, and a much higher sugar density in snacks (Wang, van der Horst, Jacquier, & Eldridge, 2016).

8.1.3 Time, place, and social factor The pattern of three meals per day is rigorously followed in many countries across the globe. Breakfast, lunch, and dinner are the common English words used to label those three main meals, with the exception of supper, which is often used interchangeably for dinner in most parts of the United States and Canada. Similar to meals, the time of consumption is an important component in defining snacks, using structured meal patterns as markers. In relation to timing, snacks are commonly consumed during the mid-morning, mid-afternoon, and evening (Chaplin & Smith, 2011). According to Chaplin and Smith, the best definition of snacking accepted by most of the participants in their study is “food or drink eaten between main meals.” Looking further into the time breakdown of meals and snacks, a US study performed by Phan (2015) to investigate motivations associated with meals and snacks of a group of about 300 people in Kansas (US) found the word “breakfast” was used by the majority of the respondents to indicate the eating that occurred between 7:00 and 9:00 a.m., while “lunch” was used for the meal between 12:00 and 3:00 p.m. For the evening meals occurring from 5:00 to 8:00 p.m., “dinner” and “supper” were the frequent names used. The respondents, who were university faculty, staff, employees, and students, also gave the name “mid-morning snack” to the eating event after breakfast and before lunch (9:00–11:00 a.m.); “mid-afternoon snack” to the eating event around 3:00–5:00 p.m.; and “late-night snack” or “evening snack” or “night-out” to describe any eating after 9:00 p.m. These findings are consistent with Phan and Chambers (2016a) and Duffey et al. (2013). Similar results are also reported by Toornvliet, Pijl, Hopman, Elte-De Wever, and Meinders (1996), where meals were

174

Context

usually consumed between 8:00 and 10:00 a.m., between 12:00 and 2:00 p.m., and between 6:00 and 8:00 p.m.; other food items consumed between meals were considered a snack. Time is, thus, a key distinguishing factor between meals and snacks (Kant & Graubard, 2015). In general, there is no clear distinction between meals and snacks in terms of where the eating occurs. Phan (2015) and Phan and Chambers (2016a) reported that both meals and snacks are consumed inside and outside the home, with offices (at work) and restaurants being the most common outside place for eating events to happen. However, differences are found between daytime and evening eating. Breakfast, mid-morning snacks, lunch, and mid-afternoon snacks occur both at home and at work, while dinner and evening snacks mostly take place at home. Another notable pattern for meals is that breakfast is consumed at home more than at work, while lunch is equally consumed in both places. A reversed pattern is seen for snacks, where a midmorning snack is more likely to be eaten outside the home, and a mid-afternoon snack is consumed both at home and at work. This defining (specific) of meals and snacks by place of consumption may only be valid for Americans with the same demographics of the participants in Phan (2015) and Phan and Chambers (2016a). The only exception is dinner, which universally takes place at home. Another aspect that can differentiate meals from snacks is that meals are in general more likely to be eaten with others (see Chapter 2 by Higgs), while snacks are generally eaten alone (de Castro et al., 1990). Wansink, Payne, and Shimizu (2010) confirmed that eating episodes are more likely to be viewed as meals if the person is eating with family versus eating alone. Because people tend to eat more when dining with others when compared with eating alone, social meals are characteristically larger than snacks. This social facilitation of eating has been investigated by various studies, and the findings have been reported and discussed in a systematic review by Herman (2015). de Castro et al. (1990) found a strong positive correlation between the amount eaten in the meal and the number of other people present at the meal. Herman, Roth, and Polivy (2003) clarified this social correlation, reporting that it is significant only for eating with family and friends; eating with more than one stranger tends not to increase the food intake.

8.1.4 Motivational factors Motivation should be understood as a constant flow of behavior that can be directed in many different ways (Petri & Govern, 2012). Motivation explains the reason that a certain behavior is exhibited. There are at least two approaches to explore these reasons: why an individual exhibits certain behaviors (ultimate causation) and how these behaviors came about (proximate causation) (Wong, 2000). According to Wong, an analysis of behavior in terms of ultimate causation is also regarded as a functional explanation, with the assumption that there is something to be gained by behaving that way. Meanwhile, proximate mechanisms that are shaped by natural selection provide an explanation concerning how certain activities occur. One commonly held characteristic of motivation is its activating properties (K€oster & Mojet, 2006; Wong, 2000). This is often seen in the production of behavior

Meal and snack: Two different contexts for foods and drinks

175

(overt responding), that is, when the observed organism is behaving in certain way in response to a stimulus. Organisms are believed to actively look for stimulation and try to maintain an optimal level of activation or arousal. The attractiveness of stimuli (e.g., food items) thus depends on their arousing properties (e.g., intensity, novelty, complexity). It should be at the optimal level of arousal, otherwise it will not be preferred (K€ oster & Mojet, 2006). Because the level of preferred arousal is different from one person to another, the same stimulus may be just right for some, but too weak or too strong for others. At least 17 motivational factors are associated with choices of foods and beverages consumed at meals and snacks (Chambers et al., 2016; Phan & Chambers, 2016a; Phan & Chambers, 2016b; Phan & Chambers, 2018; Renner, Sproesser, Strohbach, & Schupp, 2012). However, meals and snacks are different in the number of motivations involved. Choices for meals are more complicated, incorporating more motivational factors than those for snacks. Motivations for choosing foods for meals and snacks vary throughout the day, from physiological needs for day-time meals and snacks (such as need and hunger, weight control, and health), to more emotional/psychological needs for evening meals and snacks (i.e., sociability, social image, and pleasure). This might be influenced by the circadian rhythm with the levels of arousal and alertness rising during the morning and reaching a peak near midday (Gibson, 2006). Phan and Chambers (2016a, 2016b) have revealed that food choices for meals and snacks are driven by a number of different motivational factors (Fig. 8.1). For example, in addition to liking, foods for breakfast and lunch are mainly chosen because of need and hunger, convenience, and habits. Food choices for dinner are driven by variety seeking, traditional eating, price, and sociability. On the other hand, choices for morning and afternoon snacks are made based on the concerns of weight control and health, but late-night snack choices often are purely because of pleasure and visual appeal. These motivation paths can be the result of how people conceptualize meals versus snacks. A meal is a structured event with rules of combination and sequence, while a snack does not possess that characteristic (Douglas, 1972). Depending whether an eating occasion is perceived as a meal or a snack, food choice and food intake could be different to reflect the corresponding motivational factors (Wansink et al., 2010). In addition to liking and convenience, choosing foods and drinks for snacks are the result of the interplay of health, pleasure, and sociability, depending on which time of the day the snacking takes place. On the other hand, choices for meals appear to be sophisticated, with a number of interacting motivations, including price, social norms, social image, traditional eating, and variety seeking.

8.2

The dynamics of meals and snacks

When discussing meals and snacks, it is important to consider their dynamic nature. As discussed herein, meals and snacks are different in many aspects, including portion size, energy contribution, food categories represented, and so on. Thus, whether an eating occasion is perceived as a meal or a snack would impact the amount and types

176

Context 0.6 Choice Limitation

0.5

0.4

0.3

Sociability Social Image

Social Norms

Traditional Eating

Variety Seeking

DINNER

Dim 2 (40%)

0.2

0.1

Habits Price

0

Visual Appeal LATE-NIGHT SNACK

BREAKFAST

Health

–0.1 Natural Concerns

LUNCH

Convenience

Liking Pleasure

Need&Hunger

–0.2 MID-AFTERNOON SNACK

–0.3

–0.4 –0.4

Affect Regulation

Weight Control MID-MORNING SNACK

–0.2

0

0.2

0.4

0.6

0.8

1

Dim 1 (48%)

Fig. 8.1 Motivations associated with meals (breakfast, lunch, dinner) and snacks (mid-morning, mid-afternoon, late night). Adopted from Phan, U. T. X., & Chambers, E. (2016). Application of an eating motivation survey to study eating occasions. Journal of Sensory Studies, 31 (2), 114–123.

of foods one eats. This is supported by the research of Shimizu, Payne, and Wansink (2010), who found that environmental cues can alter one’s interpretation of an eating experience as a meal or as a snack and, in turn, significantly impact the amount of food intake. On the other hand, certain foods are seen to be more appropriate for meals than for snacks, and vice versa, and are more fixed into a single category—consuming foods preconceived as “meal” items would signal a “meal” perception, despite environmental influences. Still, active decisions between “meal” or “snack” foods are found to be driven by mood (Gardner, Wansink, Kim, & Park, 2014), making the conceptualization of meals and snacks somewhat affect-regulated and subject to change. This section focuses on the variations in meals and snacks from one country to another, as well as the interchangeability of meals and snacks in the case of breakfast.

8.2.1 Same eating but different meaning: How culture influences the motivations associated with meals and snacks As previously described, food choice is largely driven by motivational factors. Such motivational factors have been found to be universal, as they can be applied to explain different food choices in different countries around the world. This can be seen from application of the motivations developed in Renner et al. (2012) in Germany compared with American and Turkish people (Chambers et al., 2016; Chanadang,

Meal and snack: Two different contexts for foods and drinks

177

Phan, Chambers, & Esen, 2016). Sproesser et al. (2017) has recently confirmed the consistency of eating motives across countries with differing eating environments such as the US, India, and Germany. However, the same motives might not be associated with the same meals/snacks or same foods and beverages in different countries. Chambers et al. (2016) found similarities and differences in the motivations underlying the same meals or snacks between the two US and Turkish groups of participants, especially breakfast. The American sample was driven by convenience when deciding what to eat for breakfast, while the Turkish sample placed more emphasis on health, naturalness, and tradition. The two groups were also different in their motives for lunch, for example, while the Americans still aimed for convenience, the Turks were more driven by sociability and pleasure, which is quite similar to the French (Bildtga˚rd, 2010). The two groups are alike in their motives for dinner (traditional eating and sociability) and late-night snacks (pleasure). Mid-afternoon snacks are associated with weight control and natural concerns in the case of the American sample, but habit is the main driving factor in the case of the Turkish sample. The two groups also indicated that liking is the main driving motive for any food eating event, regardless of whether the event is a meal or snack. The two samples were also different in the motivations associated with certain food categories. Sweets are seen as “indulgent” by the American group, but are considered to be part of a “social event” by Turkish people. These findings reveal how eating patterns and motivations associated with meals and snacks can be similar, but also different, depending on the culture. The differences between meals and snacks are reflected in these culturally driven food choice motivations.

8.2.2 Interchange of meals and snacks: The case of breakfast The questions about the qualification of breakfast as a meal has been brought up by Meiselman (2009), given so many changes during the past centuries in global dietary patterns. Meiselman questions whether we still eat three meals (or more) a day, or whether breakfast has become less meal-like, possibly even falling out of the meal category. These changes could be because breakfast is often solitary, includes few food items, and is often skipped more than other meals. Before the 16th century, across Europe, breakfast was a meal taken only by children, the elderly, and workers/laborers who had to get up early to work (Albala, 2002). Breakfast was forgone by most due to the belief that it was not healthful, and was a sign of gluttony. In France, breakfast was never considered a “real” meal during the medieval era, and prior to the 18th century (Lehmann, 2002). It was classified as a “real” meal for only a brief period in England, roughly from the early 18th century to the middle of the 20th century, and was in decline by the 1960s (Lehmann, 2002). Skipping breakfast has become an increasing trend in many countries, from both the Western (Alexy, Wicher, & Kersting, 2010; Cho, Dietrich, Brown, Clark, & Block, 2003) and Eastern hemispheres (Horikawa et al., 2011; Shim, Paik, & Moon, 2007) over the past few decades. Phan and Chambers (2016a) found breakfast in the US to be light, quick, solitary, and mainly consumed at home. The food items are mainly cereals, eggs, dairy, and coffee. This is very consistent with Chapman, Melton, and Hammond (1998), who

178

Context

reported the weekday breakfast patterns of North Americans are cereal, toast, eggs, bacon, and pancakes. Pliner and Rozin (2000) suggest that many people eat the same breakfast every day; breakfast items do not seem to show the decrease in palatability with repetition shown by other food items. Phan (2015) found evidence supporting this reservation using the Food Choice Map method. Fig. 8.2 shows an example of a food map of one American participant who consumed a banana, milk, coffee, water, and a protein shake every day for breakfast. There were 26 out of 100 participants in Phan’s

Fig. 8.2 An example of the style of consuming a few similar items for breakfast every day.

Meal and snack: Two different contexts for foods and drinks

179

study who reported this style of eating a few similar items for their breakfast each day. This also supports that breakfast is the smallest meal in the US. A similar conclusion can be drawn for Turkish breakfast as well (Chambers et al., 2016). In an earlier Finnish study reported by Roos and Pr€att€al€a (1997), for example, breakfast with a warm dish was included in main meals, but breakfast composed of beverages only was included in snacks. Defining breakfast as a main meal may thus lead to an underestimation in the number of snacks and in the energy intake from snacks. Some studies have included a condition that provides participants with a mid-morning snack category option (having missed breakfast), and therefore, by definition, the snack is also their breakfast (Benton, Slater, & Donohoe, 2001; Smith & Wilds, 2009). This makes it difficult to differentiate breakfast responses from snack responses. As people have become busier with work and daily activities, time spent on preparing breakfast has been significantly reduced. For many people, convenience has become the main motivation for breakfast’s food choices. This has been well observed in Europe, the United States, and Australia. Asian breakfast, on the other hand, still holds its tradition of breakfast preparation. However, Western-style breakfast, with convenience in mind, has been increasing in this region as well. As early as in the 1990s, the main pattern observed in eight Asian countries (Japan, Korea, Thailand, Malaysia, Singapore, Hong Kong, Philippines, and Indonesia) was that rice formed the basis of traditional breakfast, including foods such as porridge, noodles, boiled rice, and glutinous rice. Western-style breakfast, at that time, was emerging in this region, with bread and coffee as the main food and drink. The situation in Vietnam is similar. Nowadays, Vietnamese people do not usually spend time preparing breakfast, due to the popularity of many different breakfast restaurants and food stands. This trend has been investigated by Lachat et al. (2009) in a study to determine the nutritional contribution of out-of-home eating in Vietnamese adolescents. In this study, out-of-home foods are defined as foods prepared outside the home, regardless of where they are consumed. By that definition, Vietnamese breakfast foods are mainly out-of-home foods with diverse options such as the infamous pho-noodles soups, banh mi (baguette with variety of stuffing such as grilled pork, fried egg, cold cuts, sausages, grilled beef, pate, and vegetables), xoi man (savory sticky rice), chao (rice congee), and so on. The Vietnamese breakfast, in many ways, is more savory in comparison to the Western breakfast. Despite the large number of varieties, Vietnamese people usually eat only one dish for breakfast, and they like eating with others. Therefore, in terms of place, preparation process, motivation (mainly, convenience), and number of food and beverage items, Vietnamese breakfast seems to be closer to a snack rather than a meal. Nevertheless, if defined by nutrients, energy intake, and sociability, Vietnamese breakfast can be considered a meal. The fact that Vietnamese breakfast no longer satisfies some of the reported criteria of being a meal has made it less-meal like than what it used to be. Turkish breakfast emerges as a rare example. Turkish women still prepare breakfast for their family, and the foods they prepare are often traditional foods, with boiled eggs, omelets, eggs, cheese, bread, butter, jam, olives, sliced tomatoes and cucumbers, and tea. Traditional dishes that are not eaten on a regular basis are sometimes offered

180

Context

at breakfast, especially when a guest is invited. Breakfast is an important meal to the Turks, and thus it does not bother them to spend time preparing it. The number of food items served at breakfast are about two to five. In some cases, seven, eight, or more items can be offered. The Turks often eat breakfast at home and with family. Therefore, Turkish breakfast does satisfy the criteria of time, place, nutrition, energy intake, sociability, as well as culture, to qualify as a meal. In short, the classification of breakfast is controversial, with support and exceptions on both sides. Depending on the criteria, breakfast can be either a meal or a snack. However, by the fact that its qualification as a meal is questionable and subject to change, breakfast has become less and less meal-like than ever before. Breakfast may just become a snack or extra eating event at some point in the near future.

8.3

Methods for studying food choice in the context of meals and snacks

This section introduces the methodology that has been used in research studies on motivations of everyday food choices under different contexts of meals and snacks, which has been reported in Phan and Chambers (2018). The methodology includes a motivations survey questionnaire developed based on The Eating Motivation Survey (TEMS) (Renner et al., 2012), and a face-to-face, in-depth interview using the Food Choice Map (FCM) (Sevenhuysen & Gross, 2003). Both TEMS and FCM have been applied in the form of a bottom-up process that starts with collecting motivations for individual choices of food and beverage items consumed at specific meals/snacks, then works its way up to generate motivations for corresponding food groups and corresponding meals/snacks. The idea is to form the general motivations from the motives associated with very specific foods and drinks at very specific eating occasions of time, place, and social settings in the respondents’ daily diets. For example, in the case of an apple consumed at lunch, motivations (reasons) such as “because it is good for health” or “because it is rich in vitamins and minerals” have been reported. In other words, for fruits and vegetables, a bottom-up approach can convey the theme of “I eat vegetables, therefore I eat healthy.” The bottom-up approach has proven to successfully guide respondents to a specific eating experience to extract the motivation constructs underlying their food choices in different contexts of meals and snacks throughout the day. This approach provides certain validation for the results based on the nature of the foods and drinks, the motivations, and contexts of eating events (see Chambers et al., 2016; Chanadang et al., 2016; Phan & Chambers, 2016a; Phan & Chambers, 2016b). For example, it is reasonable to find coffee and oatmeal are consumed at breakfast, and the reasons are “because it’s quick to prepare,” “usually have it”, or drinking water for dinner “because it’s low in calories.” The motivations are associated with specific items and eating contexts, and thus could be justified based on knowledge about that item and the eating context. The pros and cons of this approach are presented in Table 8.2.

Meal and snack: Two different contexts for foods and drinks

181

Table 8.2 The pros and cons of the bottom-up approach in studying food choice and motivation Pros

Cons

Engaging respondents to a specific eating experience with specific food/drinks items to provide valid answers Yielding numerous data at individual levels of eating experience and foods/drinks involved Incorporated in survey questionnaire can provide valid and reliable data from a large sample sizea Incorporated in the Food Choice Map helps provide various motivations underlying people’s choices of certain foods/drinks for certain meals/snacksb

Can be very time consuming depending on the number of eating experiences and food/ drink items the respondents reported Data coding and analysis can be complex and time consuming Can result in high drop-out rate when incorporated in survey questionnaire due to the number of questions to answer Can be laborious when incorporated in the Food Choice Map due to the amount of work put into reconstructing the food map and interviewing for every single item on the map

a

See Section 8.3.1. See Section 8.3.2.

b

8.3.1 The eating motivations survey questionnaires using the bottom-up approach The survey questionnaire was developed based on the brief version of TEMS (Renner et al., 2012), which included 15 motivation factors and incorporated two additional factors into the questionnaire, that is, choice limitation and variety seeking. Table 8.3 presents the complete 17 motivational factors and their subscales. The main part of the questionnaire asks respondents to report their most recent eating event by choosing from a list of six eating occasions including breakfast, mid-morning snack, lunch, mid-afternoon snack, dinner, and late-night snack. If their eating event is not one of these six options, they can choose option “other” and specify the name of that event. After that, they are asked to specify how many food and beverage items they consumed for that eating event, and provide the names of all of those items. The respondents are then provided with one TEMS questionnaire per item to indicate the reasons why they chose to eat that food or beverage item for that specific eating event. In other words, if someone reports eating three items for breakfast, for example “a fried egg,” “a cup of coffee,” and “an apple,” then she will fill out three TEMS questionnaires for those three items to report all motivations underlying those choices. Check-All-That-Apply (CATA) is the technique employed in this questionnaire. To facilitate the data processing of the study, seven is the maximal number of items one person can report. So if one eats more than seven items, he or she then is instructed to report only seven representative items that are the most important part of the eating event.

182

Context

Table 8.3 The modified version of the eating motivation questionnaire used in Phan and Chambers (2016a, 2016b) Factor

Motivation subscales

Liking

because I have an appetite for it because it tastes good because I like it because I’m accustomed to eating it because I usually eat it because I am familiar with it because I need energy because it is pleasantly filling because I’m hungry to maintain a balanced diet because it is healthy because it keeps me in shape (e.g., energetic, motivated) because it is quick to prepare because it is the most convenient because it is easy to prepare because someone made it for me and it is the choice because I enjoy it in order to indulge myself in order to reward myself because it belongs to certain situations out of traditions (e.g., family traditions, special occasions) because I grew up with it because it is natural (e.g., not genetically modified) because it contains no harmful substances (e.g., pesticides, pollutants, antibiotics) because it is organic because it is social so that I can spend time with other people because it makes social gatherings more comfortable because it is inexpensive because I don’t want to spend any more money because it is on sale because the presentation is appealing (e.g., packaging) because it spontaneously appeals to me (e.g., situated at eye level, appealing colors) because I recognize it from advertisements or have seen it on TV because it is low in calories because I watch my weight because it is low in fat because I am sad because I am frustrated because I feel lonely because it would be impolite not to eat it to avoid disappointing someone who is trying to make me happy because I am supposed to eat it

Habit

Need and Hunger Health

Convenience

Pleasure

Traditional Eating Natural Concerns

Sociability

Price

Visual Appeal

Weight Control

Affect Regulation Social Norms

Meal and snack: Two different contexts for foods and drinks

183

Table 8.3 Continued Social Image

Choice Limitation Variety Seeking

because because because because because because because

it is trendy it makes me look good in front of others others like it it was what was served it is the only choice I like to eat a variety of different foods each day I don’t like to eat the same food for the same meal everyday

This questionnaire consists of seventeen motivation factors and their corresponding fifty motivation subscales.

8.3.2 The food choice map using the bottom-up approach The Food Choice Map (FCM) method developed by Sevenhuysen and Gross (2003) is a qualitative interview procedure that records the frequency of food consumption and the reasons for particular food choices. In an FCM interview, respondents are asked to recall foods they ate often in a usual week, followed by questions about the meal or snack, foods eaten less frequently at those meals, and a variety of aspects related to those foods and their frequencies of consumption, such as where purchased, when and with whom consumed, likely important for health, perception of cost, and other aspects of interest to the respondent. To construct the food map, Sevenhuysen and Gross employed an artist to quickly draw a picture of the food items reported by the respondents during the interview. The drawing only serves as generic stimuli to convey the food items. In Phan and Chambers’ studies, real pictures of food and drink items were used. A total of 700 pictures were printed on 5  9 cm name cards (Avery, CA, USA), and were handed to the respondents. The respondents were asked to pick out the food images that represent the typical foods and beverages they often consume in a week. Then, on an 84.1  118.4 cm worksheet posted on the wall, they used temporary glue to stick the food images in a place to reflect the time of consumption and the number of days in a usual week (1—once time per week; 7—every day) that they eat the food for the same eating event. In case a food was not found from the pictures provided, the respondents can write the name of the food on a post-it note to place it on the map. Fig. 8.3 presents an example of a food choice map created by this method. Upon completion of the food choice map of their weekly diet, the respondents are asked to indicate: (1) name, time, and location for all the eating events reported on the map, (2) reasons (motivations) for choosing each food or beverage by responding to the question “What reasons do you have for choosing this (food/beverage) for (breakfast, lunch, dinner, etc.)?” using their own words. To facilitate this technique, the interviews should be conducted by interviewers who have been trained for either interviewing or focus group moderating. A detailed interviewer’s guide in which the core questions are highlighted should also be developed and provided to the interviewers to aid the interview process. Probing questions for different scenarios should be provided in the guide to prepare the interviewers for situations when the respondents either keep repeating the same reason, such as “because I like it,” or are less willing to talk. Depending on the number of

184

Context

Fig. 8.3 An example of a Food Choice Map. Columns 1 to 7 contain the foods/beverages that were consumed once to seven times in a usual week. Column 0 contains the foods/beverages that were consumed less than once a week. The rows present the time of the day at which the food items were consumed.

foods and drinks reported, this FCM interview could be time consuming. Table 8.4 presents some examples of the types of reasons resulting from a FCM interview.

8.4

Implication for future research

Based on dictionary definitions and empirical evidence, food and drink consuming experiences can be placed in one of two accepted categories: snacks and meals. Each, however, has its own specifications in terms of content, location, social aspects, and

Meal and snack: Two different contexts for foods and drinks

185

Table 8.4 Examples of some reasons for choosing specific foods resulting from an FCM interview (Phan, 2015) Type of reason

Examples

Weight control

I struggle with my weight, so I do try to drink 64 fl. oz. water every day because it keeps you full Brown rice is healthy, easy to cook, easy to go with side dishes When I am really hungry, and I’m in a rush, I go to fries and a hamburger

Health perception Convenience

associated motivations. Both contexts play important roles, satisfying both physical and psychological needs, but may be driven by different motivations in response to one’s circadian rhythm. Day-time meals or snacks have been found to be associated with more physical motivations such as need and hunger, health, weight control, and cost. On the other hand, night-time meals and snacks are more influenced by psychological motivations such as sociability, social image, and social norms; the need of sharing food and interacting with others at the dinner table or during a night-out is universal. Therefore, research in the domain of food choice should take into account these two different contexts to fully understand how a specific choice is made. Considering how time of day plays a role in food choice motivations also presents opportunities for driving healthier choices. With convenience and versatility being higher drivers of day-time eating, nutritionists and health behavior specialists can recommend healthful, low-calorie foods that people can buy in large portions and have available during the day. Making these healthful foods available and accessible could significantly impact diet because it is anticipated that those foods would be consumed up to four times per day. Evening meals and snacks have less opportunity to be altered because of sociability motivations. Pleasure is the biggest obstacle for any attempt to change night-time snacking because it fulfills people’s “soul.” Nutritionists and health educators should take into account this aspect of people’s food choice, and allow room for those less-healthful foods in people’s diets at a certain limit to maintain the balance between physical and psychological needs. While eating experiences can be defined differently and driven by different motivational factors between individuals, food remains integral to the human experience. Understanding how and why food is chosen can open opportunities for influence, or for new product markets, making it a valuable area for future research. This includes developing new research methodology such as the bottom-up approach presented in this chapter to give insights into the dynamics of food choice and its elements.

References Albala, K. (2002). Hunting for breakfast in medieval and early modern Europe. In H. Walker (Ed.), The meal (pp. 20–30). Totnes, Devon, England: Oxford Symposium/Prospect. Alexy, U., Wicher, M., & Kersting, M. (2010). Breakfast trends in children and adolescents: Frequency and quality. Public Health Nutrition, 13(11), 1795–1802.

186

Context

Ballard-Barbash, R., Thompson, F. E., Graubard, B. I., & Krebs-Smith, S. M. (1994). Variability in percent energy from fat throughout the day: Implications for application of total diet goals. Journal of Nutrition Education, 26(6), 278–283. Bellisle, F. (2014). Meals and snacking, diet quality and energy balance. Physiology & Behavior, 134, 38–43. Bellisle, F., Dalix, A. M., Mennen, L., Galan, P., Hercberg, S., de Castro, J. M., et al. (2003). Contribution of snacks and meals in the diet of French adults: A diet-diary study. Physiology & Behavior, 79(2), 183–189. Bellisle, F., Monneuse, M. O., Steptoe, A., & Wardle, J. (1995). Weight concerns and eating patterns: A survey of university students in Europe. International Journal of Obesity and Related Metabolic Disorders, 19(10), 723–730. Benton, D., Slater, O., & Donohoe, R. T. (2001). The influence of breakfast and a snack on psychological functioning. Physiology and Behavior, 74, 559–571. Bildtga˚rd, T. (2010). What it means to “eat well” in France and Sweden. Food and Foodways, 18(4), 209–232. Bisogni, C. A., Falk, L. W., Madore, E., Blake, C. E., Jastran, M., Sobal, J., et al. (2007). Dimensions of everyday eating and drinking episodes. Appetite, 48(2), 218–231. Blake, C. E., Bisogni, C. A., Sobal, J., Devine, C. M., & Jastran, M. (2007). Classifying foods in contexts: How adults categorize foods for different eating settings. Appetite, 49(2), 500–510. Blake, C. E., Fisher, J. O., Ganter, C., Younginer, N., Orloski, A., Blaine, R. E., et al. (2015). A qualitative study of parents’ perceptions and use of portion size strategies for preschool children’s snacks. Appetite, 88, 17–23. Chambers, D., Phan, U. T. X., Chanadang, S., Maughan, C., Sanchez, K., Di Donfrancesco, B., et al. (2016). Motivations for food consumption during specific eating occasions in Turkey. Food, 5(2), 39. Chanadang, S., Phan, U. T. X., Chambers, D., & Esen, E. (2016). Motivation of everyday food choices: Differences between American and Turkish respondents. In: D. H. Nguyen, D. Valentin, T. M. Le, A. D. L. Nguyen, S. Chollet, & H. Abdi (Eds.), The proceedings of SPISE2016, Taste and try before you fry: From product formulation to consumer experience, SPISE2016, Ho Chi Minh City, pp. 63–73. Chaplin, K., & Smith, A. P. (2011). Definitions and perceptions of snacking. Current Topics in Nutraceuticals Research, 9(1/2), 53. Chapman, G. E., Melton, C. L., & Hammond, G. K. (1998). College and university students’ breakfast consumption patterns: Behaviours, beliefs, motivations and personal and environmental influences. Canadian Journal of Dietetic Practice and Research, 59(4), 176–182. Chiva, M. (1997). Cultural aspects of meals and meal frequency. British Journal of Nutrition, 77 (S1), S21–S28. Cho, S., Dietrich, M., Brown, C. J., Clark, C. A., & Block, G. (2003). The effect of breakfast type on total daily energy intake and body mass index: Results from the third National Health and Nutrition Examination Survey (NHANES III). Journal of the American College of Nutrition, 22(4), 296–302. de Castro, J. M., Brewer, E. M., Elmore, D. K., & Orozco, S. (1990). Social facilitation of the spontaneous meal size of humans occurs regardless of time, place, alcohol or snacks. Appetite, 15, 89–101. de Graaf, C. (2000). Nutritional definitions of the meal. In H. L. Meiselman (Ed.), Dimensions of the meal: The science, culture, business, and art of eating (pp. 47–59). Gaithersburg, MD: Aspen Publishers.

Meal and snack: Two different contexts for foods and drinks

187

Douglas, M. (1972). Deciphering a meal. Daedalus, 1, 61–81. Douglas, M., & Nicod, M. (1974). Taking the biscuit: The structure of British meals. New Society, 30(637), 744–747. Duffey, K. J., Pereira, R. A., & Popkin, B. M. (2013). Prevalence and energy intake from snacking in Brazil: Analysis of the first nationwide individual survey. European Journal of Clinical Nutrition, 67(8), 868–874. Fjellstr€om, C. (2004). Mealtime and meal patterns from a cultural perspective. Scandinavian Journal of Nutrition, 48(4), 161–164. Gardner, M. P., Wansink, B., Kim, J., & Park, S. B. (2014). Better moods for better eating? How mood influences food choice. Journal of Consumer Psychology, 24(3), 320–335. Gibson, E. L. (2006). Emotional influences on food choice: Sensory, physiological and psychological pathways. Physiology & Behavior, 89, 53–61. Herman, C. P. (2015). The social facilitation of eating. A review. Appetite, 86, 61–73. Herman, C. P., Roth, D. A., & Polivy, J. (2003). Effects of the presence of others on food intake: A normative interpretation. Psychological Bulletin, 129(6), 873–886. Higgs, S. (2015). Social norms and their influence on eating behaviours. Appetite, 86, 38–44. https://doi.org/10.1016/j.appet.2014.10.021. Horikawa, C., Kodama, S., Yachi, Y., Heianza, Y., Hirasawa, R., Ibe, Y., et al. (2011). Skipping breakfast and prevalence of overweight and obesity in Asian and Pacific regions: A metaanalysis. Preventive Medicine, 53(4–5), 260–267. Jacquier, E. F., Gatrell, A., & Bingley, A. (2017). “We don’t snack”: Attitudes and perceptions about eating in-between meals amongst caregivers of young children. Appetite, 108, 483–490. Kant, A. K., & Graubard, B. I. (2015). 40-year trends in meal and snack eating behaviors of American adults. Journal of the Academy of Nutrition and Dietetics, 115(1), 50–63. Kerr, M. A., Rennie, K. L., McCaffrey, T. A., Wallace, J. M. W., Hannon-Fletcher, M. P., & Livingstone, M. B. E. (2009). Snacking patterns among adolescents: A comparison of type, frequency and portion size between Britain in 1977 and Northern Ireland in 2005. British Journal of Nutrition, 101, 123–131. Kjærnes, U., Holm, L., Gronow, J., M€akel€a, J., & Ekstr€om, M. P. (2009). The study of Nordic meals: Lessons learnt. In H. L. Meiselman (Ed.), Meals in science and practice: Interdisciplinary research and business applications (pp. 69–91). Cambridge: Woodhead Publishing Ltd. K€ oster, E., & Mojet, J. (2006). Theories of food choice development. In L. Frewer & H. Trijp (Eds.), Understanding consumers of food products (pp. 93–124). Cambridge: Woodhead Publishing Ltd. Lachat, C., Le, B. K., Nguyen, C. K., Nguyen, Q. D., Nguyen, D. V. A., Roberfroid, D., et al. (2009). Eating out of home in Vietnamese adolescents: Socioeconomic factors and dietary associations. The American Journal of Clinical Nutrition, 90(6), 1648–1655. Lawless, H. T. (2000). Sensory combinations in the meal. In H. L. Meiselman (Ed.), Dimensions of the meal: The science, culture, business and art of eating (pp. 92–106). Gaithersburg, Maryland: Aspen Publishers, Inc. Lehmann, G. (2002). Meals and mealtimes, 1600–1800. In H. Walker (Ed.), The meal. Totnes, Devon (pp. 139–154). England: Prospect Books. M€akel€a, J. (1991). Defining a meal. In E. L. F€urst, R. Pr€att€al€a, M. Ekstr€ om, L. Holm, & U. Kjaernes (Eds.), Palatable worlds: Sociocultural food studies (pp. 87–95). Oslo: Solum Forlag. M€akel€a, J. (2000). Cultural definitions of the meal. In H. L. Meiselman (Ed.), Dimensions of the meal: The science, culture, business, and art of eating (pp. 7–18). Gaithersburg: Aspen.

188

Context

Meiselman, H. L. (2009). Meals in science and practice: An overview and summary. In H. L. Meiselman (Ed.), Meals in science and practice: Interdisciplinary research and business applications (pp. 16–34). Cambridge: Woodhead Publishing Ltd. Ochs, E., & Shohet, M. (2006). The cultural structuring of mealtime socialization. New Directions for Child and Adolescent Development, (111), 35–49. Ovaskainen, M. J., Tapanainen, H., & Pakkala, H. (2010). Changes in the contribution of snacks to the daily energy intake of Finnish adults. Appetite, 54, 623–626. Ovaskainen, M. L., Reinivuo, H., Tapanainen, H., Hannila, M. L., Korhonen, T., & Pakkala, H. (2006). Snacks as an element of energy intake and food consumption. European Journal of Clinical Nutrition, 60(4), 494–501. Petri, H., & Govern, J. (2012). Motivation: Theory, research, and application. Belmont, CA: Wadsworth, Cengage Learning. Phan, U. T., & Chambers, E. (2018). Motivations for meal and snack times: Three approaches reveal similar constructs. Food Quality and Preference, 68, 267–275. Phan, U. T. X. (2015). Motivation of every day food choices. Ph.D. thesisKansas State University. Available from. https://krex.k-state.edu. Accessed 26 December 2017. Phan, U. T. X., & Chambers, E. (2016a). Application of an eating motivation survey to study eating occasions. Journal of Sensory Studies, 31(2), 114–123. Phan, U. T. X., & Chambers, E. (2016b). Motivations for choosing various food groups based on individual foods. Appetite, 105, 204–211. Pliner, P., & Rozin, P. (2000). The psychology of the meal. In H. L. Meiselman (Ed.), Dimensions of the meal: The science, culture, business, and art of eating (pp. 19–46). Gaithersburg, MD: Aspen Publishers, Inc. Prescott, S. L., & Logan, A. C. (2017). Each meal matters in the exposome: Biological and community considerations in fast-food-socioeconomic associations. Economics & Human Biology, 27, 328–335. Renner, B., Sproesser, G., Strohbach, S., & Schupp, H. T. (2012). Why we eat what we eat. The eating motivation survey (TEMS). Appetite, 59(1), 117–128. Ritchie, L. D. (2012). Less frequent eating predicts greater BMI and waist circumference in female adolescents. American Journal of Clinical Nutrition, 95(2), 290–296. Roos, E., & Pr€att€al€a, R. (1997). Meal patterns and nutrient intake among adult Finns. Appetite, 29, 11–24. Sevenhuysen, G. P., & Gross, U. (2003). Documenting the reasons people have for choosing their food. Asia Pacific Journal of Clinical Nutrition, 12(1), 30–37. Shim, J. E., Paik, H. Y., & Moon, H. K. (2007). Breakfast consumption pattern, diet quality and health outcomes in adults from 2001 National Health and nutrition survey. Korean Journal of Nutrition, 40(5), 451–462. Shimizu, M., Payne, C. R., & Wansink, B. (2010). When snacks become meals: How hunger and environmental cues bias food intake. International Journal of Behavioral Nutrition and Physical Activity, 7(1), 63. Smith, A. P., & Wilds, A. (2009). The effects of cereal bars for breakfast and mid-morning snacks on mood and memory. International Journal of Food Science and Nutrition, 60 (s4), 63–69. Sobal, J. (2000). Sociability and meals: Facilitation, commensality, and interaction. In H. L. Meiselman (Ed.), Dimensions of the meal: The science, culture, business and art of eating (pp. 119–133). Gaithersburg, MD: Aspen Publishers Inc. Sproesser, G., Ruby, M. B., Arbit, N., Rozin, P., Schupp, H. T., & Renner, B. (2017). The eating motivation survey: Results from the USA, India and Germany. Public Health Nutrition, 21, 515–525. https://doi.org/10.1017/S1368980017002798.

Meal and snack: Two different contexts for foods and drinks

189

Summerbell, C. D., Moody, R. C., Shanks, J., Stock, M. J., & Geissler, C. (1995). Sources of energy frommeals versus snacks in 220 people in four age groups. European Journal of Clinical Nutrition, 49, 33–41. Toornvliet, A. C., Pijl, H., Hopman, E., Elte-De Wever, B. M., & Meinders, A. E. (1996). Serotoninergic drug-induced weight loss in carbohydrate craving obese patients. International Journal of Obesity and Related Metabolism Disorders, 20, 917–920. Wadhera, D., & Capaldi, E. D. (2012). Categorization of foods as “snack” and “meal” by college students. Appetite, 58(3), 882–888. Walker, H. (2002). The meal. Totnes, Devon, England: Prospect Books. Wang, D., van der Horst, K., Jacquier, E., & Eldridge, A. L. (2016). Snacking among US children: Patterns differ by time of day. Journal of Nutrition Education and Behavior, 48(6), 369–375. Wansink, B., Payne, C. R., & Shimizu, M. (2010). “Is this a meal or snack?” situational cues that drive perceptions. Appetite, 54(1), 214–216. Wong, R. (2000). Motivation: A biobehavioural approach. Cambridge, UK: Cambridge University Press. Younginer, N. A., Blake, C. E., Davison, K. K., Blaine, R. E., Ganter, C., Orloski, A., et al. (2016). “What do you think of when I say the word ‘snack’?” towards a cohesive definition among low-income caregivers of preschool-age children. Appetite, 98, 35–40.

This page intentionally left blank

The meal as the proper context for food and drinks

9

Johanna Ma€kela€, Mari Niva Faculty of Educational Sciences, University of Helsinki, Helsinki, Finland

9.1

Defining proper meals

Why do meals, and especially proper meals matter? Because in addition to offering substance, meals are symbols, rituals, arenas for socialization and commensality, and markers of both similarities and differences. Meals have a profound meaning to human life and culture. Everyday patterns of eating are embedded and intertwined in the cultural, social, and economic contexts and conditions of societies. Therefore, nutritional, cultural, and social aspects of eating carry meanings related to “properness.” In this chapter, we take a social scientific, and in particular, sociological approach, and focus on the complexities and contexts of a proper meal by analyzing its contents, ingredients, and commensality, especially in the Western world. We pay particular attention to what a proper meal is and how it has changed, and how it might alter in the future. Therefore, we explore the ways in which it is challenged and redefined. Yates & Warde (2015, p. 300) define the cultural complex of the meal as consisting of foodstuffs, dishes, patterns of eating events (the structure of the episodes), event formats (the organization of dishes in parallel and in series), preparation and provisioning, and social occasion (place, company, and social context). Earlier, a Nordic study (M€akel€a et al., 1999) constructed an “eating system” that included three dimensions: eating pattern, meal format, and the social organization of eating. Here, the eating pattern includes the rhythm and the number of eating events, as well as the alternations of hot and cold eating events. The meal format refers to both the composition of the main course, and the sequence of the whole meal. The social organization is comprised of where and with whom people eat, and who did the cooking. The scholarly interest in the structure and context of meals was initially sparked when Douglas and Nicod (1974) analyzed the structure of British working-class meals in the 1970s. This well-cited classic work explored the structure and grammar of meals. In their classification of meals, they used oppositions such as savory and sweet, bland and spiced, hot and cold, and liquid and dry; but also focused on complexity, copiousness, and ceremoniousness of meals. In another famous article, Douglas (1972/1997) pointed out that the meaning of a meal is based on a system of repeated analogies, and that meals are connected to a wider social system. According to Douglas (1972/1997, p. 41), a meal requires “a table, a seating order, restriction on movement, and on alternative occupations.” Context. https://doi.org/10.1016/B978-0-12-814495-4.00009-X Copyright © 2019 Elsevier Inc. All rights reserved.

192

Context

A meal is not, however, the only type of eating that Douglas and Nicod presented. They identified four different types of eating: first, a “food event;” second, a “structured event” organized by rules concerning time, place, and sequence of action; third, a “meal,” where food is eaten as part of a structured event obeying rules of combination and sequence; and finally, a “snack,” which, in contrast to a meal, is an unstructured food event without any rules of combination and sequence. (See Chapter 8 by Phan in this edition for a detailed description of meals and snacks.) The meal system constructed by Douglas and Nicod consists of three types of meals: (A) a major meal/the main meal, (B) a minor meal/the second meal, and (C) an even less significant meal/the third meal (Douglas, 1983; Douglas & Nicod, 1974). The most used of these meal types is A, which has the same basic structure based on a staple (potato), a center (meat, fish, or egg), trimmings (vegetables), and dressing (gravy). This structure was investigated further in the 1980s by Anne Murcott (1982), who found that in South Wales, a “proper meal” was a plateful, and included one course based on meat, two vegetables, and gravy, which tied the other ingredients into a proper meal. The British research on meals has influenced research in many other countries, particularly in the Nordic countries. In the 1980s, studies on food and dietary habits in Denmark showed that typical hot meals consisted of meat, potatoes, vegetables, and sauce (Haraldsdo´ttir, Holm, Jensen, & Møller, 1987). In a Swedish study, a cooked meal had four components. The main ingredient was usually animal protein. The first trimming was the starchy base of the meal, and the second trimming included vegetables. Extra trimmings could be vegetables or condiments (Ekstr€om, 1990). Somewhat later, in Norway, a proper meal was a cooked one with vegetable trimmings (Bugge & Døving, 2000). In Finland, working mothers’ ideas of a proper meal consisted of three factors: a hot dish, a salad, and company (M€akel€a, 1996), emphasizing not only the contents of the meals, but also commensality. Generally, these studies in the UK and the Nordic countries show that a proper meal in these countries typically includes a main course only, which is in stark contrast to France, where a proper meal consists of several, typically three or four, courses (Poulain, 2002). It should be noted that research on meals in Western Europe has dominated meals research literature, and this focus should be extended. Yet, it is evident that the content of the concept of a proper meal is demarcated in culture-specific ways. For example, when Torbj€ orn Bildtga˚rd (2010) interviewed Swedes and the French about a related concept, “eating well,” he discovered that despite some shared themes, there were also obvious differences in the “alimentalities” (i.e., alimentary mentalities) of the two countries (ibid. p. 210). For both Swedes and the French, eating well included the objective of reaching a nutritional balance through eating healthy foods. The Swedes referred to following the “the plate model” or “the food wheel,” both developed as pedagogical tools in nutrition science, as a means to achieve balance; whereas the French mentioned “the complete meal” (starter, main course, cheese, dessert, i.e., a proper meal) as a guarantee for balance. In the latter, it was the traditional structure of the meal that was thought to ensure nutritional variety. A further difference was found in the way in which commensality was tied in with “eating well.” The French

The meal as the proper context for food and drinks

193

talked about conviviality and sharing the pleasure of eating, whereas the Swedes hardly referred to commensality as part of eating well. Since the work of Douglas, it has been evident that meals carry several connotations, not only in everyday parlance, but in scholarly use and research, too. For example, Lamont (1992) has identified four possible meanings of a meal: (1) “a timely repast,” food eaten at a customary eating time, (2) an objectified structure of food components, (3) action following a “script,” or social action aiming for particular ends, and (4) a social event that is meaningful for its participants. According to Lamont, the first two construct the meal as an object, while the two latter meanings emphasize the meal as a social event, as lived experience. He further contends that in studying meals, the physiological (taste perception), psychological (the omnivore’s paradox), and sociological (socio-cultural influences) aspects should be taken into account.

9.2

What elements are regarded as essential in a proper meal?

As noted herein, there are cultural conventions, routines, and ideals regulating what food elements are essential for a meal to be considered proper. In a Nordic population study, the data for which were collected in Denmark, Finland, Norway, and Sweden in 2012 (N ¼ 8248; see, e.g., Holm et al., 2016), we asked the respondents to reflect upon, among other things, what elements they considered as important in a proper hot meal. They were asked the following question: “Think about an ordinary day, at home. If you were to eat a proper hot meal, what would be included?” The respondents could choose as many options as they wanted in a list including (1) meat or fish, (2) potatoes, rice, or pasta, (3) bread, (4) cooked vegetables, (5) salad or fresh vegetables, (6) condiments (sauces etc.), (7) starter, (8) dessert), (9) other, and (10) don’t know. Table 9.1 shows the percentages of respondents choosing each item separated for gender, three age groups, and three educational groups in the four countries. The results for items “other” and “don’t know” are not reported in the table, but are briefly referred to in the text that follows. The results show both national and socio-demographic similarities and differences in the elements that were regarded as essential in a proper hot meal. First, based on the results, the most important element of the meal was meat or fish (93% selected this item), irrespective of country, gender, or education. Some differences can be seen in age: older respondents chose meat or fish a little more often than the young. However, the differences were not very large. The Nordic citizens also seemed to value a staple in a hot meal, that is, potatoes, rice, or pasta. Around four in five respondents in all countries chose this item; Danes and Swedes a little more often than Norwegians and Finns, and men somewhat more often than women. Other socio-demographic differences were small. The largest national differences were found in bread as a meal element. For a large majority of Danes, Norwegians, and Swedes, bread was not a significant item at all.

Table 9.1 Percentage of respondents choosing each food type as a part of a proper home-cooked meal (multiple options possible) (due to rounding of cell counts, total number of valid cases N ¼ 8246–8251)a Gender Man

Woman

Age P

Meat or fish (N 5 4102) (N 5 4146) DK (N ¼ 2060) NO (N ¼ 2079) SE (N ¼ 2065) FI (N ¼ 2044) All countries (N ¼ 8248)

91 95 94 93 93

92 94 94 90 93

Bread DK (N ¼ 2060) NO (N ¼ 2079) SE (N ¼ 2065) FI (N ¼ 2044) All countries (N ¼ 8248)

87 76 85 82 83

79 75 81 73 77

ns. ns. ns. * ns.

11 6 9 31 14

50–80

P

90 90 89 84 88

93 95 94 94 94

91 97 96 94 95

*** ns. ** *** ***

84 69 82 74 77

83 75 83 80 80

83 80 84 78 81

ns. *** *** *** ***

14 4 7 34 14

16 7 11 43 19

15 11 15 48 23

Middle

High

P

91 94 94 90 92

92 95 96 92 94

90 95 91 93 92

(N 5 8248) ns. ns. ** ns. **

(N 5 2557) (N 5 3369) (N 5 2322) ns. *** ns. * **

(N 5 2166) (N 5 2656) (N 5 3426) *** ** *** *** ***

Low

All

(N 5 2557) (N 5 3369) (N 5 2322)

(N 5 2166) (N 5 2657) (N 5 3425)

(N 5 4102) (N 5 4146) 19 9 15 56 25

35–49

(N 5 2166) (N 5 2656) (N 5 3425)

Potatoes, (N 5 4102) (N 5 4146) rice, or pasta DK (N ¼ 2060) NO (N ¼ 2079) SE (N ¼ 2065) FI (N ¼ 2043) All countries (N ¼ 8247)

15–34

Education

87 72 85 77 80

84 78 84 79 81

78 75 80 75 77

(N 5 8248) *** * * ns. ***

(N 5 2557) (N 5 3369) (N 5 2322) ns. *** *** *** ***

18 8 17 53 25

15 9 12 42 19

12 6 8 31 14

91 94 94 92 93***

83 75 83 77 80***

(N 5 8248) * ns. *** *** ***

15 8 12 43 19***

Cooked vegetables DK (N ¼ 2060) NO (N ¼ 2079) SE (N ¼ 2065) FI (N ¼ 2044) All countries (N ¼ 8248)

(N 5 4102) (N 5 4146) 58 63 45 37 51

65 74 53 51 61

(N 5 2166) (N 5 2656) (N 5 3426) ** *** ** *** ***

Salad or raw (N 5 4102) (N 5 4146) vegetables DK (N ¼ 2060) NO (N ¼ 2079) SE (N ¼ 2065) FI (N ¼ 2044) All countries (N ¼ 8248)

52 46 65 68 58

64 59 75 80 69

46 54 56 48 51

36 47 55 37 44

63 73 50 37 56

66 71 53 52 60

*** *** *** ** ***

(N 5 2166) (N 5 2656) (N 5 3426) *** *** *** *** ***

Condiments (N 5 4102) (N 5 4146) (sauces, etc.) DK (N ¼ 2060) NO (N ¼ 2079) SE (N ¼ 2065) FI (N ¼ 2044) All countries (N ¼ 8248)

54 58 43 40 49

(N 5 2556) (N 5 3369) (N 5 2322)

55 45 69 70 59

61 49 68 73 63

58 60 71 77 67

44 43 57 40 46

39 54 57 42 48

41 53 52 43 47

61 70 47 43 55

68 74 52 48 61

** *** ns. ns. ***

(N 5 2557) (N 5 3370) (N 5 2321) ns. *** ns. * ***

(N 5 2166) (N 5 2656) (N 5 3426) *** ** ns. *** ***

58 61 48 43 53

(N 5 8248)

52 46 69 70 59

61 52 69 74 64

63 59 72 79 68

(N 5 8248) *** *** ns. ** ***

(N 5 2557) (N 5 3370) (N 5 2322) ns. *** ns. ns. ns.

46 58 59 46 52

43 50 58 42 48

32 43 49 38 41

62 68 49 44 56***

58 52 70 74 64***

(N 5 8248) *** *** *** * ***

41 51 55 42 47***

Continued

Table 9.1 Continued Gender Man

Starter DK (N ¼ 2060) NO (N ¼ 2079) SE (N ¼ 2065) FI (N ¼ 2044) All countries (N ¼ 8248)

Dessert DK (N ¼ 2060) NO (N ¼ 2079) SE (N ¼ 2065) FI (N ¼ 2044) All countries (N ¼ 8248)

Woman

Age P

(N 5 4102) (N 5 4146) 3 3 4 3 3

2 1 2 2 2

2 9 4 16 8

35–49

50–80

P

(N 5 2166) (N 5 2656) (N 5 3426) ns. * * ns. ***

(N 5 4102) (N 5 4146) 6 12 6 18 11

15–34

Education

2 1 3 2 2

1 1 2 3 2

3 3 4 4 4

5 11 5 16 9

3 8 3 13 7

4 12 7 21 11

ns. * ns. ns. ***

2 1 2 2 2

High

P

2 2 3 3 3

3 3 3 4 3

(N 5 8248) ns. * ns. * **

(N 5 2557) (N 5 3369) (N 5 2322) ns. * * *** ***

5 13 3 19 10

***P < .001, **P < .01, *P < .05, Chi Square test. For column “all,” the stars refer to statistical differences between the countries.

a

Middle

(N 5 2557) (N 5 3370) (N 5 2322)

(N 5 2166) (N 5 2656) (N 5 3426) *** * ns. ns. ***

Low

All

4 9 6 15 9

3 9 6 19 9

2 2 3 3 3 (ns.)

(N 5 8248) ns. * * ns. *

4 11 5 17 9***

The meal as the proper context for food and drinks

197

Finns differed from others: nearly half of them considered bread as an item to be included in a proper meal. Particularly men, those in the oldest age group and those with least education, were fond of bread. Such socio-demographic differences were most prominent in Finland, but can also be seen in Sweden. The difference between Finland and the other three countries can perhaps be traced to the role of bread in national food cultures. For centuries, rye bread was a central component of Finnish peasant diets, and today whole grain, fiber-rich bread is recommended in the “plate model” of the national nutrition recommendations as part of hot meals (National Nutrition Council, 2014). Many Finns consider rye bread made of 100% rye (with no other grains) as the only “proper” bread, and Finns living abroad typically long for this “Finnish” rye bread. Indeed, in 2017, rye bread was voted as the Finnish national food (http://www.elo-saatio.fi/finlands-national-food-is-rye-bread). Interestingly though, in an analysis on meal complexity on actual meals, the differences in bread eating between the Nordic countries were very small, and Finland did not differ from the other countries (Kahma, M€akel€a, Niva, & Lund, 2014). This suggests that for Finns, bread as part of a meal may be more important than for other Nordic people as an ideal, but not necessarily as a practice. The results also revealed curious national differences in terms of what is regarded as the proper way of preparing vegetables for a proper meal. In Denmark and Norway, around two-thirds of the respondents thought that a proper meal would include cooked vegetables, whereas less than half of Swedes and Finns did the same. The opposite was true for salads or fresh vegetables: more Finns and Swedes than Danes and Norwegians regarded salad or fresh vegetables as an essential component of a proper meal. Women, the older respondents, and those with higher education tended to be more prone to think of vegetables—either cooked or fresh—as an essential component, possibly reflecting the higher health-orientation among these groups (e.g., Konttinen, Sarlio-L€ahteenkorva, Silventoinen, M€annist€ o, & Haukkala, 2012). Very few respondents regarded a starter as an essential element of a proper meal. Dessert was somewhat more popular, particularly among Finns, and to a lesser extent, Norwegians. In both starters and desserts, the socio-demographic differences were less pronounced than for the other items. Swedes tended to regard condiments as essential somewhat more often than others, particularly Danes and Finns, and men and those with lower education more often than women, and those with higher education. The share of respondents selecting the open response option “other” was 2%–3% in all countries (no table). The open responses contained a variety of items, some of which would be categorized among the preceding items. Items mentioned several times included chicken (showing that people don’t always think of chicken as meat, or that they may think “meat” refers only to red meat), egg, cheese, pulses, soya products, bulgur or quinoa, or drinks, such as water, milk, beer, wine, and coffee. In Norway and Finland, the youngest age group, and in Denmark, Sweden, and Finland, the groups with the highest education selected the item “other” somewhat more often than others, possibly reflecting a higher share of special diets such as vegetarianism or veganism among the young and the highly educated. The share of respondents selecting the item “don’t know” was 1%–2% (no table).

198

Context

The results presented in the table represent Nordic ideas and ideals of a proper meal prepared and eaten at home on an “ordinary day.” The results may have been different had we asked about the components of a festive meal or a weekend meal, and the listed components do not reveal anything about the variety and complexity of the dishes that they would be used to prepare. Indeed, Bugge and Alma˚s (2006) noted that what is proper for a family meal depends on the day of week: they describe how their Norwegian interviewees used minced meat as an ingredient, and found clear differences in their ideas about how it is properly used on weekdays, Saturdays, and Sundays. The dishes carry different meanings, and a minced meat dish served on Monday may not be suitable for Sunday. Similar context-dependent differentiation about “proper” cooking techniques and dishes probably applies to many other ingredients. It is not only in ideals, but also in practice, that the components of a meal differ temporally, for example, in the weekly cycle. In Britain, Yates and Warde (2015) found that on weekdays, the most popular combinations of evening meal components included fish and potatoes in some form (15% of the 1419 respondents), mixed, minced, or chopped meat and potatoes (15%), poultry and potatoes (12%), and pasta with mixed, minced, or chopped meat (11%). On weekends, the most popular combinations were beef and potatoes (19%), poultry and potatoes (13%), cheese and bread (13%), and fish and potatoes (9%). Young people more often had Italian and Asianinspired dishes. In general, the most popular combinations suggested that meals with a center and a staple were still popular, although they also included dishes in which the ingredients were mixed.

9.3

The proper context of a meal

Meal studies generally indicate that for most people, the properness of a meal relates not only to what is on the plate, that is, the material elements of the meal, but also to sharing the meal with other people. The Western ideal type of a proper meal includes the notion of company, most often the nuclear family (e.g., M€akel€a, 2009; Morrison, 1996; Yates & Warde, 2015). In meal studies, this is referred to as commensality, which Sobal et al. (2002, p. 378) somewhat bluntly define as “eating with other people,” and Fischler (2011, p. 529) as “eating at the same table.” Commensality is a central part of sociability in everyday life (see the chapters by Higgs and by Holm in this edition). Commensality is usually viewed positively as enhancing interaction and cohesion between the people who share the meal, symbolizing a sense of belonging in the group, and signifying a respect for shared norms (Giacoman, 2016). But it involves both inclusion and exclusion, and manifests equality or hierarchy (Fischler, 2011). Furthermore, the ideal of a nuclear family, that is, married parents and their children, is challenged as families and partnerships are more diverse than ever. Douglas (1972/1997, p. 255) remarks that “the meal puts its frame on the gathering,” that is, the rules that regulate social interaction at the meal are reflected in how the meal itself is organized. Meals bear more social significance than drinks,

The meal as the proper context for food and drinks

199

and are thus more likely to be enjoyed with those we are close with, such as family and friends, rather than strangers. In Douglas’s (1972/1997) terms, distance and intimacy operate in the social organization of the meal. The social aspects and the properness of the meal interact in several ways. First, the meal seems to be experienced as more proper if it is shared with other people (M€akel€a, 2009), and eating alone is disapproved of (Fischler, 2011). Even though eating alone and the number of single households are increasing, meals have social significance, and for many people, a meal without company is not experienced as a “real” meal. But not anyone counts as company. As Sobal et al. (2002, p. 391) note, “proper meals include eating with the proper person (or proper people).” Who such proper people may include is an empirical question, and may vary in different cultural contexts. In Western societies, the most obvious candidates for “proper” meal companions include members of the nuclear family, including the spouse and/or children. Indeed, in the public imagery of the ideal family, all family members assemble together to eat the “family dinner” on a regular basis. The family dinner is an institution the purported demise of which has aroused worry and moralization, although social scientists have pointed out that family dinner is largely a myth, the origin of which doesn’t reach very far in history (e.g., Yates & Warde, 2015). Nevertheless, the social meanings ascribed to family dinner include idealized visions of what a proper family is: the dinner is seen to advance family coherence, conviviality, relaxation after the day’s work, sociability, and for children, learning table manners, etiquette, and new tastes. Eating together is one of the important ways in which the family is made and sustained, and commensality is a key aspect in “family meal scripts” (Blake, Bisogni, Sobal, Jastran, & Devine, 2008, p. 658). Indeed, the family dinner is seen to be archetype of commensality (Mestdag, 2005; Sobal et al., 2002). (The family meal is discussed in more detail in Chapter 11 of this edition.) Sobal et al. (2002) note that while family is one of the key “commensal units” (groups of people who gather together to eat or drink), such units may also include institutional groups, for example, at work, where people habitually go out for lunch together or have a cup of coffee together. Colleagues are thus “proper” people to eat with, too (see also Giacoman, 2016). Commensal units may be extended to “commensal circles,” that is, networks of people eating together on a more or less regular basis. The getting together of commensal circles includes negotiations on the inclusion and exclusion of people, particularly when mixing different circles at, for example, a dinner party (Sobal et al., 2002). The circle may include not only family, colleagues, friends, and relatives, but in formal and structured situations, also strangers who are members of, for example, work-related networks. Second, people often make more effort to prepare a hot meal with proper elements when the meal is to be shared. Hot meals eaten alone have been found to be less complex than shared (hot) meals (Kahma et al., 2014), and when alone, many people more easily settle for sandwiches, fruit, or savory or sweet snacks instead of a full meal. The properness of a meal thus ties together the elements and the company, and a lack of one or the other diminishes the probability of the meal to be experienced as proper (see, e.g., Bugge & Alma˚s, 2006). In their study of newly married or cohabiting

200

Context

couples, Sobal et al. (2002) found that for the spouses interviewed, the ideal was to eat a proper meal whenever they could eat together. The meals that the couples had together were made of more complex ingredients, and included more dishes and were prepared with more care and attention than meals eaten alone. Having someone at home to share the meal with encouraged the informants to have meals that they described as “decent,” “actual,” or “structured,” in contrast to “grabbing,” eating “on the run,” or skipping meals altogether when they were alone. Marshall and Anderson (2002) made similar findings in their study of how proper meals were seen by young married or cohabiting couples. For their informants, eating properly was a metaphor for family life, and the ideal was reflected in both the form and structure of the meal, and the sociability of the event. Those who ate with their partner had more regular and “proper” meals, spent more time at the meal, and ate larger meals. As the preceding examples show, couples and families seem to agree that it is, in many ways, nicer to eat in company than alone, although it has also been found that family dinners are not always convivial and sociable events, but involve many kinds of tensions (e.g., Neumark-Sztainer, Story, Ackard, Moe, & Perry, 2000). However, a British study (Thomas & Emond, 2017) found that elderly people living alone actually valued the solitude of eating alone. While eating, they could watch TV or read the newspaper, which they saw as a form of company or as a way to relax. Their meals at home were in most cases variations of the British proper meal par excellence: “meat and two veg.” Third, the more social and cultural importance there is to the meal, the more attention is paid to the properness of its elements. As Douglas (1972/1997, p. 42) writes, “meals are ordered in scale of importance and grandeur through the week and the year.” Meals at annually repeating festivities, such as birthday parties, Christmas, Easter, or Thanksgiving in the United States, are planned carefully, and their ingredients, the way they are prepared, the setting of the table. and gathering all the relevant people in one’s commensal unit or network around the table are paid considerable attention. Such elements have to go together and to match in order for the meal to be experienced as proper. Studies in Western countries during the past decades have suggested that compared with breakfasts and lunches, evening meals are more complex and structured, and their contents more varied (Yates & Warde, 2015); and while lunches are typically cold in some countries and hot in others, dinner is most commonly hot (e.g., Holm et al., 2015). Actually, as Sobal and Nelson (2003, p. 185) point out, breakfast is a commensal anomaly, as it is often eaten alone. Mestdag (2005, p. 72) concludes that “the meal is still characterized as a social occasion.” Because most meals are eaten at home, the company at meals still most often consists of nuclear family members (Holm et al., 2016; Lhuissier et al., 2013). Not surprisingly, one of the strongest factors impacting commensality is living arrangements, which is why the rising number of one-person households tends to increase eating alone. At the same time, we increasingly eat with people who are not family members (Mestdag, 2005; Mestdag & Glorieux, 2009). New forms of sociability around eating are evolving that transcend the traditional family ties. Today, many restaurants offer a communal table where solo diners can meet other guests and share the meal with other people in a proper context.

The meal as the proper context for food and drinks

9.4

201

Changing and challenging the properness of eating

For some time, it has been speculated that the meanings of proper meals may be changing. Indeed, it would be strange if nothing happened to our food ideals while other spheres of life are in rapid transformation in many ways. For example, changes in work life, digitalization of everyday life, increasing economic opportunities to travel, eat out, and buy more expensive options, and the increase in one-person households all make an impact on how eating is organized as part of the time–space configuration of everyday life. Furthermore, it should be noted that food consumption is still one of the areas where social inequality is concretized both within and between societies. As Yates and Warde (2015, p. 300) have noted, meals continue to “hold enormous social significance, but their attributes, meanings, and other dimensions, such as how they are eaten and with whom, are subject to creeping change.” Such changes are bound to impact ideals of properness in terms of the ingredients used, their level of manufacturing, the dishes made of them, and forms of sociability. In her famous article, Mary Douglas wrote that “a meal incorporates cereals, vegetables, and animal proteins” (Douglas, 1972/1997, p. 41). In the nearly fifty years that have passed since the publication of the article, at least one of these three things has changed: it is no longer as self-evident as before that a meal must include animal protein. Although many people still find meat or fish essential—as shown herein— the necessity of meat and other animal proteins is increasingly questioned. Plant-based protein is no longer seen as only vegetarian or vegan food, but as an option for all those who wish to cut down their meat consumption. Recently, flexitarianism, that is, a diet that is primarily based on vegetables but does not totally exclude consumption of meat as fish, has gained popularity, and there are signs that in some developed countries, meat consumption per capita has reached a peak (Henchion, McCarthy, Resconi, & Troy, 2014). But as we have noted in an earlier article (M€akel€a & Niva, 2016), probably intermediaries such as food services at schools and workplaces are needed to normalize novel alternative proteins in order to make them more palatable in everyday life. In addition, when promoting sustainability, it is evident that there are different meanings of sustainability. People do not necessarily agree about what is “right” and “wrong” food, for whom, in which contexts, and who has the right to define or control what we eat (cf. Kirveennummi, M€akel€a, & Saarimaa, 2013). Due to sustainability concerns, the meanings of a proper meal are bound to start including novel dimensions. In addition to meat and other animal-based food, one of the most important considerations relates to food waste. During the past few years, the role of food waste in all parts of the food system has been acknowledged as one of the challenges that must be solved in order to diminish the environmental load caused by food (e.g., Evans, 2014). From this perspective, could the ideals and practices of a proper meal also take into account the need for avoiding food waste? There are already restaurants using leftover foods from grocery stores as their main ingredients, and applications through which consumers can book leftover lunch food at restaurants and pick it up to eat home in the evening. Although such innovations can reduce food waste only to a limited extent, they can change the image of leftover food and

202

Context

demonstrate the value of food also to home cooks so that instead of ending up as waste and pollutant, leftover food becomes part of proper and respectable meals. Properness has for long been closely related to the ideal of a meal made from scratch. The worry about the decline of a family meal has been paralleled with a concern about a deterioration of cooking skills, and a substitution of ready-made meals for home-made food. The self-made meal is seen as a synonym for a proper meal, because people often believe that home-prepared meals are healthier, more natural, and more authentic than convenience foods, and that by cooking from scratch, one can show love and care for their family members (e.g., Daniels, Glorieux, Minnen, van Tienove, & Weenas, 2015). Because convenience foods are easy and quick to get to the table, they are seen to lack the aspect of care. This is something that the food industry is struggling with when trying to convince consumers about the benefits of their ready-made meals. For example, a Finnish food manufacturer, Saarioinen (https://www.saarioinen.fi/saarioinen/yritys/english-info/), has promoted their convenience foods in commercial campaigns purporting that their products are “food made by mothers,” and featuring pictures of their women employees. The women wearing their work apparel seem like real people—not models or actors—and put forward an image of real, proper mothers making real, proper food for those of us who cannot make it ourselves. The ideal of a mother preparing food for her family is here turned into a mother who also feeds other families—or enters the dining table as a substitute mother. The bringing of the mother and the carer into the marketing of industrially produced food may be interpreted as an effort to transform the image of industrial food to fit in with ideals of properness, but also as a commercial fun-making at the expense of the age-old superior image of home-made food. There are signs that commensality as one of the core elements of the proper meal may be changing in more fundamental ways than only in relation to whom we eat with. Recently we have seen an emergence of solo dining in restaurants, and in some countries, even new restaurants targeting particularly solo diners (Lahad & May, 2017). While an eater without company is by no means an unfamiliar sight in coffee houses, bars, and luncheonettes, having a “proper” evening meal at a restaurant alone is by many people, particularly women, experienced as awkward (Heimtun, 2010). But now solo dining is marketed as an experience for people who wish to concentrate on the flavors and aesthetics of the meal itself, instead of socializing with friends or family. The solo diner may engage in sociability with the restaurant staff, in which case the focus of the interaction is on the meal, the combination of ingredients and their origin, the sequence of dishes, the matching of food and drinks, the ways of preparation, or the beauty of the food on the plate. It is thus possible that solo dining becomes an experience in which the eater concentrates on the palate, and immerses her/himself in new food adventures. Warde, Martens, and Olsen (1999) found in their study on eating out in Britain that those in the upper social classes visited a larger variety of restaurants than others, suggesting that the privileged part of the population used their opportunities for enjoying a variety of taste experiences. We may speculate whether such omnivorousness in eating out might also include an openness to different interpretations of what a proper meal means. If one enjoys very different kinds of cuisines, one

The meal as the proper context for food and drinks

203

might perhaps more easily categorize the meals offered as proper, even if they differed substantially from the customary ones. This is not to say that 20 years after Warde et al.’s study, an openness to different ideas of proper meals would be confined to those in the higher social positions. Rather, we suggest that the proliferation of a variety of restaurants in Western countries that represent and sometimes mix different cuisines, may open up and widen interpretations of properness as regards, for example, how the food elements of the meal are organized synchronically and diachronically.

9.5

Moving contexts: What will be proper in the future?

Several scholars have suggested that the conceptualizations of a (proper) meal are dynamic, and tied in with the social and cultural changes in late modern societies (e.g., Lhuissier et al., 2013; Wise, 2011). Most evident changes in both ideals and practices of a proper meal at home may well be related to the role of cooking from scratch and using convenience foods. In Western countries, the use of pre-prepared ingredients in cooking is today part of everyday life for many home cooks, although it is a practice that seems to require justification and explanation (see Bugge & Alma˚s, 2006). Although cooking from scratch is seen as “optimal cooking” (Lavelle et al. 2016, p. 383), factors such as time pressure, desire for convenience, or family preferences, may form insurmountable barriers for cooking with basic ingredients (see also Szabo, 2011). At the same time, we think it is more uncertain how the ideal of a proper meal as a social and shared event might be affected by actual changes in the sociability of meals during the past few decades. In many countries, we have seen a small, but detectable, increase in eating alone (Holm et al., 2016; Mestdag & Glorieux, 2009), and it may well be that such a change in the practice of commensality also transforms the ideal of eating together as an inherent part of a proper meal. On the other hand, research suggests that although commensality is not always attainable in everyday life, it is still highly valued as a goal (e.g., Fischler, 2011; Sobal et al., 2002; Sobal & Nelson, 2003). As noted herein, sharing does not necessarily mean a family meal at home. According to Traphagan and Brown (2002), in Japan, family dinners are an exception rather than a rule for most families, because the father often works until late in the evening. Due to this, a family meal outside of the home, often in a fast food restaurant, may be a rare and valuable occasion for the family. In Europe, too, it is today more socially acceptable than before for families to spend leisure time at consumption spaces such as cafes and restaurants, and thus continue their pre-children urban lifestyles (Karsten, Kamphuis, & Remeijnse, 2015). Such changes in eating in and eating out as a family probably gradually make ideals and practices of properness more flexible and amenable to variation. The social aspect is now built into the somewhat trite saying “you are what you eat,” and paraphrased as “you are who you eat with” (Sobal & Nelson, 2003, p. 188), and “we are the people we eat with because we share the same food” (Giacoman 2016, p. 466). But even if food is used in efforts to cross cultural

204

Context

borderlines, it isn’t always easy to avoid food racism (Wise, 2011), or anxieties relating to eating culturally novel food. It can, however, build bridges and create new communities, so that meals, proper or not, are a means to facilitate intercultural sociabilities (ibid.). In increasingly multicultural societies, food may be one of the ways to create connections and belonging among people from different ethnic backgrounds. Eating together creates a feeling of community and solidarity among people, and it is through such connectivity that proper food gains new meanings and creates new social ties. In the future, technology may help in supporting such meal sociability, and the experience of enjoying proper meals, even for those living alone. Grevet, Tang, and Mynatt (2012) developed an “eating alone together probe,” a touch screen tablet interface, which a small group of friends living alone used to let each other know when they were making dinner or eating it at home or elsewhere. The experiment showed that the participants got “a pleasant feeling of sociability” just knowing that others were cooking or eating at the same time, even if they did not communicate with each other. Such systems supporting “shared temporality and social presence” (ibid. p. 105) through technological means may be one of the ways to create commensality without physical contact. As noted herein, it is inevitable that what we eat and how much we waste must change in order for the environmentally detrimental effects of eating to diminish. The global production and consumption of animal protein cannot continue growing, and a larger share of food produced must be consumed instead of wasted. Proper meals and proper eating thus become entangled with environmental considerations, possibly in ways we cannot yet foresee. Still it is necessary to analyze developments such as solo dining restaurants and novel technological means to socialize during meals, not only as signs of increasing diversity and choice. Instead of uncritically embracing such phenomena, we should perhaps also ask to what extent such phenomena may indicate increasing polarization of eating. Do people have equal possibilities to make choices about what to eat and with whom, and are we choosing possibilities or necessities? Currently, it seems that new commensal communities around cooking, sharing food, and eating together are being created. Typically, these new types of commensal circles extend the nuclear family and are often characterized by transience: people gather together for sharing food for a better world in pop up happenings that fight against food waste or celebrate the joy of eating together. Such events do not necessarily require “properness” of the meal components or forms sociability in the same sense as “conventional” shared meals, but can perhaps be interpreted as a token of political consumerism, collective responsibility-taking (Stolle & Micheletti, 2013), or alternative hedonism (Soper, 2008). Yet, it seems that commensality is, despite the continuous and apparent changes, an indispensable denominator of a proper meal. It is reasonable to forecast that the commensality of meals is not going to disappear easily. Probably the food on our plates is bound to change more rapidly than the idea of sharing food with other human beings. Even though the category of a meal seems theoretically and empirically identifiable, the definitions of its content, location, and company change over time.

The meal as the proper context for food and drinks

205

References Bildtga˚rd, T. (2010). What it means to “eat well” in France and Sweden. Food and Foodways, 18, 209–232. Blake, C. E., Bisogni, C. A., Sobal, J., Jastran, M., & Devine, C. M. (2008). How adults construct evening meals. Scripts for food choice. Appetite, 51, 654–662. Bugge, A. B., & Alma˚s, R. (2006). Domestic dinner. Representations and practices of a proper meal among young suburban mothers. Journal of Consumer Culture, 6(2), 203–228. Bugge, A., & Døving, R. (2000). Det norske ma˚ltidsmønsteret – ideal og praksis. Forskningsrapport nr. 2. Lysaker (in Norwegian): SIFO. Daniels, S., Glorieux, I., Minnen, J., van Tienove, T. P., & Weenas, D. (2015). Convenience on the menu? A typological conceptualization of family food expenditures and food-related time patterns. Social Science Research, 51, 205–218. Douglas, M. (1972/1997). Deciphering a meal. In C. Counihan & P. van Esterik (Eds.), Food and culture. A reader (pp. 36–54). London and New York: Routledge. Originally published in Daedalus 1972, 101: 61–81. Douglas, M. (1983). Culture and food. In M. Freilich (Ed.), The Pleasures of Anthropology (pp. 74–101). New York: New American Library. Douglas, M., & Nicod, M. (1974). Taking the biscuit. The structure of British meals. New Society, 30(637), 744–747. € Umea˚ Studies in Sociology No 98. Umea˚ (in Swedish): Ekstr€om, M. (1990). Kost, klass och kon. Umea˚ Universitet. Evans, D. (2014). Food waste: home consumption, material culture and everyday life. London & New York: Bloomsbury Academic. Fischler, C. (2011). Commensality, society and culture. Social Science Information, 50, 528–548. Giacoman, C. (2016). The dimensions and role of commensality: A theoretical model drawn from the significance of communal eating among adults in Santiago, Chile. Appetite, 107, 460–470. Grevet, C., Tang, A., & Mynatt, E. (2012). Eating alone, together: New forms of commensality. In Proceedings of the 17th ACM international conference on supporting group work, Sanibel Island, Florida, USA, October 27–31, 2012. Haraldsdo´ttir, J., Holm, L., Jensen, J. H., & Møller, A. (1987). Danskernes kostvaner 1985. 2. Hvem spiser hvad? København (in Danish): Levnedsmiddelstyrelsen. Heimtun, B. (2010). The holiday meal: Eating out alone and mobile emotional geographies. Leisure Studies, 29(2), 175–192. Henchion, M., McCarthy, M., Resconi, V. C., & Troy, D. (2014). Meat consumption: Trends and quality matters. Meat Science, 98(3), 561–568. Holm, L., Lauridsen, D., Lund, T. B., Gronow, J., Niva, M., & M€akel€a, J. (2016). Changes in the social context and conduct of eating in four Nordic countries between 1997 and 2012. Appetite, 103, 358–368. Holm, L., Lauridsen, D. S., Gronow, J., Kahma, N., Kjærnes, U., Lund, T. B., et al. (2015). The € mer food we eat in the Nordic countries – some changes between 1997 and 2012. In Mat ar € mat. Samhallsvetenskapliga € an perspektiv pa€ mat och ma˚ltider. Gothenburg: Gothenburg University. Kahma, N., M€akel€a, J., Niva, M., & Lund, T. B. (2014). Associations between meal complexity and social context in four Nordic countries. Anthropology of Food [Online], S10. http://aof. revues.org/7666.

206

Context

Karsten, L., Kamphuis, A., & Remeijnse, C. (2015). ‘Time-out’ with the family: the shaping of family leisure in the new urban consumption spaces of cafes, bars and restaurants. Leisure Studies, 34(2), 166–181. Kirveennummi, A., M€akel€a, J., & Saarimaa, R. (2013). Beating unsustainability with eating: Four alternative food-consumption scenarios. Sustainability: Science, Practice, & Policy, 9(2), 83–91. Konttinen, H., Sarlio-L€ahteenkorva, S., Silventoinen, K., M€annist€ o, S., & Haukkala, A. (2012). Socio-economic disparities in the consumption of vegetables, fruit and energy-dense foods: The role of motive priorities. Public Health Nutrition, 16(5), 873–882. Lahad, K., & May, V. (2017). Just one? Solo dining, gender and temporal belonging in public spaces. Sociological Research Online, 22(2), 12. Lamont, M. P. (1992). Deciphering a meal again, or the anthropology of taste. Social Science Information, 31(1), 69–86. Lavelle, F., McGowan, L., Spence, M., Caraher, M., Raats, M. M., Hollywood, L., et al. (2016). Barriers and facilitators to cooking from ‘scratch’: A qualitative interview study. Appetite, 107, 383–391. Lhuissier, A., Tichit, C., Caillavet, F., Cardon, P., Masullo, A., Martin-Fernandez, J., et al. (2013). Who still eats three meals a day? Findings from a quantitative survey in the Paris area. Appetite, 63, 59–69. M€akel€a, J., & Niva, M. (2016). Citizens and sustainable culinary cultures. In A. Paloviita & M. J€arvel€a (Eds.), Climate change adaptation and food supply chain management (pp. 171–182). Milton Park, Abingdon, Oxon: Routledge. M€akel€a, J. (1996). Kunnon ateria. P€a€akaupunkiseudun perheellisten naisten k€asityksi€a [A proper meal: exploring the views of women with families]. Sosiologia, 33(1), 12–22 (in Finnish). M€akel€a, J. (2009). Meals: The social perspective. In H. L. Meiselman (Ed.), Meals in science and practice. Interdisciplinary research and business applications. Cambridge: Woodhead Publishing Limited. M€akel€a, J., Kjærnes, U., Ekstr€om, P., Marianne, F., L’Orange, E., Gronow, J., et al. (1999). Nordic meals: Methodological notes on a comparative survey. Appetite, 32, 73–79. Marshall, D. W., & Anderson, A. S. (2002). Proper meals in transition: Young married couples on the nature of eating together. Appetite, 39, 193–206. Mestdag, I. (2005). Disappearance of the traditional meal: Temporal, social and spatial destructuration. Appetite, 45, 62–74. Mestdag, I., & Glorieux, I. (2009). Change and stability in commensality patterns: A comparative analysis of Belgian time-use data from 1966, 1999 and 2004. The Sociological Review, 57, 703–726. Morrison, M. (1996). Sharing food at home and school: Perspectives on commensality. The Sociological Review, 648–674. Murcott, A. (1982). On the social significance of the “cooked dinner” in South Wales. Social Science Information, 21(4–5), 677–696. National Nutrition Council. (2014). Health from food. Finnish nutrition recommendations 2014. Helsinki: National Nutrition Council (in Finnish). Neumark-Sztainer, D., Story, M., Ackard, D., Moe, J., & Perry, C. (2000). The “family meal”: Views of adolescents. Journal of Nutrition Education, 32(6), 329–334. Poulain, J.-P. (2002). The contemporary diet in France: “De-structuration” or from commensalism to “vagabond feeding”. Appetite, 39(1), 43–55. Sobal, J., Bove, C. F., & Rauschenbach, B. (2002). Commensal careers at entry into marriage: Establishing commensal units and managing commensal circles. The Sociological Review, 50(3), 378–397.

The meal as the proper context for food and drinks

207

Sobal, J., & Nelson, M. K. (2003). Commensal eating patterns: A community study. Appetite, 41, 181–190. Soper, K. (2008). Alternative hedonism, cultural theory and the role of aesthetic revisioning. Cultural Studies, 22(5), 567–587. Stolle, D., & Micheletti, M. (2013). Political consumerism. Global responsibility in action. Cambridge Books Online. Available at http://ebooks.cambridge.org/. Szabo, M. (2011). The challenges of “re-engaging with food.” Food, Culture & Society, 14, 547–566. Thomas, N., & Emond, R. (2017). Living alone but eating together: Exploring lunch clubs as a dining out experience. Appetite, 119, 34–40. Traphagan, J. W., & Brown, L. K. (2002). Fast food and intergenerational commensality in Japan: New styles and old patterns. Ethnology, 41(2), 119–134. Warde, A., Martens, L., & Olsen, W. (1999). Consumption and the problem of variety: Cultural omnivorousness, social distinction and dining out. Sociology, 33, 105–127. Wise, A. (2011). Moving food: Gustatory commensality and disjuncture in everyday multiculturalism. New Formations, 74, 82–107. Yates, L., & Warde, A. (2015). The evolving content of meals in Great Britain. Results of a survey in 2012 in comparison with the 1950s. Appetite, 84, 299–308.

This page intentionally left blank

The value of studying laboratory meals

10

France Bellisle Nutritional Epidemiology Research Team (EREN), Paris 13 University, INSERM (U1153), INRA (U1125), CNAM, Bobigny, France

10.1

Introduction

A meal is a physiological, psychological, sensory, and ethological event. In animal species living in societies it is also a social event. While many other adjectives could be added to the list, the recognition that food intake behavior in animals (including humans) is associated with many diverse aspects of an individual’s experience creates very complex demands for any scientific attempt to describe, understand, and possibly explain and control this very basic, life-sustaining event. One critical aspect of food intake behavior is its periodic nature. Even under conditions of constant access to nutrient sources, intake is organized in a series of eating episodes, interspersed by other moments when the animal does not eat. These episodes, mostly occurring during the species-specific circadian activity phase, vary in number, in size, and in duration. In human societies, they also vary in composition, social context, and many other dimensions. Humans like to call their eating episodes “meals” or “snacks.” These terms are highly imprecise (Bellisle, 2014) and their definitions vary in different cultures, let alone different scientific contexts. “Laboratory meals” are specimens of such behavior. They allow the consumption responses of an individual to be examined in a particular context in which significant aspects of the environment are held constant, while others (sensory, social, psychological, etc.) vary under the strict control of the experimenters. Another critical aspect of food intake behavior is that it is essential to life, covering the bodily needs for energy and nutrients, with long-term repercussions on health and body weight control. The frequency of obesity and associated metabolic diseases has reached the proportions of a globalized epidemic. While experts debate the relative contributions of decreasing physical effort in developed societies and changes in the diet, it appears important to test a variety of hypotheses, particularly those involving critical mechanisms, under the strict conditions of the laboratory. A large number of determinants of meal consumption can be manipulated under laboratory settings: time of day, time of week, sensory stimulation, physiological status, ambience, temperature, social facilitation, and so forth. The laboratory is a privileged environment in which to conduct a scientific investigation of behavior under strict, reproducible conditions; to observe behavior repeatedly in the same individuals and study temporal changes or stability; to establish dose/ effect or exposure/effect relationship between causal factors and effect; and to draw Context. https://doi.org/10.1016/B978-0-12-814495-4.00010-6 Copyright © 2019 Elsevier Inc. All rights reserved.

210

Context

parallels between animal observations and human responses to well-defined influences. Tightly controlled laboratory tests offer a high degree of sensitivity and control over the intervention and the outcome measures (Blundell et al., 2010). In the following pages, we will look at the historical roots of the laboratory study of food intake behavior in humans. We will then examine a few characteristics of the laboratory context itself as it developed from early pioneer days to the present. Illustrative examples of the methods and results typical of the laboratory studies will be presented in the examination of the “Satiety Cascade” and the “preload paradigm.” A section will cover the specific demands of the laboratory study of the human appetite. Finally, the important question of the generalization of laboratory observations and other difficulties of the laboratory approach will be discussed.

10.2

A brief historical perspective

The laboratory investigation of appetite was born in the 19th century physiology laboratory. In the mid-19th century, Claude Bernard posed the foundations of the study of experimental medicine (Bernard, 1865) and realized important physiological studies (particularly in the physiology of nutrition) that led to the concept of the “milieu interieur,” the internal environment. In animal species, the constancy of the milieu interieur is an essential condition for survival. Many parameters of this internal status need to be regulated within the narrow limits of what Cannon (1932) later called “homeostasis”: the body temperature and glycemia for example. It soon became evident that while powerful physiological processes come into play to insure regulation, they need to be complemented by the animal’s active behavior in order to complement internal processes. While the pancreas releases hormones to maintain glycemia within regulated limits, food intake has to take place periodically in order to supply the organism with adequate energy and nutrients. Following the scientific investigation of the various internal mechanisms involved in energy and nutrient regulation under strict laboratory conditions, it became evident that research had to extend its domain to behavioral responses that complement internal ones. In this perspective, behavior appears to be a natural extension of regulatory mechanisms. For many heirs of the physiological tradition, the laboratory appeared a logical place to study its role, under the strict rules of the experimental method. The search of physiological factors that affect and even define motivational constructs such as hunger and satiety (Blundell, 1979; Le Magnen, 1992) was at the origin of laboratory research. The nature of the “hunger signal’ that triggers food intake has generated much research. In turn, identifying the laws governing behavior, not only eating but also other regulatory or non-regulatory behaviors, owes a lot to the laboratory investigation of physiological responses associated with eating. The discovery and mechanistic analysis of the conditioned reflex by Pavlov (1927) is the example par excellence. A dog salivates when meat is ingested (an unconditioned response) and also in anticipation of eating, when various characteristics of the environment have become associated with the imminent access to meat (a conditioned response). Conditioned responses such as salivation and other aspects of the “cephalic phase of

The value of studying laboratory meals

211

digestion” (Powley & Berthoud, 1985; Teff, 2011) in turn affect digestive processes, and food motivation in future ingestive episodes. The sensory characteristics of familiar foods themselves become conditioned cues that predict the postingestive effects of consumption. Their predictive value is constantly updated via successive exposures so that they determine a person’s food preferences and motivation at various moments of the lifetime, what recent science calls “liking” and “wanting” (Berridge, 2004; Finlayson, King, & Blundell, 2007). The study of gastric secretions at the time of meals in dogs led to the demonstration of a far reaching mechanism of adaptation to the environment, the conditioned reflex, that applies in numerous areas of life. Subsequent developments in behavioral science used laboratory tests of food consumption in animals or humans to demonstrate and elucidate the mechanisms of instrumental conditioning. From Watson (1926) to Skinner (1938), the demonstration of the laws of instrumental learning, which clearly reach well beyond food consumption responses, were studied in laboratory settings, where independent factors (type, number, duration, frequency, intensity of food rewards) could be demonstrated to elicit measurable and predictable changes in strictly measured, dependent consumption responses (number, frequency, intensity, persistence, etc.). Examining animals in their natural environment confirmed the validity of laws of learning identified from laboratory observations, and showed how species-specific fixed action patterns and environment-specific influences modulated the performance of learned responses (Domjan & Burkhard, 1986). In the history of behavioral science, food intake was viewed as the behavioral mechanism insuring regulation of critical parameters of the internal milieu. Classic theories of food intake control focused on specific regulated parameters. Jean Mayer’s glucostatic theory of eating behavior held that, while the blood glucose level is regulated within narrow limits by hormonal mechanisms, food consumption is triggered periodically by decreases in the rate of use of glucose (Mayer, 1953). The most direct test of the glucostatic hypothesis, the observation of a hungry animal’s or person’s eating behavior following an injection of glucose, can only be performed under strict laboratory conditions. Mayer’s early works have led to the search for glucose sensors in the brain, and of the brain structures that command eating behavior as a response to changes in glucose utilization. Early research identified a “hunger center” and a “satiety center” in the hypothalamus, whose activation/inhibition stimulated or inhibited eating (Anand & Brobeck, 1952; Hetherington & Ranson, 1940; Hoebel & Teitelbaum, 1962; Stellar, 1954). Other physiology-oriented theories of food intake control followed. The lipostatic theory, originally proposed by Kennedy (1953), held that the amount of fat in the body was the regulated parameter that stimulated or inhibited eating in order to maintain the body fat mass constant. The discovery of leptin, the “hormone of satiety” secreted by the adipose tissue and capable of inhibiting food intake, has since brought support to the lipostatic theory (Zhang et al., 1994). From the early days of the regulatory theories up to the present, the notion that eating is triggered and controlled by the fluctuations of physiological parameters in the brain or in the periphery of the body has led to the development of laboratory methods to test the influence of nutrients, hormones, peptides, and so forth on characteristics of eating behavior in the context of a laboratory in which both independent and dependent variables can be measured with precision.

212

Context

10.3

The laboratory context

Studying meals in the laboratory context is a potent method when a precise mechanism of action of a particular factor of interest on food consumption has to be elucidated, allowing a high degree of control. For example, when a dose/response function is to be quantified between the levels of a particular factor (fasting duration, portion size, intensity or diversity of sensory stimulation, etc.) and some aspect of responding (meal size, rate of eating, experience of palatability, etc.) the laboratory context offers the possibility of experimentally manipulating the levels of the factor of interest and quantifying precise changes in the response. It makes it possible also to examine how the consumption response itself affects other aspects of appetite, for example, satiety. In all these situations, clear operational definitions of independent and dependent variables must be used in an experimental protocol designed to test a causal hypothesis. Physiological or sensory effects on consumption can be quantified in this fashion, and also social or affective influences (presence/absence of other persons at the time of eating (Cruwys, Bevelander, & Hermans, 2015); influence of parents, friends, or unfamiliar persons (Hermans, Herman, Larsen, & Engels, 2010; Mennella, Griffin, & Beauchamp, 2004), and environmental conditions (Bellisle & Dalix, 2001). Many important contributions of laboratory work to the understanding of human appetite have been published over the years. Examples of such situations are found in -

-

-

-

-

-

-

studies addressing how specific tastes and flavors influence liking and consumption of foods (Yeomans, 1998) and how these effects are mediated by certain brain peptides (Yeomans & Gray, 2002). the inventory of the numerous factors (physiological, sensory, psychological, social) present at the time of ingestion that affect various aspects of consumption (meal size, ingestion rate, experienced palatability and satiation, etc.) (Blundell, 2017). the observation and assessment of a purely sensory mechanism affecting satiation and satiety, independently of postingestive metabolic effects. “Sensory-specific satiety” was first identified in human laboratory works by Rolls, Rolls, Rowe, and Sweeney (1981), and then studied in many other laboratories (Hetherington, 2013). the measurement of various aspects of the “cephalic phase of digestion” that occurs at the beginning of ingestion, and can exert a variety of effects on later appetite and metabolism (Bellisle, Louis-Sylvestre, Demozay, Blazy, & Le Magnen, 1983; Teff, 2011). the scientific discussion of the satiating value of energy ingested in liquid form versus solid foods (Allison, 2014; Allison & Mattes, 2009). the parallel investigation of the action of certain influences in different species, for example when methods derived from animal work are used in a human study for measuring the reward value of foods and the motivation to consume (Hogenkamp, Shechter, St-Onge, Sclafani, & Kissifeff, 2017). the recent development of the study of brain responses at the time of food stimulation. A recent report showed how neuromodulation directed at the prefrontal cortex of obese individuals decreased their snack food intake and hunger (Heinitz et al., 2017). In this study, transcranial direct current stimulation, a noninvasive technique used to modulate brain activity, modified subjects’ responses in a vending machine paradigm and during snack food taste tests. the investigation of changes in ingestive responses over time in the same individual, by repeated testing in a controlled eating situation. Learning effects, habituation effects, or

The value of studying laboratory meals

-

213

changes occurring in the participant’s mental or physical status can then be quantified. Changes in acceptance of novel flavors in bottle-fed infants, for example, have been followed over several months in laboratory tests (Mennella et al., 2004). Learning, based on previous experience with foods, has been shown to modulate “expected satiety” at the beginning of a meal and affect the amount of food an individual ingests (Yeomans, McCrickerd, Brunstrom, & Chambers, 2014). the optimization of the sensory characteristics of foods and beverages investigated in numerous academic and industry laboratories (for example, in the field of salt effects on palatability, Bolhuis, Lakemond, de Wijk, Luning, & de Graaf, 2010, 2012).

10.3.1 The “Satiety Cascade” Food consumption is a periodic behavior. It is triggered at various moments of the day by a number of converging factors (time of day, need state, sensory stimulation, social context, etc.). As eating progresses, inhibitory influences of many origins (sensory, gastric, hormonal, neural, as well as cognitive) develop and finally bring the meal to an end. Satiation is the complex inhibitory process that integrates these influences and terminates a meal. Satiation determines meal size. After the end of one eating episode, many factors contribute to inhibiting further eating until the next meal. These stimulatory and inhibitory influences were conceptualized as the elements of the “Satiety Cascade” first described over 30 years ago (Blundell, Rogers, & Hill, 1987) and regularly updated since then (Blundell et al., 2010; see Fig. 10.1). The Satiety Cascade integrates sensory, cognitive, postingestive, and postabsorptive factors that inhibit the motivation to eat again for a certain time. Because satiation and satiety have to do with the inhibition of appetite, they are considered potent mechanisms determining total daily energy intake and, on the long-term, body weight control. Studying the causal factors involved in satiation and satiety is therefore central to the understanding of appetite and ingestive responses. Laboratory studies have largely contributed to quantifying the action of many factors at various moments of the satiety cascade (Blundell, 2017).

10.3.2 The preload paradigm In order to investigate satiety influences, laboratory studies of meal intake use highly standardized paradigms that facilitate the quantification of causal relationships. The preload paradigm is one typical methodological approach that has been applied in a large number of human and animal works. In the “preload paradigm,” a load of food is ingested at a fixed interval before a subject (animal or human) has ad libitum access to food (Fig. 10.2). It is therefore possible to measure how the parameters of the load (nutrient composition, energy content, sensory aspects, etc.) will affect ingestive responses. For example, it is possible to see whether any energy or nutrient compensation occurs at the ad libitum meal to adjust for the contents of the preload. When testing human subjects, changes in subjective appetite sensations can be followed in the interval between preload and ad libitum intake. Using a slightly modified version of the paradigm, it is also possible to study the duration of the inhibition of intake induced by a preload: in these

214

Context

Energy balance

Meal quality Expectations Reward/Pleasure Recognition Associations

Stretch Osmotic load CCK GLP-1 PYY Ghrelin

Sensation + prior beliefs and associations

Sensory

FOOD Satiation

Insulin Leptin Adiponectin(?)

Meal quantity

Stomach + intestines

Cognitive

Nutrient status Insulin Oxidation Glucose Amino acids

Fat-free mass Fat mass RMR

Liver + metabolites

Postingestive

Postabsorptive

early

late

Satiety Original Satiety Cascade modified by Mela and Blundell

Fig. 10.1 The satiety cascade. The satiety cascade was originally presented in Blundell et al. (1987). It is periodically up-dated to integrate new scientific findings. Source: Blundell, J. E., de Graaf, K., Hulshoff, T., Jebb, S., Livingstone, B., Lluch, A., et al. (2010). Appetite control: Methodological aspects of the evaluation of food. Obesity Reviews, 11, 251–270.

Preload

Appetite ratings Hunger, fullness, desire to eat, thirst

Ad libitum meal

Variable interval: 15 min to 6 h

Time

Expectation: Subjects ingesting a preload will eat less at next meal

Fig. 10.2 The preload paradigm. Typically, under repeated-measure designs, participants are tested at the same time of day, under identical deprivation conditions. Standardized preloads are ingested. After a variable time delay, the effects on spontaneous food intake are measured. Subjective measures of appetite are often obtained at predetermined time intervals after the preload and/or the test meal. The preload can be overtly or covertly manipulated in order to test the satiating potency of many factors: energy content, energy density, sensory characteristics, nutrient composition, presence of potential satiety-enhancing factors, etc.

The value of studying laboratory meals

215

circumstances, the postmeal interval is not determined by the experimenter, but is allowed to vary until the subject will spontaneously initiate the following meal. Over the years, the preload paradigm has given rise to a substantial inventory of factors that influence the amount of food eaten, and other aspects of satiety (Blundell, 2017), among which -

-

-

The energy density of the preload has a major effect on satiety, for a given energy content (Rolls, Bell, & Waugh, 2000). For an equivalent energy load, a hierarchy of satiating potency exists between the macronutrients: protein induces stronger satiety than carbohydrates, which in turn induce more satiety than fats (Veldhorst et al., 2008). The presence of fiber in a food enhances its satiety effects (Drapeau & Tremblay, 2000; Lluch et al., 2010). Physical phase affects satiety: energy ingested in liquid form elicits less satiety than the same energy load ingested in solid form. This robust observation, confirmed in numerous controlled laboratory studies, suggests a “passive overconsumption” of energy ingested in beverages facilitating weight gain (Almiron-Roig et al., 2013; Mattes, 2006). Various non-nutrients and bioactive food constituents, such as caffeine, enhance satiety (Tremblay & Bellisle, 2015). The importance of texture: viscosity of a beverage contributes to increasing its satiating power (Mattes & Rothacker, 2001).

In the preload paradigm, the temporal succession of events has to be strictly controlled. The preload paradigm requires laboratory conditions in order to yield valid results. It is important, to test how well the results obtained under laboratory settings generalize to the free-living conditions. One very active area of research is the exploration of how satiety effects generalize to long-term, free-living, food consumption, and possibly confirm that satiety enhancement is indeed a means to decrease long-term food intake and beneficially affect body weight control (Tremblay & Bellisle, 2015). Many publications have brought support to this idea. For example, Rolls, Roe, Beach, and Kris-Etherton (2005) provided foods with low energy density to dieters for one year, and observed that weight loss was improved by 50% compared with control.

10.3.3 The laboratory as a microcosm of the eating environment In a laboratory study, the experimental environment reproduces one eating situation where behavior is measured. In animal studies, the laboratory situation cannot possibly reproduce all the characteristics of the wild eating environment. It is highly important, however, to make the laboratory eating environment compatible with the animal’s species-specific action patterns. Among several crucial dimensions, respecting the circadian activity/rest phase is important. Nocturnal animals should be tested during the dark phase of their circadian cycle. Many other ethological factors, such as preferred food texture (a strong acceptance factor in rodents, for example), should be taken into consideration. In human studies, a small fraction of the numerous influences that affect eating in free-living conditions can be imported into the lab. This requirement will always raise

216

Context

legitimate questions about the generalization of the observed responses from the lab to naturalistic conditions. But it will also allow specific influences to be manipulated in controlled ways in order to quantify their impact on specific aspects of the response. This cannot be done with the same degree of control in the free-living environment, where so many influences vary at the same time in unpredictable and unreproducible ways. The human food intake laboratory can take many forms. In its simplest form, it can look like a sensory evaluation booth. Very elementary food stimuli, at times liquid foods of known sensory and nutrient value, can be presented to a human subject, and his/her consumption behavior can be quantified (amount, duration, rate of eating, etc.) Even such basic forms of laboratory environment have yielded interesting information about human appetite. Under such circumstances, it has been shown, for example, that the preferred intensity of sweetness will differ depending on whether the experimental food, let us say a yogurt, is swallowed or not (Lucas & Bellisle, 1987). Other food consumption laboratories can be much more complex. Many laboratories are disguised as small dining rooms or even small restaurants or cafeterias. In preload studies, the participant’s self-selected food choices at the ad libitum test meal are often studied, as well as the energy or nutrient composition. In such situations, a cafeteria-like buffet is served for the subjects to select their preferred food options in a realistic context. The buffet style allows assessment of the potential effect of a preload manipulation on food choices (in terms of preference and/or avoidance, selection of foods based on sensory characteristics and/or energy density, etc.), even if the energy intake at the meal remains constant (Blundell et al., 2010). In recent years, many food consumption laboratories have developed into highly complex environments, replicating plausible eating places for human consumers at certain moments of the day. One example of such sophisticated environments is the Living Lab of the Institut Paul Bocuse, in Lyon, where study participants have their meals in a pleasant restaurant-like environment, while their behaviors are measured using video recordings, and food options are designed to test specific hypotheses about human appetite. Validation studies have shown that the parameters of intake recorded under identical test conditions in this environment are reproducible and sensitive to variations of hunger states (Allirot, Saulais, Disse, Roth, & Cazal, 2012). In most academic laboratories, however, meal menus are not as elaborate as the Bocuse dishes, and merely represent a selection of popular food options (pizza, macaroni and cheese, cookies, candy, and water) (Carnell et al., 2018). These options are not different in the context of a lab or in naturalistic settings. In naturalistic conditions, consumers select foods that have at least middling palatability levels (de Castro, Bellisle, & Dalix, 2000; de Castro, Bellisle, Dalix, & Pearcey, 2000). In the laboratory, most food is usually selected to have at least acceptable palatability, or it is ascertained that the participating subjects will have at least minimal acceptance of the test foods. In some cases, low palatability food stimuli can be used, particularly in tests of the influence of sensory factors on some aspects of the ingestive responses. The circumstances of eating in free-living conditions are highly diverse, and humans in their history have found themselves eating in all sorts of social and material conditions where the eater has limited control over his/her environment. The laboratory is one more such situation. The experimenter selects a sub-set of factors to

The value of studying laboratory meals

217

examine their influence on meal eating. Although the laboratory can be a very simplified environment, it can also integrate a large number of variables of interest: not only types and characteristics of foods, but also time of day (or week or season), social environment (eating alone or in company), ambiance factors, and characteristics of the consumers (age, sex, ethnicity, motivational attitudes such as dietary restraint or emotional eating, etc.). Whatever the level of sophistication, care must be taken to keep important influences under strict control: -

the social context is important, so the laboratory should make sure that no unwanted social influences impinge on the consumer’s responses during the meal; attention to the meal is important, so potential distractors should be strictly controlled; temperature and other ambiance factors should be constant, particularly in repeated testing situations.

10.3.4 The human meal as an independent or a dependent variable The laboratory offers a controlled environment in which the numerous influences that affect meal intake can be studied; it also allows an investigation of how meal intake itself can affect other aspects of appetite or behavior. In other words, in a laboratory context, meal intake can be studied as an independent or as a dependent variable. For example, meal size and eating rate can be affected by factors such as palatability or texture (Bellisle & Le Magnen, 1980). Conversely, eating rate at meal time can affect postprandial satiety effects (Ferriday et al., 2015). Some ambitious protocols use food intake under laboratory conditions as both an independent and a dependent variables. For example, a recent study looked at morning and afternoon appetite in obese individuals with or without binge eating disorder (Carnell et al., 2018). After an 8 h fast, the participants first received a standardized liquid meal of fixed composition, then a stress test. Access to an ad libitum buffet was allowed 2 h 40 min after the liquid load. Appetite and stress were monitored using rating scales, and blood was drawn for hormone measures. This study illustrated why and how the afternoon/evening represented a high-risk period for overeating, particularly in individuals with binge eating disorder exposed to stress.

10.4

The demands of laboratory testing of human food consumption

The study of human meal intake under laboratory conditions is not a novel area of research. Over the years, researchers have developed specific tools and gained expertise from repeated testing in laboratory conditions. The emerging picture is that the laboratory is a highly controlled environment to gain access to some of the determining influences of human appetite, but also a very demanding one. A number of review papers have been published about the methodological demands associated with the

218

Context

study of human appetite under laboratory settings (for example Blundell et al., 2009, 2010; Chapelot, 2013).

10.4.1 The design of the laboratory test meal The value of data obtained under laboratory conditions will critically depend on the hypothesis tested, and the adequacy of the protocol. Operational definitions of the critical variables should be established at the outset of the project. It is important to insure adequate statistical power with a sufficient number of participants, and an adequate statistical analysis plan that will permit significant effects to be identified. The validity (internal and external) and reliability of the critical variables should guide the elaboration of the testing situation. The food stimuli and the conditions of their presentation should be thoroughly considered, as well as the time sequence of all relevant events taking place in the experimental situation. A special attention should be paid to potential confounders, which are numerous when considering human behavior in general, and human eating behavior in particular (time of testing, appropriateness of the food stimuli, individual characteristics such as body weight status or psychological attitudes, including dietary restraint, etc.).

10.4.2 The demands of preload studies The preload paradigm has benefited from the cumulative experience of research teams working on satiety or other aspects of postingestive appetite. Detailed advice for efficient standardization has been published (Blundell et al., 2010; Chapelot, 2013; Kissileff, 1985). Standardization of preloads in terms of energy content, macronutrient composition, energy density, physical state (solid vs. liquid), weight or volume, and sensory and cognitive characteristics is important, as well as the full characterization of the test situation: conditions of the ad libitum meal (single food versus buffet, social surroundings, etc.), time sequence of preload and test meal administration, and the nature and duration of the appetite measurements between preload and test meal, or even following the test meal (Blundell et al., 2010).

10.4.3 Assessing subjective motivation associated with meals The use of self-report tools for the assessment of subjective appetite prior to, during, or following a laboratory meal is also a frequent procedure used in laboratory settings. These tools have developed from paper-and-pencil instruments to hand-held electronic devices. They typically address such aspects of motivation as experienced hunger or fullness at specific times relative to ingestive events. Many formats have been standardized and validated over the years. The most frequently used is the horizontal visual analogue scale marked at both ends with anchors expressing the extremes of a given sensation (hunger or fullness, for example). These validated tools provide data with good sensitivity, reliability, and validity (Flint, Raben, Blundell, & Astrup,

The value of studying laboratory meals

219

2000), especially when appetite ratings are expressed over several hours as area under the curve (AUC), rather than absolute value of discrete points.

10.4.4 The laboratory as a nonnatural environment The laboratory testing of meal intake in human subjects by no means implies that the circumstances of the test have to be “artificial” relative to free-living food consumption. The foods served do not have to be “artificial” foods. Of course, simplified or elementary food stimuli have to be used under certain circumstances (such as liquid foods ingested from a straw), but they can also be anything from cocktail-size sandwiches to elaborate dishes. Many satiety and functional claims have been tested under laboratory conditions using novel foods, or novel recipes of familiar foods, designed by the industry for the purpose of the test (see, for example, the yogurt studies by Lluch et al., 2010). The measuring instruments can be wearable sensors designed to be as light and non-obtrusive as possible. Considerable progress has been made in this area over the years (Bellisle & Le Magnen, 1980; Fontana et al., 2014). Studies used discrete cameras that scan a dining room, or hidden scales that continuously weigh the amount of food being ingested without any interference with the subject’s behavior. The experimental room where intake behavior takes place does not need to look like a stern sensory evaluation booth or a hospital room. It can be decorated as a restaurant or cafeteria typical of places where people frequently have their “free-living” meals. Recent works have developed the notion of a “simulated context,” a naturalistic consumption situation reproduced under laboratory conditions (Holthuysen, Vrijhof, De Wijk, & Kremer, 2017) as a cost-effective procedure that combines increased experimental control with the realism of the simulated context. Various means can be used in order to evoke a realistic meal situation in the laboratory, among which is the use of recent technological advancements in virtual reality or augmented reality, which can create increasingly realistic and immersive environments (Zandstra & Lion, 2018).

10.5

Limitations of the laboratory approach

10.5.1 External validity issues The study of human behavior under laboratory conditions is often criticized for being “artificial” and non-representative of spontaneous, free-living responses. Questions about external validity are legitimate, but are by no means limited to laboratory testing. Just because a behavior has been observed out of the laboratory does not guarantee that it generalizes to all other free-living circumstances. A lot can be done to increase the “naturalness” of the test laboratory and, as long as this is compatible with the demands of the protocol, quite a few of the characteristics of a naturalistic eating environment can be imported into the laboratory. The external validity of laboratory findings remains an important question that requires adequate testing under the

220

Context

relevant free-living conditions. It is highly important, for example, to determine whether the potent satiety effects that have been observed under laboratory conditions will be reproduced in free-living consumers, and whether the enhancement of satiety will be sufficient to affect long-term appetite and body weight control (Bellisle & Tremblay, 2011).

10.5.2 The ‘Hawthorne effect’ One obvious limitation of studying human behavior under laboratory conditions is the potential loss of spontaneity and associated distortions. The “Hawthorne effect,” the propensity to change one’s behavior as a result of being observed, is inherent in all scientific studies of human behavior, whether in the field or laboratory. Naturalistic observation conditions can, at times, make the observation procedures less salient than they are in controlled laboratory environments. In some circumstances, however, the observation tools clearly affect the observed behavior. Under naturalistic conditions, the simple fact of keeping a food diary makes people eat less; in behavioral therapy programs for weight loss, the food diary is used not only for dietary assessment, but also as a tool to decrease intake (Butryn, Webb, & Wadden, 2011). Likewise, an individual whose behavior is observed under laboratory conditions, and who is fully aware that his/her behavior is being observed, can consciously or unconsciously modify his/her behavior to please the experimenter, or to displease the experimenter, or simply to give the best possible image of him/herself. Again, this is not only true of laboratory conditions, but the salience of measurement procedures under laboratory settings could amplify the distortion.

10.5.3 Expectations and demands of the experimental settings Scientific research requires approval by ethical committees and, in human studies, participants must be fully informed from the outset of what is expected from them in a particular experimental test. Although this is true in every human experiment, informing participants of the nature of the test without influencing participants’ behavior is often a major challenge. Laboratory testing can create expectations, perhaps worries, and consequent distortions in the behavior. Experimenters should remain aware of these potential sources of bias, and try to minimize their effect from the elaboration of the protocol to the final interpretation of the results. One strategy that can be used, although it has substantial costs, is to have the participants go through the experimental procedure once, without any testing being done, before the start of the actual experiment.

10.5.4 Duration limits The duration of any laboratory test of behavior is obviously limited. Human subjects, adults or children, cannot possibly remain under laboratory conditions for more than a few hours or a few days. Most laboratory tests of human food intake last for a few hours: the time required to record the actual meal consumption, plus a few minutes

The value of studying laboratory meals

221

prior to the meal (for example in preload studies) and/or up to a couple of hours after the meal (to study the early parts of the satiety cascade). While the information gathered from such studies has been immense, they cover only a small part of the feeding circumstances in a human life. In order to increase the control over important variables, many researchers extend the observation or experimentation phase to longer durations. Typically, participants arrive at the laboratory early in the morning after a night fast, have breakfast, remain under controlled conditions during the morning, have a preload, then are served a test meal 1–2 h later, and remain in the laboratory for a while after lunch. Each of these events can be manipulated for experimental purposes, and/or can be selected for the measurement of some critical aspect of the response. One example is found in a test of a hand-held electronic data capture method for the continuous monitoring of subjective appetite sensations: participants arrived at the lab prior to breakfast, in the fasted state, and then had a fixed breakfast; an ad libitum lunch was served 4 h after breakfast; and the participants’ appetite sensations were rated every thirty minutes until after lunch termination (Gibbons, Caudwell, Finlayson, King, & Blundell, 2011). Studies of the adaptation of energy intake to changing conditions of the food supply can be continued over several days in subjects housed in wards where their food intake can be observed and measured. Food intake behavior over 24 days was examined in response to the covert 25% decrease in energy density of the diet, showing a slow, progressive adaptation of spontaneous consumption (Porikos, Hesser, & Van Itallie, 1982). In a study of genetic factors affecting body weight, twins were maintained for 100 days in a metabolic ward where they were overfed by 1000 kcal a day (Bouchard et al., 1990; Bouchard, Tchernof, & Tremblay, 2014): in these circumstances, food intake was used as a controlled independent variable whose effects were studied on weight changes and associated metabolic responses.

10.6

Conclusions and consideration for future developments

The laboratory meal is a highly useful context for studying several aspects of appetite control in human consumers. Its high internal validity, the reliability of validated tools, the control over independent variables, and the precision of operational definitions are a few of its merits. It simplifies the array of factors susceptible to act on the behavior of interest, and it is optimally used over short periods of time. It has produced large amounts of precious information about the determinants of food intake. Particularly, biological factors affecting hunger and satiety have been characterized, and causal links have been quantified. The value of the laboratory context for the investigation of human appetite depends on the aims of the hypothesis being tested and the quality of techniques used by the experimenter. Improvements in methodology have occurred over time to address some of the limitations classically associated with laboratory studies of human behavior. The foods served and the circumstances of eating can often be very close to what they would be under every day eating conditions. Important laboratory works have been published

222

Context

using representative foods, served in realistic eating places, including social interactions, all under the strict experimental control allowed by the laboratory. Over the years, laboratory meals have become increasingly similar to free living eating events as experimenters became more informed and skillful. In appetite research, compromises have to be made about the requirements for internal and external validity, between precision and naturalness (Blundell et al., 2010). There is no absolute border with laboratory science on one side and free living observation on the other. Many features of the free living environment can be imported into the appetite lab, and measurement instruments initially developed in the lab context can be exported to free living research contexts. While the optimal protocol is likely to vary depending on the hypothesis and the type of factors and mechanism at play, overlapping laboratory and free-living protocols can often be used in a variety of contexts in order to address appetite from complementary perspectives.

References Allirot, X., Saulais, L., Disse, E., Roth, H., & Cazal, C. (2012). Validation of a buffet meal design in an experimental restaurant. Appetite, 58, 889–897. Allison, D. A. (2014). Liquid calories, energy compensation, and weight: what we know and what we still need to learn. British Journal of Nutrition, 111, 384–386. Allison, D. A., & Mattes, R. D. (2009). Nutritively sweetened beverage consumption and obesity. Journal of the American Medical Association, 301, 318–320. Almiron-Roig, E., Palla, L., Guest, K., Ricchiuti, C., Vint, N., Jebb, S. A., et al. (2013). Factors that determine energy compensation: A systematic review of preload studies. Nutrition Reviews, 71, 458–473. Anand, B. K., & Brobeck, J. R. (1952). Food intake and spontaneous activity of rats with lesions in the amygdaloid nuclei. Journal of Neurophysiology, 15, 421–430. Bellisle, F. (2014). Meals and snacking, diet quality and energy balance. Physiology & Behavior, 134, 38–43. Bellisle, F., & Dalix, A. M. (2001). Cognitive restraint can be offset by distraction, leading to increased meal intake in adult women. American Journal of Clinical Nutrition, 74, 197–200. Bellisle, F., & Le Magnen, J. (1980). The analysis of human feeding patterns: The edogram. Appetite, 1, 141–150. Bellisle, F., Louis-Sylvestre, J., Demozay, F., Blazy, D., & Le Magnen, J. (1983). Reflex insulin response associated to food intake in human subjects. Physiology & Behavior, 31, 515–521. Bellisle, F., & Tremblay, A. (2011). Satiety and body weight control. Promise and compromise. Comment on ‘Satiety, No way to slim’. Appetite, 57, 769–771. Bernard, C. (1865). Introduction a` l’ etude de la m edecine exp erimentale. Paris: Baillie`re. Berridge, K. C. (2004). Motivational concepts in behavioral neuroscience. Physiology & Behavior, 81, 179–209. Blundell, J. E. (1979). Hunger, appetite and satiety – Constructs in search of identities. In M. Turner (Ed.), Nutrition and lifestyles (pp. 21–42). London: Applied Sciences Pub Ltd Proceedings of the British Nutrition Foundation. First annual conference. Blundell, J. E. (2017). The contribution of behavioural science to nutrition: Appetite control. Nutrition Bulletin, 42, 236–245.

The value of studying laboratory meals

223

Blundell, J. E., de Graaf, C., Finlayson, G., Halford, J. C. G., Hetherington, M., King, N., et al. (2009). Measuring food intake, hunger, satiety, and satiation in the laboratory. In D. A. Allison & M. L. Baskin (Eds.), Handbook of assessment methods for eating behaviors and weightrelated problems (pp. 283–325). Thousand Oaks, USA: Sage Publications (Chapter 8). Blundell, J. E., de Graaf, K., Hulshoff, T., Jebb, S., Livingstone, B., Lluch, A., et al. (2010). Appetite control: Methodological aspects of the evaluation of food. Obesity Reviews, 11, 251–270. Blundell, J. E., Rogers, P. J., & Hill, A. J. (1987). Evaluation the satiating power of foods: Implications for acceptance and consumption. In J. Solms, D. A. Booth, R. M. Pangbourne, & O. Raunhardt (Eds.), Food acceptance and nutrition (pp. 205–219). London: Academic Press. Bolhuis, D. P., Lakemond, C. M., de Wijk, R. A., Luning, P. A., & de Graaf, C. (2010). Effect of salt intensity on ad libitum intake of tomato soup similar in palatability and on salt preference after consumption. Chemical Senses, 35, 789–799. Bolhuis, D. P., Lakemond, C. M., de Wijk, R. A., Luning, P. A., & de Graaf, C. (2012). Effect of salt intensity in soup on ad libitum intake and on subsequent food choice. Appetite, 58, 48–55. Bouchard, C., Tchernof, A., & Tremblay, A. (2014). Predictors of body composition and body energy changes in response to chronic overfeeding. International Journal of Obesity, 38, 26–242. Bouchard, C., Tremblay, A., Despres, J. P., Nadeau, A., Lupien, P. J., Theriault, G., et al. (1990). The response to long-term overfeeding in identical twins. New England Journal of Medicine, 322, 1477–1482. Butryn, M. L., Webb, V., & Wadden, T. A. (2011). Behavioral treatment of obesity. Psychiatric Clinic of North America, 34, 841–859. Cannon, W. B. (1932). The wisdom of the body. New York: WW Norton & Company Inc. 294 p. Carnell, S., Grillot, C., Ungredda, T., Ellis, S., Mehta, N., Holst, J., et al. (2018). Morning and afternoon appetite and gut hormone responses to meal and stress challenges in obese individuals with and without binge eating disorder. International Journal of Obesity, 42, 841–849. Chapelot, D. (2013). Quantifying satiation and satiety. In J. E. Blundell & F. Bellisle (Eds.), Satiation, satiety and the control of food intake (pp. 12–39). Cambridge: Woodhead Publishing. Cruwys, T., Bevelander, K. E., & Hermans, R. C. J. (2015). Social modeling of eating: A review of when and why social influence affects food intake and choice. Appetite, 86, 3–18. de Castro, J. M., Bellisle, F., & Dalix, A. M. (2000). Palatability and intake relationships in freeliving humans: Measurement and characterization in the French. Physiology & Behavior, 68, 271–277. de Castro, J. M., Bellisle, F., Dalix, A. M., & Pearcey, S. M. (2000). Palatability and intake relationships in free-living humans: Characterization and independence of influence in north Americans. Physiology & Behavior, 70, 343–350. Domjan, M., & Burkhard, B. (1986). The principles of learning and behavior. Belmont: Brooks/ Cole Publishing Company. Drapeau, V., & Tremblay, A. (2000). Diet and body weight reduction. In D. H. Lockwood & T. G. Heffner (Eds.), Obesity: pathology and therapy (pp. 237–258). Berlin: Springer. Ferriday, D., Bosworth, M. L., Lai, S., Godinot, N., Martin, N., Martin, A. A., et al. (2015). Effects of eating rate on satiety: A role for episodic memory? Physiology & Behavior, 152, 389–396. Finlayson, G., King, N., & Blundell, J. E. (2007). Liking vs. wanting food: Importance for human appetite control and weight regulation. Neuroscience & Biobehavioral Reviews, 31, 987–1002.

224

Context

Flint, A., Raben, A., Blundell, J. E., & Astrup, A. (2000). Reproducibility, power, and validity of visual analogue scales in assessment of appetite sensations in single test meal studies. International Journal of Obesity, 24, 38–48. Fontana, J. M., Higgins, J. A., Schuckers, S. C., Bellisle, F., Zhaoxing, P., Melanson, E. L., et al. (2014). Energy intake estimation from counts of chews and swallows. Appetite, 85, 14–21. Gibbons, C., Caudwell, P., Finlayson, G., King, N., & Blundell, J. E. (2011). Validation of a new hand-held electronic data capture method for continuous monitoring of subjective appetite sensations. International Journal of Behavioral Nutrition and Physical Activity, 8, 57. https://doi.org/10.1186/1479-5868-8-57. Heinitz, S., Reinhardt, M., Piaggi, P., Weise, C. M., Diaz, E., Stinson, E. J., et al. (2017). Neuromodulation directed at the prefrontal cortex of subjects with obesity reduces snack food intake and hunger in randomized trial. American Journal of Clinical Nutrition, 106, 1347–1357. Hermans, R. C. J., Herman, C. P., Larsen, J. K., & Engels, R. C. M. E. (2010). Social modeling effects on snack intake among young men. The role of hunger. Appetite, 54, 378–383. Hetherington, A. W., & Ranson, S. W. (1940). Hypothalamic lesions and adiposity in the rat. Anatomical Record, 78, 149–172. Hetherington, M. (2013). Sensory-specific satiation and satiety. In J. E. Blundell & F. Bellisle (Eds.), Satiation, satiety and the control of food intake (pp. 253–269). Cambridge: Woodhead Publishing. Hoebel, B. G., & Teitelbaum, P. (1962). Hypothalamic control of feeding and self-stimulation. Science, 149, 452–453. Hogenkamp, P. S., Shechter, A., St-Onge, M.-P., Sclafani, A., & Kissifeff, H. R. (2017). A sipometer for measuring motivation to consume and reward value of foods and beverages in humans: Description and proof of principle. Physiology & Behavior, 171, 216–227. Holthuysen, N. T. E., Vrijhof, M. N., De Wijk, R. A., & Kremer, S. (2017). Welcome on board: Overall liking and just-about-right ratings of airplane meals in three different consumption contexts – Laboratory, re-created airplane, and actual airplane. Journal of Sensory Studies, 32, e12254. Kennedy, G. C. (1953). The role of depot fat in the hypothalamic control of food intake in the rat. Proceedings of the Royal Society B, 140, 578–592. Kissileff, H. R. (1985). Effects of physical state (liquid-solid) of foods on food intake: Procedural and substantive contributions. American Journal of Clinical Nutrition, 42, 956–965. Le Magnen, J. (1992). Neurobiology of feeding and nutrition. San Diego: Academic Press. Lluch, A., Hanet-Geisen, N., Salah, S., Slas-Salvado, J., L’Heureux-Bouron, D., & Halford, J. C. G. (2010). Short-term appetite-reducing effects of a low-fat dairy product enriched with protein and fibre. Food Quality & Preference, 21, 402–409. Lucas, F., & Bellisle, F. (1987). The measurement of food preferences in humans: Do taste-andspit tests predict consumption? Physiology & Behavior, 39, 739–744. Mattes, R. D. (2006). Beverages and positive energy balance: The menace is the medium. International Journal of Obesity (London), 30, S60–S65. Mattes, R. D., & Rothacker, D. (2001). Beverage viscosity is inversely related to postprandial hunger in humans. Physiology & Behavior, 74, 551–557. Mayer, J. (1953). Glucostatic mechanisms of regulation of food intake. New England Journal of Medicine, 249, 13–16. Mennella, J. A., Griffin, C. E., & Beauchamp, G. K. (2004). Flavor programming during infancy. Pediatrics, 113, 840–845. Pavlov, I. (1927). Conditioned reflexes. Oxford: Oxford University Press.

The value of studying laboratory meals

225

Porikos, K. P., Hesser, M. F., & Van Itallie, T. B. (1982). Caloric regulation in normal-weight men maintained on a palatable diet of conventional foods. Physiology & Behavior, 29, 293–300. Powley, T. L., & Berthoud, H. R. (1985). Diet and cephalic phase insulin responses. American Journal of Clinical Nutrition, 42, 991–1002. Rolls, B. J., Bell, E. A., & Waugh, B. A. (2000). Increasing the volume of a food by incorporating air affects satiety in men. American Journal of Clinical Nutrition, 72, 361–368. Rolls, B. J., Roe, L. S., Beach, A. M., & Kris-Etherton, P. M. (2005). Provision of foods differing in energy density affects long-term weight loss. Obesity Research, 13, 1052–1060. Rolls, B. J., Rolls, E. T., Rowe, E. A., & Sweeney, K. (1981). Sensory specific satiety in man. Physiology & Behavior, 27, 137–142.0. Skinner, B. F. (1938). The behavior of organisms. New York: Appleton-Century-Crofts. Stellar, E. (1954). The physiology of motivation. Psychological Reviews, 61, 5–22. Teff, K. L. (2011). How neural mediation of anticipatory and compensatory insulin release helps us tolerate food. Physiology & Behavior, 103, 44–50. Tremblay, A., & Bellisle, F. (2015). Nutrients, satiety, and control of energy intake. Applied Physiology, Nutrition, and Metabolism, 40, 971–979. Veldhorst, M., Smeets, A., Soenen, S., Hochstenbach-Waelen, A., Hursel, R., Diepvens, K., et al. (2008). Protein-induced satiety: Effects and mechanisms of different proteins. Physiology & Behavior, 94, 300–307. Watson, J. B. (1926). Behaviorism: A psychology based on reflexes. Archives of Neurology and Psychiatry, 15, 185–204. Yeomans, M. R. (1998). Taste, palatability and the control of appetite. Proceedings of the Nutrition Society, 57, 609–615. Yeomans, M. R., & Gray, R. X. (2002). Opioid peptides and the control of human ingestive behaviour. Neuroscience and Biobehavioral Reviews, 26, 713–728. Yeomans, M. R., McCrickerd, K., Brunstrom, J. M., & Chambers, L. (2014). Effects of repeated consumption on sensory-enhanced satiety. British Journal of Nutrition, 111, 1137–1144. Zandstra, E. H., & Lion, R. (2018). In home testing. In H. Meiselman (Ed.), Context: The effects of environment on product design and evaluation. San Diego, USA: Elsevier. Zhang, Y., Proenca, R., Maffei, M., Barone, M., Leopold, I., & Friedman, J. M. (1994). Positional cloning of the mouse obese gene and its human homologue. Nature, 372, 425–432.

Further reading Blundell, J. E., & Bellisle, F. (Eds.), (2013). Satiation, satiety and the control of food intake. Cambridge: Woodhead Publishing. French, J. R. P. (1953). Experiments in field settings. In L. Festinger & D. Katz (Eds.), Research methods in the behavioral sciences (pp. 98–135). New York: Holt, Rinehart & Winston (Chapter 3). Sclafani, A. (2004). Oral and postoral determinants of food reward. Physiology & Behavior, 81, 773–779.

This page intentionally left blank

Are family meals declining? The example of Denmark

11

Lotte Holm, Thomas Bøker Lund Department of Food and Resource Economics (IFRO), University of Copenhagen, Frederiksberg, Denmark

11.1

Introduction

The fate of family meals has been a recurrent theme in public debate and in research for several decades. While in the late 1990’s a decline in family meals was often anticipated, empirical research has shown that the fear that families have stopped eating together is not warranted, at least in a European context (Davidson & Gauthier, 2010; Fjellstrom, 2009; Jarosz, 2017). While it is debated to what extent the universality of the family meal was ever a historical reality (Cinotto, 2007; Ekstr€om, 1990; Murcott, 2013), there is less doubt that the family meal is an idea that continues to be of monumental significance, sometimes described as a romanticized myth created to moralize about modern eating patterns (Sobal, Bove, & Rauschenbach, 2002) or modern family life (Murcott, 2013). Still, research shows that also in current modern households, family meals are reported to be viewed as a priority (Brannen, O’Connell, & Mooney, 2013), and regardless of family form, eating together at home in daily routine meals is seen as a hallmark of group membership ( Julier, 2013). Family meals may take many forms, but in current research, family meals are generally defined as those occasions when food is eaten simultaneously in the same location by more than one or all family members (Martin-Biggers et al., 2014). The significance of family meals is a theme in the research literature within social science, evolutionary and developmental studies, and in public health. In this chapter, we present how family meals are linked to discussions of social behavior, to child socialization, and to public health in the broadest sense. Even though the practice of assembling family members on a regular basis for a shared meal has not vanished, there may have been changes in the frequency of, and the way family meals are practiced. This could potentially imply that the importance ascribed to family meals has changed too. We use Denmark as a case to discuss current changes in the frequency, organization, and conduct of family meals in modern life. Denmark is a small affluent society, which, together with Finland, Norway, and Sweden, represents what has been labeled a “social-democratic welfare society’ (Esping-Andersen, 1990) with universal access to health services and education, and where female participation in the workforce has been high since the late 1960s (Rasmussen & Brunnbech, 2017). Denmark, therefore, is a relevant case for studying family meals in the context of modern households where women are active in the labor market. Households are relatively small in Context. https://doi.org/10.1016/B978-0-12-814495-4.00011-8 Copyright © 2019 Elsevier Inc. All rights reserved.

228

Context

Denmark, and more than two generations living together is extremely rare. Thus, the Danish case is about family meals in nuclear families.

11.2

The significance of family meals

In social research, family meals are linked to theories of commensality, which is seen as an essential dimension of the common meal that finds it’s most salient expression in its daily social occurrence (Fischler, 2011). Commensality creates bonds between those sharing their food, and meals regulate social life and individual behavior, and they contribute to socializing individuals into specific rules involving cooperation (Fischler, 2011). Numerous studies highlight the significance of sharing meals for family cohesion. Family meals are seen as the medium through which the family unit is produced and reproduced on a daily basis (DeVault, 1991), and they are, in many households, viewed as a priority in spite of practical obstacles in daily life (Brannen et al., 2013; Bugge & Døving, 2000; Ekstr€ om, 1990; Green et al., 2009; Iversen & Holm, 1999; Jansson, 1993). It has been suggested that this is because family meals are linked to conventions and norms that facilitate intimacy and collective identity (Gutierrez, Price, & Arnould, 2008), and the regularity of family meals serves as a predictable and stable aspect of family life that helps to promote security within the family (Davidson & Gauthier, 2010). Family meals are daily sites for socialization of children (Ochs & Shohet, 2006). Through participation in meals and through mealtime communication, children learn their own position relative to others, they are socialized into culturally divergent symbolic, moral, and emotional meanings associated with food and eating, and they learn about ways of acting, thinking, and feeling in the world (Ochs & Shohet, 2006). Social interaction and conversation are often highlighted as the important factor that ensures the significance of family meals for family cohesion and child socialization. However, it has been maintained that the food itself is an important feature. For family meals to be arenas for moral socialization, they need to be events where foods are shared, that is, participants need to eat from the same pot. When several people share the food, issues of fairness and morality become important: each individual needs to make sure that enough is left for the others to be satisfied, and this is central for how family meals are arenas for learning pro-social behavior. This aspect is thought to be more decisive than whether or not people eat at the same time: even if meal participants eat at different points in time, they still need to exert the selfcontrol necessary to ensure that the food is shared in a fair manner (De Backer et al., 2014). In a similar vein, others argue that the benefits of having a family meal can be undermined if the family consumes fast food (Martin-Biggers et al., 2014), which is often available in individually packed, ready-made meals (Ochs & Beck, 2013). In public health literature, empirical studies highlight positive effects of eating family meals, especially for children and adolescents. Thus, positive associations are reported between frequent family meals and overall diet quality (Fulkerson

Are family meals declining? The example of Denmark

229

et al., 2014; Woodruff & Hanning, 2008), and suggested for lower risk of obesity (Hammons & Fiese, 2011; Martin-Biggers et al., 2014; Valdes et al., 2013). While the evidence for the link between family meals and obesity is still inconsistent and weak (Casazza et al., 2015; Martin-Biggers et al., 2014; Valdes et al., 2013), family meals have become an important component of health promotion for children (Fruh et al., 2011, p. 235), including childhood obesity prevention in the US (Rao, 2008). Further, frequency of family meals is found to be inversely associated with teen substance abuse (tobacco, alcohol, and drugs), and with disordered eating practices, and positively associated with good language skills, academic performance, and family cohesion and connectedness (Martin-Biggers et al., 2014). In the public health literature, mechanisms behind the positive impacts of family meals are rarely discussed, but TV watching during family meals is seen to impede the positive impacts of family meals (Feldman et al., 2007; Fiates, Amboni, & Teixeira, 2008; Fuller-Tyszkiewicz et al., 2012; Liang, Kuhle, & Veugelers, 2009). This suggests then, that also in public health, social interaction and conversation are seen as crucial for the positive effects of family meals. In all, in research, family meals are assigned a significant role in health, in children’s literacy, language experience, and school performance, in the cohesion of families, in the socialization of younger generations not only into cultural norms and values related to eating, but also into pro-social behavior and different ways of acting, feeling, and thinking in the world. It is worth underlining that, as the practice of having family meals is quite widespread, it is not possible to decide whether the virtues and benefits assigned to the family meal are, in fact, linked to ordinary family life in general, and not to the family meal as such. Still, the question of whether or not the practice of having family meals change is of interest, whether or not it is seen as the central factor for child socialization, health, and social cohesion, or merely a an indication of ordinary family life.

11.3

Family meals in decline?

Whether or not family meals are in decline has for decades been a topic on public agendas in many countries, but studies confirm that in daily life, the idea of sharing meals on a daily basis with family members is, and has still been, important, not only in Denmark and other Nordic countries (Bugge & Døving, 2000; Ekstr€om, 1990; Iversen & Holm, 1999; Jansson, 1988), but also in other European countries (De Backer et al., 2014; Jarosz, 2017; Suggs et al., 2018) and in other parts of the world (Giacoman, 2016; Jingxiong et al., 2007; Sato et al., 2016; Trofholz et al., 2018; Utter et al., 2013). At the same time, several authors have questioned whether family meals were ever a regularly practiced and stable event, as often anticipated in public debate. Thus, Ekstr€ om underlined that the shared meal was not, historically speaking, shared by everyone, as not everybody belonged to the sharing group. Thus, women, servants, and children each had a different social status and “everybody did not eat from the same pot” (Ekstr€ om, 1990, p. 75). Murcott proposes a similar argument in her discussion of how family forms and household structures have varied

230

Context

considerably between social classes, leaving considerable variation with respect to, for example, whether women or children were part of family meals (Murcott, 2013), and in a small case study of households in early 20th century West Yorkshire, Jackson and colleagues found great variations in whether and how families ate together ( Jackson, Olive, & Smith, 2009). Murcott suggests that the ideal of the family that is expressed in worry about the decline of the family meal is probably an “ideal-typical model of the middle-class and (respectable) working-class family” (Murcott, 2013). Cinotto shows in a historical account of family meals in the US that current ideals of family meals refer to a brief history of middle-class practices in the 1950’s. She suggests that media representations of family life from this period is the reason why these specific practices have come to epitomize traditional family meals (Cinotto, 2007). Parallel to this, Jackson argues that the family meal is a venerated social institution, a myth we live by, more an ideal than a daily practice. Thus, the persistent idea that family meals are on the decline is a case of moral panic, a proliferation of poorlyinformed public debate with disregard for solid evidence ( Jackson et al., 2009). Time-use studies provide data that are suited to analyze how the prevalence of family meals develops over time in more recent history. But to our knowledge, there are only a few studies addressing this question, and results are equivocal, because studies are conducted in different national settings, and at different time periods. Mestag and Vandeweyer found that in Belgium over a thirty-year period from the mid-1960s to 1999, time spent eating with family went down by almost half—from 51 to 27 min per day (Mestdag & Vandeweyer, 2005), but from 1988 to 1999, there were no signs of further destructuration of the social organization of eating, and the nuclear family was still the most important commensal unit (Mestdag, 2005). Cheng and colleagues found that in the United Kingdom, time spent eating in the home declined between 1975 and 2000, but as the duration of eating episodes was stable, the authors interpreted that eating took place in the company of others, and concluded that the temporal pattern of eating and drinking at home remained remarkably stable between 1975 and 2000. However, no data about the company of eating episodes were presented in this study (Cheng et al., 2007). On the basis of these two studies, it is not possible to draw any conclusions about current developments in family eating, and there is thus a lack of more recent data and comprehensive analyses.

11.4

Social organization of family meals

In addition to the question of whether family meal eating is a current and prevalent phenomenon is the question of whether the form of such meals has changed. It has been suggested that codes of conduct and rules of etiquette have undergone a process of informalization, according to which, norms and rules have become more lenient and less formal, and a recent Nordic study confirmed identified trends supporting this ideal (Holm et al., 2016). But is this also the case for family meals? Evidently, the processes of socialization that are supposed to take place at family meals would be impacted if new practices with little or no conversation between family members at meals is emerging.

Are family meals declining? The example of Denmark

231

There are strong, and most likely idealized, images of what a family meal should look like: All family members should be present, they should sit around a table, eating the food that is prepared by the woman in the family. They should engage in friendly conversation about events in their daily lives and matters of the world, and they should enjoy the emotional bonds between family members that are created and confirmed during this daily encounter. Already Bourdieu, in his analysis of festive meals in the French working class and bourgeoisie of the 1960s, showed that the codes of conduct at working class meals were often structured differently. The timing of eating would not necessarily be synchronized between those sharing the meal, youngsters may appear later at the table, and young children would perhaps eat their dessert in front of the television (Bourdieu, 1984). While this analysis does not relate to ordinary everyday meals, newer studies indicate that current family meals often display similar lenient codes of conduct. Thus, Brannen and colleagues found that in dual-earner British families, shared family meals would often take place with the adults sitting on the sofa, the children on the floor, and everybody enjoying TV programs (Brannen et al., 2013). In some American middle-class families, eating in separate rooms, at different times, and eating different foods catering to individual tastes have been found to be characteristics of family meals, a practice that is supported by the plethora of individually packed convenience foods that can be found in American supermarkets, and stocked in large quantities in households (Ochs & Beck, 2013). It is a stable phenomenon in most research about family meals that the responsibility for arranging family meals is assigned to women (Wood, 1995), and providing loving care to family members through food and family meals have been described as central in feminine identity (F€ urst, 1997). However, overall, the gendering of cooking has changed in recent years (Holm et al., 2015), and the question is whether responsibility for familly meals is now increasingly shared between women and men. While research confirms that ideas about the organization of family meals is quite persistent ( Jackson, 2009), there are qualitative studies that indicate that realities in modern life are quite different (Ochs & Beck, 2013). Large-scale representative population studies may shed more light on the practical realities of family life. In the following, data from one such study will be analyzed.

11.5

Aim of analysis

We analyze family meals in Denmark in 1997 and 2012. We focus on how typical such meals were, how they were distributed across populations, and aspects of the organization and conduct of such meals. Families eat different kinds of meals together, and in research, definitions of family meals vary with respect to which kind of meals and how many family members need be present (Martin-Biggers et al., 2014). In sociological studies, the term “family meal” often refers to cooked meals eaten in the evening, when family members have returned home from their daily activities outside the household. In studies, this meal is often viewed as the main meal of the day in terms of time spent preparing and eating it,

232

Context

the degree of ceremony and value and the importance assigned to the meal in terms of gathering family members (DeVault, 1991; Fjellstrom, 2009; Julier, 2013). In the present study, we adopt a rather strict definition of family meals, closely related to the ideals of main meals in families highlighted in social research. We are interested in family meals as a context for child socialization, family cohesion, and social interaction, and we have therefore chosen to focus on events that are shared by all household members. Thus, we operationalize family meals as cooked meals (hot meals is the term used in our study) that are eaten at home in multi-person households with all family members present and eating at the same time. We analyze the frequency of having family meals in 1997 and 2012, and how they distribute across socio-demographic groups. We also investigate whether or how the content and conduct of family meals have changed over the fifteen years between our two studies. We analyze whether key aspects of meals, which have been shown to change between 1997 and 2012, also changed for family meals. Thus, we ask whether the gendering of cooking family meals has changed (Holm et al., 2015), and whether the duration, sitting arrangements, and conduct of family meals have changed (Holm et al., 2016), including whether families shared the same food, and whether they watched TV while having family meals (De Backer et al., 2014; Martin-Biggers et al., 2014).

11.6

The data

The analysis is based on data from a large project about meal patterns in Nordic populations. The project included two cross-sectional surveys conducted across four Nordic countries in 1997 and 2012, respectively, using largely identical questionnaires. Prompted in part by an interest in a possible de-structuration and individualization of everyday eating, the studies were designed to register, in detail, one day of eating (the day before the interview) in representative samples of the Nordic populations (Holm et al., 2012). Data were collected during a one-week time frame, in both years a week at the end of April where there were no national specific “special days” (such as holidays) in any of the countries, and equal distribution of interviews across days of the week was ensured. The data in the 1997 study (N ¼ 4823) were collected using computer-assisted interviews (CATI), while the 2012 study (N ¼ 8248) used an internet-based questionnaire. Study populations in both years were individuals between 15 and 80, and in both years there are minor variations from country-specific census data on gender, region, and age, and the lowest-educated population segment (i.e. individuals having completed only compulsory school or short-term specialization courses) is, in general, under-represented. Despite this, we chose to work with an unweighted data sample in the present analysis, because we used a subsample only of individuals living in multi-person households (N ¼ 758 and N ¼ 1417 in 1997 and 2012, respectively). The main part of the questionnaire focused on respondents’ eating events the day before the interview. Questions about each eating event were posed in chronological order, to a maximum of ten events. The questions focused on number, timing, and structure of eating events, as well as social context in terms of where and with whom

Are family meals declining? The example of Denmark

233

the meal was eaten (M€akel€a et al., 1999). Finally, if meals were eaten at home in the company of others, respondents were asked who had cooked the meal. Questions about eating events on the previous day were followed by enquiries about respondents’ socio-demographic background, and about their attitudes toward food and eating-related issues. The data analyses employed (which include chi2 tests, Spearman’s correlation coefficients, logistic regression, and predicted probabilities from logistic regression) will be described in the relevant passages of the results section.

11.7

Results

11.7.1 Have family meals become more or less frequent from 1997 to 2012? The first analysis showed that there was no significant change in the frequency with which Danish people ate family meals over the fifteen years of study. Thus, in 1997, 61% had them on weekend days, and 64% on weekdays, and in 2012, the frequencies were 63% and 65%, respectively. Small increases in frequency are suggested, but as they are insignificant, results show that practicing family meals is a stable phenomenon in Denmark. By means of a logistic regression analysis of socio-demographic variations in eating family meals, we found that age and household composition were significantly related to having family meals in Denmark. Next, we therefore take a closer look at how different combinations of these two variables relate to family meals. In Fig. 11.1, the probability of having a family meal across different theoretically defined

Fig. 11.1 Family meals and life-phase groups 1997 and 2012. Prevalence (in percent) of family meal according to age and household composition (with or without either small or teenage children) (N ¼ 1997:658 and 2012:1327).

234

Context

life phase groups is presented. Results are based on predicted probabilities from logistic regression conducted separately for each year. Judging by the patterns observed in Fig. 11.1, the frequency of having family meals is influenced by a combination of family composition and age, which again reflects phases of life. No group is below 50% and no group is above 75% in frequency of family meals. The fewest family meals are found among younger households without children. When small children (aged 0–13 years) enter the picture, family meals are more frequent, and then they decline again in households with teenage children (14 years or older), only to increase somewhat in the older households without children, and even more so in the eldest households. The figure also shows that while there is no change in the households with small children, there are small increases in all other household types between the two years. As stated earlier, the differences are not significant, however.

11.7.2 The conduct of family meals We now take a closer look at more details of family meals. First, we turn to the question as to whether or not families actually shared family meals, that is, shared the same food, or, whether they ate different foods, accommodating individual tastes and preferences. We then focus on the arrangement and conduct of eating, and who cooked the meal. The importance of sharing the same food at family meals has been highlighted both from a theoretical point of view (De Backer et al., 2014), and based on empirical studies of the difficulties with arranging family meals in American households where a plethora of individually packed, ready made meals are available (Ochs & Beck, 2013). Following an interest in this, we asked in the 1997 survey whether, at family meals, “anyone was served specific food” and in 2012 “Did everybody eat the same food?” The two questions are not identical, but both address the issue of sharing food versus accommodating individual preferences by serving individualized meals. In 1997, only 12% of family meals had individual preferences accommodated, while at 88%, all family members ate the same food. In 2012, this had increased to 94%. The change was (P < .001). In terms of arrangement of the food, family meals in Denmark thus represent the ideal of sharing the same food among all family members, and over the fifteen year period between the two surveys, this pattern has been strengthened. Next, we focus on the conduct of eating in terms of whether or not parallel activities went on while eating, and the duration of eating events, the setting of the eating in terms of whether or not the meal took place at a dining or a sofa table, and finally, the social organization of eating in terms of the gendering of cooking family meals. Fig. 11.2 shows changes in arrangements of meals and conduct of eating between 1997 and 2012. The figure shows that in Denmark, the gendering of cooking family meals has changed. There is a large and marked decline in women cooking such dinners alone (from 69% to 50% of dinners), and a considerable increase in men cooking family meals (from 19% to 28%). Family meals cooked by men and women together increased also, as did meals cooked by others. Further, family meals increasingly took

Are family meals declining? The example of Denmark

235

Fig. 11.2 Family meals in Denmark, 1997 and 2012: Average percentages of arrangements of meals and conduct of eating (Activities while eating*; Duration of eating, Place of eating; Gendering of cooking).

place at sofa tables instead of at proper dining tables. Most typically in both years, family meals would last between 21 and 30 min, but during the fifteen years between the two studies, long family meals (41 min or longer) have become much more rare in Denmark (from 32% to 7%), and the shorter family meals (20 min or less) have become markedly more frequent (from 11% to 27%). Also, there has been a large increase in family meals taking place while watching television, while listening to the radio, and reading, while eating has gone down. Finally, family meals with no such parallel activities have become less frequent in 2012 than in 1997 (from 64% to 57%). Still, it is noteworthy that the majority have family meals in which the meal itself appears to be the main activity. To learn whether there are socio-demographic differences insofar as the family meal characteristics outlined in Fig. 11.2 are concerned, we conducted logistic regression analysis using generalized estimating equations to take into account that some households had more than one family meal (data not shown). Many of these characteristics do not vary systematically between social groups in the population. However, watching television while eating family meals, and variables linked to this (eating on a sofa and not at a dinner table) is systematically related to age, in that the younger the age group, the more frequent is this practice. Watching television while enjoying family meals diverges from the ideal image of the family meal. We will therefore take a closer look at this activity, by analyzing the frequency of eating family meals while watching television among the life-phase groups outlined earlier. See Fig. 11.3, where the probability of watching TV during a family meal is presented across life phases based on predicted probabilities from logistic regression conducted separately for each year.

236

Context

Fig. 11.3 Family meals while watching TV and life-phase groups 1997 and 2012. Prevalence (in percent) of watching TV during the family meal according to age and household composition (with or without either small or teenage children). (N ¼ 1997:509 and 2012:971).

Fig. 11.3 shows that watching TV during the family meals varies a lot between life phases, from a little over 10% to almost 60% of family meals. In both years, prevalence is highest in the youngest age group, where there are no children, after which it drops considerably when/if children enter the household. Further, the figure shows that the changes between the two years are not the same in the life phases. There is a marked raise in the youngest age group without children. While in 1997, 39% of this life phase group watched TV during family meals, 58% did so in 2012. TV became more common also at family meals in families with teenage children, the frequency in this group almost tripled over the fifteen years, among families aged 49–59 with no children, it almost doubled, and among the oldest age group, watching TV while having family meals more than doubled. For households with small children, the pattern of change is remarkably different. Here there was a decrease in TV at family meals from 20% to 13%.

11.8

Discussion

In this analysis, we defined family meals as hot meals eaten in the home with all household members present. We found that, overall, the frequency of eating family meals in Denmark was very stable over the fifteen years between 1997 and 2012. This finding is in contrast to concerns expressed both in public debate and research that family meals are in decline, but there are not many empirical studies of change with which we can compare our findings. A study of changes in family meals in Belgium from 1960 to 1999 based on time-use data (Mestdag & Vandeweyer, 2005) shows a marked decline in family meals, but the time period is different from ours, and whether or not a similar decline took place in Denmark in that same period is not known. A study from the United Kingdom from 1975 to 2000 suggests that family meals

Are family meals declining? The example of Denmark

237

may not have declined in the period from 1975 to 2000, but most likely increasingly moved to locations outside the household, but no clear evidence is given as to the social company of meals (Cheng et al., 2007). To our knowledge, no other studies address change in family meals in periods similar to ours. Studies from other countries are thus necessary to decide whether the stability of having family meals is a Danish phenomenon, or, whether it is a more general trend. In Denmark, the frequency of having family meals was related to life phase. We found a general pattern in both years according to which the frequency of family meals is low among young households without children, increases when small children enter the picture, decreases again when children are older (teenagers), only to increase again when children have left the households, and to reach it’s peak among the oldest households. Thus, family meals are least popular among the young. The pattern suggests that the presence of small children in households promotes the arrangement of family meals, and that such meals are maintained in older households without children. When children reach teenage hood it appears that the accommodation of family members’ individual activities becomes more important. This confirms findings from qualitative research that suggests that teenagers’ individual activities are prioritized over family meals, as the socialization of teenagers into becoming autonomous subjects is considered vital among Danish parents (Iversen & Holm, 1999). The analysis showed clear signs that the organization and codes of conduct for family meals changed in the period. From 1997 to 2012, the gendering of cooking family meals changed, as more men did so, and fewer women. Thus, it appears that the caring for family members through food and meals is a commitment that increasingly is shared by men and women. Further, the data shows that as family meals became shorter, they tended to move from the proper dinner table to the sofa, and to increasingly take place in parallel to watching television, and decreasingly while listening to the radio. The sharing of food, and the conversation taking place while doing so, is in much research seen as crucial for the significance of family meals. Our results point in different directions. First, it appears that family meals in Denmark are meals where food is shared, not served as individualized menus accommodating personal preferences. This underlines that Danish family meals have the potential of serving as arenas for teaching children pro-social behavior. On one hand, it appears that television is becoming an element in family meals, and this may hamper meal conversation and the building of family cohesion. However, this needs not be the case, as empirical studies have shown that some families manage to talk and watch television at the same time, and eating in other places than around a dinner table, may signal intimacy and a break with hierarchy that may promote family cohesion (Scagliusi et al., 2016). Still, it is worth underlining that our results show that in families with small children, TV is not a frequent guest at family meals. Our analysis is quite unique in that it builds on data that address the habits and practicalities around daily eating. It thus gives more direct and detailed information about eating practices, than is the case with studies based on time-use. Compared with analyses based on time-use data, our study stands out in suggesting that family meals are stable and important elements of everyday eating in Denmark, and, that unlike other

238

Context

studies, we do not find any signs of decline. Still, the character of family meals appears to be changing, toward more lenient and informal codes of conduct. This is in line with findings that, in general, Nordic meal patterns appear to have become more informal (Holm et al., 2016). The present analysis shows that this tendency is characteristic also of family meals. It transpires then, that while the significance and value of having family meals that have been highlighted in qualitative studies appear to reflect actual practices in the Danish population, the specific way of organizing and orchestrating such meals does seem to be changing. The consequences of these changes are not unequivocal. Future detailed studies of meals in families are needed to learn more about this.

Acknowledgments The analysis presented in this chapter was based on two Nordic studies: A Day of Food in Nordic Countries, A comparative investigation of eating habits in modern everyday life (1997), and Food in Nordic Everyday Life: A comparative survey of change and stability in eating patterns (2012). Both studies were supported financially by the Joint Committee for Nordic Research Councils for the Social Sciences (NOS-S) and for the humanities and the Social Sciences (NOS-HS). The study designs were devised together by the authors and project partners: Marianne Pipping Ekstr€om, University of Gothenburg; Jukka Gronow, Johanna M€akel€a and Mari Niva, University of Helsinki; Unni Kjærnes, National Institute for Consumer Research, Oslo. Drude Lauridsen and Jukka Gronow are thanked for their contribution to analysis and for valuable comments on the manuscript at various stages of its preparation.

References Bourdieu, P. (1984). Distinction. A social critique of the Judgement of taste. London: Routledge & Kegan Paul. Brannen, J., O’Connell, R., & Mooney, A. (2013). Families, meals and synchronicity: Eating together in British dual earner families. Community, Work & Family, 16(4), 417–434. Bugge, A., & Døving, R. (2000). Det norske ma˚ltidsmønsteret - Ideal og praksis The Norwegian mealpattern - ideal and practice Lysaker: Statens Institutt for Forbruksforskning. Casazza, K., et al. (2015). Weighing the evidence of common beliefs in obesity research. Critical Reviews in Food Science and Nutrition, 55(14), 2014–2053. Cheng, S.-L., et al. (2007). The changing practice of eating: Evidence from UK time diaries, 1975 and 2000. The British Journal of Sociology, 58(1), 39–61. Cinotto, S. (2007). “Everyone would be around the table”: American family mealtimes in historical perspective, 1850-1960. New Directions for Child and Adolescent Development, 115, 83–97. Davidson, R., & Gauthier, A. H. (2010). A cross-national multi-level study of family meals. International Journal of Comparative Sociology, 51(5), 349–365. De Backer, C. J. S., et al. (2014). “Our” food versus “my” food. Investigating the relation between childhood shared food practices and adult prosocial behavior in Belgium. Appetite, 84, 54–60. DeVault, M. L. (1991). Feeding the family. The social organization of caring as gendered work. Chicago: University of Chicago Press. € kost klass och kon € (1st ed.). Umea˚: Sociologiska Institutionen. Ekstr€om, M. (1990). Ekstrom

Are family meals declining? The example of Denmark

239

Esping-Andersen, G. (1990). The three worlds of welfare capitalism. Cambridge: Polity Press. Feldman, S., et al. (2007). Associations between watching TV during family meals and dietary intake among adolescents. Journal of Nutrition Education and Behavior, 39(5), 257–263. Fiates, G. M. R., Amboni, R. D. M. C., & Teixeira, E. (2008). Television use and food choices of children: Qualitative approach. Appetite, 50(1), 12–18. Fischler, C. (2011). Commensality, society and culture. Social Science Information, 50(3–4), 528–548. Fjellstrom, C. (2009). The family meal in Europe. In H. Meiselman (Ed.), Meals in science and practice (pp. 219–235). Oxford: Woodhead Publishing Limited. Fruh, S. M., et al. (2011). The surprising benefits of the family meal. The Journal for Nurse Practitioners, 7(1), 18–22. Fulkerson, J. A., et al. (2014). A review of associations between family or shared meal frequency and dietary and weight status outcomes across the lifespan. Journal of Nutrition Education and Behavior, 46(1), 2–19. Fuller-Tyszkiewicz, M., et al. (2012). The associations between TV viewing, food intake, and BMI. A prospective analysis of data from the longitudinal study of Australian children. Appetite, 59(3), 945–948. F€ urst, E. L. (1997). Cooking and femininity. Women’s Studies International Forum, 20(3), 441–449. Giacoman, C. (2016). The dimensions and role of commensality: A theoretical model drawn from the significance of communal eating among adults in Santiago, Chile. Appetite, 107. Green, T., et al. (2009). Making healthy families? In P. Jackson (Ed.), Changing families, changing food (pp. 205–225). Houndsmills: Palgrave Macmillan. Gutierrez, K., Price, L., & Arnould, E. J. (2008). Consuming family dinner time. In Advances in consumer research. . Hammons, A. J., & Fiese, B. H. (2011). Is frequency of shared family meals related to the nutritional health of children and adolescents? Pediatrics, 127(6), e1565–e1574. Holm, L., et al. (2012). The modernisation of Nordic eating studying changes and stabilities in eating patterns. Anthropology of Food, S7, 2–14. Holm, L., et al. (2015). Who is cooking dinner? Food, Culture & Society, 18(4), 589–610. Holm, L., et al. (2016). Changes in the social context and conduct of eating in four Nordic countries between 1997 and 2012. Appetite, 103, 358–368. Iversen, T., & Holm, L. (1999). Ma˚ltider som familieskabelse og frisættelse [Meals and the making of family - and individualisation]. Tidsskriftet Antropologi, (39), 53–64. Jackson, P. (2009). Introduction food as a lens on family life. In P. Jackson (Ed.), Changing families, changing food (pp. 1–16). Houndmills: Palgrave Macmillan. Jackson, P., Olive, S., & Smith, G. (2009). Myths of the family meal: Re-reading Edwardian life histories. In P. Jackson (Ed.), Changing families, changing food (pp. 131–145). Basingstoke: Palgrave Macmillan. € 40(Suppl 2), 1–203. Jansson, S. (1988). Maten och myterna [Food and myths]. Va˚r Foda, Jansson, S. (1993). Food practices and social interaction: The case of becoming a family. In: Paper for: Current research into eating practices, Potsdam, pp. 1–8. Jarosz, E. (2017). Class and eating: Family meals in Britain. Appetite, 116, 527–535. Jingxiong, J., et al. (2007). Influence of grandparents on eating behaviors of young children in Chinese three-generation families. Appetite, 48(3), 377–383. Julier, A. P. (2013). Meals: “Eating in” and “Eating out.”. In P. Murcott, A. Belasco, & W. Jackson (Eds.), The handbook of food research (pp. 338–351). London: Bloomsbury Academic.

240

Context

Liang, T., Kuhle, S., & Veugelers, P. J. (2009). Nutrition and body weights of Canadian children watching television and eating while watching television. Public Health Nutrition, 12(12), 2457–2463. M€akel€a, J., et al. (1999). Nordic meals: Methodological notes on a comparative survey. Appetite, 32, 73–79. Martin-Biggers, J., et al. (2014). Come and get it ! A discussion of family mealtime literature and factors affecting obesity risk. Advances in Nutrition, 5, 235–247. Mestdag, I. (2005). Disappearance of the traditional meal: Temporal, social and spatial destructuration. Appetite, 45, 62–74. Mestdag, I., & Vandeweyer, J. (2005). Where has family time gone? In search of joint family activities and the role of the family meal in 1966 and 1999. Journal of Family History, 30 (3), 304–323. Murcott, A. (2013). Family meals: A thing of the past? In P. Caplan (Ed.), Food, health and identity (pp. 32–49). London: Routledge. Ochs, E., & Beck, M. (2013). Dinner. In T. Kremer-Sadlik & E. Ochs (Eds.), Fast-forward family: Home, work, and relationships in middle-class America. Oakland (pp. 48–66). University of California Press. Ochs, E., & Shohet, M. (2006). The cultural structuring of mealtime socialization. New Directions for Child and Adolescent Development, (111), 35–49. Rao, G. (2008). Childhood obesity: Highlights of AMA expert committee recommendations. American Family Physician, 78(1), 56–63. Rasmussen, S. H., & Brunnbech, P. Y. (2017). Familieliv og kvindefrigørelse (Family life and women’s liberation). Danmarkshistorien.dk. Available at. http://danmarkshistorien.dk/ historiske-perioder/kold-krig-og-velfaerdsstat-1945-1973/familieliv-og-kvindefrigoerelse/. Accessed 22 December 2017. Sato, P. d. M., et al. (2016). Family meals and eating practices among mothers in Santos, Brazil: A population-based study. Appetite, 103, 38–44. Scagliusi, F. B., et al. (2016). Eating at the table, on the couch and in bed: An exploration of different locus of commensality in the discourses of Brazilian working mothers. Appetite, 103, 80–86. Sobal, J., Bove, C., & Rauschenbach, B. S. (2002). Commensal careers at entry into marriage: Establishing commensal units and managing commensal circles. The Sociological Review, 50(3), 378–397. Suggs, L. S., et al. (2018). Is it better at home with my family? The effects of people and place on children’s eating behavior. Appetite, 121, 111–118. Trofholz, A. C., et al. (2018). Family meals then and now: A qualitative investigation of intergenerational transmission of family meal practices in a racially/ethnically diverse and immigrant population. Appetite, 121, 163–172. Utter, J., et al. (2013). Family meals among New Zealand young people: Relationships with eating behaviors and body mass index. Journal of Nutrition Education and Behavior, 45(1), 3–11. Valdes, J., et al. (2013). Frequency of family meals and childhood overweight: A systematic review. Pediatric Obesity, 8(1), e1–e13. Wood, R. C. (1995). The sociology of the meal. Edinburgh: Edinburgh University Press. Woodruff, S. J., & Hanning, R. M. (2008). A review of family meal influence on adolescents’ dietary intake. Canadian Journal of Dietetic Practice and Research, 69(1), 14–22.

Studying natural meals: What are the benefits of the living lab approach?

12

` s Giboreau* Anestis Dougkas*, Laure Saulais*,†, Agne *Institut Paul Bocuse Research Centre, Ecully, France, †Department of Agri-food Economics and Consumer Science, Laval University, Quebec, QC, Canada

12.1

Introduction

This chapter discusses the importance of studying natural meals and of considering the role of context when studying food consumers. It also introduces Living Lab approaches, where researchers base their data collection on an existing eating environment, and have as many relevant contextual factors as possible—such as social (e.g., number and type of guests at the table), environmental (e.g., atmospherics, space, time), and personal (e.g., recruited consumers on precise targeted criteria)— fixed and controlled. Eating behavior encompasses all the factors that determine what is eaten, through the field of food choice, and how much is eaten, through the field of appetite regulation (Blundell, 2017). The behavioral science of food choice considers, primarily, factors such as geography, religion, ethnicity, and economics, among others (Emilien & Hollis, 2017), and less frequently, biological factors. The latter has been linked to homeostatic and hedonic principles of regulating the amount of food eaten, nutrients ingested, and consequently, their effects on physiology, metabolism, and health (Blundell, 2017). As Rozin (1998) stated “Because behavior is so central to nutrition, the behavioral sciences play an especially important role in the understanding of what we eat and why we eat it. The study of what is in food is extremely important, but all of this knowledge amounts to little if we cannot persuade people to eat what is good for them and to avoid what will do them harm.” The study of appetite control and satiety is crucial in understanding food intake. Briefly, satiety is the feeling of fullness that persists after eating, and influences the period of time between eating episodes (frequency of meals), while satiation is the process that causes one to stop eating, and determines the amount of food consumed at each eating episode (meal size) (Benelam, 2009). The “satiety cascade” (Blundell, Rogers, & Hill, 1987) (see Bellisle’s chapter in this book) includes both psychological processes such as cognition, beliefs, and expectations, and physiological processes, such as release of hormones from the stomach and gastrointestinal tract (Blundell et al., 2010). In addition, sensory aspects are involved in hedonic appetite control related to the influence of reward, pleasure, and palatability Context. https://doi.org/10.1016/B978-0-12-814495-4.00012-X Copyright © 2019 Elsevier Inc. All rights reserved.

242

Context

while eating (Finlayson, King, & Blundell, 2007). The majority of people in Western societies eat according to social rhythms, and for pleasure. In many cases in developed countries, the pleasurable taste of food influences food preferences and consumption, illustrating the considerable functional overlap between homeostatic and hedonic processes in the control of food intake (Finlayson et al., 2007). However, eating behavior cannot be restricted to nutritional issues (i.e., ingestion of sufficient nutrients to satisfy biological requirements) and sensory issues (i.e., pleasure due to enjoyment of sensory properties of food). A psycho-affective dimension associated with the “context” of the food intake is also involved (Divert et al., 2015; Rozin & Tuorila, 1993). A definition of the “context as the set of events and experiences that are not part of the reference event (i.e., eating a target food) but have some relationship to it” was proposed by Rozin and Tuorila (1993). Meiselman, Johnson, Reeve, and Crouch (2000) further described context (environment or situation) as all the variables in an eating occasion, such as the food quality, service, price, decoration, and the type of individuals expected to be found in an eating environment. The location refers to the physical characteristics of size, space, color, noise, light, brightness, and so forth, of the eating environment (Meiselman et al., 2000). The context in which the food choice is made (i.e., temporal, social, and physical surroundings), the food or product of interest (i.e., intrinsic/extrinsic properties), and the eater or person-related factors (i.e., cultural, emotional, psychological, physiological) are three main elements related to food experience. They influence food preferences, acceptability, enjoyment of foods, duration of meals, and food intake (K€oster, 2009; Meiselman, 2006; Wansink, 2010). While the factors related to the food product and eater have been the focus of research in the past few decades, the context or environment and its impact on nutrition, appetite, and food intake has remained a relatively understudied area, but has been gaining more attention recently. The growing interest in food context in various scientific fields could contribute to a better understanding of the relationship between the various factors at play in eating behavior.

12.2

Why is it important to take context into account in food research?

The influences of context on food acceptance, consumer perception, food intake, and overall eating behavior have been highlighted by several authors (de Graaf et al., 2005; Meiselman, 2006; Rozin & Tuorila, 1993; Weber, King, & Meiselman, 2004). This has led people to consider context to increase the validity and predictive ability of models of consumers’ judgment, decision-making, and behavior within the applied goal of assessing the feasibility of specific food product development (de Andrade, Giongo, de Barcellos, & Deliza, 2017; Korzen & Lassen, 2010). Beyond the general acknowledgement of the need to integrate context in the models of food research, the theoretical perspectives on how such an integration should be conducted differ across disciplines, depending on whether the focus is on food perception, consumption, or food choice.

Studying natural meals: What are the benefits of the living lab approach?

243

12.2.1 The importance of context in food perception and intake research In the field of food perception research, context studies have vastly focused on the empirical identification of factors influencing the measurement of perception. Several dimensions of context have been investigated in studies comparing the effects of variations of context, with a greater focus on social and physical aspects. Social factors are among the contextual influences that have been empirically found to impact the measures of food liking and intake (see the review of social factors by Higgs, Ruddock, and Darcel in this book). This includes social interactions, with the presence of other people during eating, compared with eating alone (Cruwys, Bevelander, & Hermans, 2015). For example, the liking and acceptability of foods are enhanced by social influence (Addessi, Galloway, Visalberghi, & Birch, 2005). Food intake is also higher when eating with others, although this effect could be modulated depending on the nature of the relationship with these people (de Castro, 1994; de Castro & de Castro, 1989). Several factors of the physical environment have also been studied with similar approaches. Location and the various stimuli related to the physical environment such as visual cues (Quartier, Vanrie, & Van Cleempoel, 2014), sound (Kantono et al., 2016), and odor (Stroebele & De Castro, 2004) could also impact food preferences. Although these studies of isolated contextual factors have shown promising results, the quality of evidence remains generally too weak to draw more general principles of how these factors affect food preferences and eating behaviors. Several confounding factors may explain the unclear conclusions drawn out of this empirical data. Eating, indeed, occurs in various places, including restaurants, cafes, bars, workplace canteens, school and university cafeterias, residential homes, and many others (Redden et al., 2015; Silva et al., 2017; van Kleef, Otten, & van Trijp, 2012; Visschers & Siegrist, 2015). By changing the physical location, several other parameters change, including population, type of service, degree of choice, cost, food handling, and storage time (Meiselman et al., 2000). Several authors have therefore advocated in favor of comparing complete eating situations. Meiselman et al. (2000) compared the acceptability of the same food served in different eating locations and reported that the liking of the meal was higher in a training restaurant compared with a campus student dining center. In a follow-up study, food acceptability was measured in three different types of eating environments. Overall acceptance and organoleptic characteristics of the same meal served in those three locations followed the order: student training restaurant > food science laboratory > student dining center (Meiselman et al., 2000). More specifically, there was a contextual effect for salad, bread, and dessert, but not for beverage, because it was a pre-prepared product, and people have similar expectations of such products in various locations. As the authors stated, the role of customer’s expectations is an important contextual factor in food acceptance, as customers rate not only the actual food properties, but also the expectations of these foods (Cardello, 1994). Similar results were observed by Edwards, Meiselman, Edwards, and Lesher (2003) in an attempt to separate the confounding effect of customer/eater with location by serving the same population

244

Context

in different eating environments. Another study found that self-reported expectations of liking of a given meal were significantly different in different evoked eating environments, and were ordered as followed: home > traditional, full service restaurant > diner/fast food > school food service > army food service> airline foodservice ¼ hospital food service (Cardello, Bell, & Kramer, 1996). There have been limited efforts to integrate this empirical evidence into models, and to explain the mechanisms at play. Some attempts have been made from the perspective of expectations and experience linked to the eating situation. For example, Grunert, Bredahl, and Brunsø (2004) propose a model of subjective evaluation of food quality that integrates the role of intrinsic and extrinsic cues, product information, context, expectations, and previous experiences (Grunert, 1995; Grunert, 1997; Grunert et al., 2004). The role of expectations and memories of a meal in satiety and food intake has also been examined in a series of studies by Brunstrom et al. (Brunstrom, Collingwood, & Rogers, 2010; Brunstrom, Shakeshaft, & ScottSamuel, 2008), and Higgs, (Higgs, 2005) respectively. Brunstrom and Rogers (2009) showed that consumers’ expectations of the satiating properties from a range of 16 commonly consumed foods predicted the energetic content of self-selected served portions at lunch time. However, expected liking of those foods explained less than 1% of the variance in self-selected portions under controlled laboratory conditions (Brunstrom & Rogers, 2009). On the contrary, actual measures of palatability of foods when assessed in the laboratory has a great impact on the amount of food intake, while it has a modest effect, explaining less than 5% of the variance in consumers’ food intake in their natural environments (de Castro, 2010). This could be attributed to self-selection of foods causing a ceiling effect in palatability. In a natural environment, people choose and select the foods they are willing to consume, but in the laboratory, the selection process is removed, and the range of palatability presented is not the most representative of the real world. Therefore, direct comparisons of the results obtained in different eating environments should be done with considerable caution given the various confounding contextual factors, and a careful decision must be made between laboratory and free-living studies in the study of eating behavior.

12.2.2 The importance of context in food decisions research The study of consumers’ economic behavior on the markets for food products has given increasing attention to context in recent decades, with the rise of the behavioral economics theories of individual decision-making. The standard consumer theory, in neoclassical microeconomics, considers the economic behavior of a rational, optimizing agent on a market. This view provides a framework that is normative by nature—it assumes full rationality of consumers. However, its descriptive value is questionable as the predictions of these theories, relying on the axioms of rational choice, are often violated in real markets (Thaler, 1980). The standard approach is based on assumptions about the nature of consumer preferences. It does not take into account cognitive processes and is typically grounded in the theory of revealed preferences, which states that the choices observed on the market reflect the (unobserved) preferences of the

Studying natural meals: What are the benefits of the living lab approach?

245

consumers, thus presupposing that the behaviors of individuals on markets are consistent (Camerer, Loewenstein, & Prelec, 2005). The observation that in real markets, consumers are, on the contrary, often inconsistent, impatient, and impulsive, has prompted the development of a descriptive, behavioral theory of choice inspired by recent advances in the fields of Psychology and Neurosciences. Behavioral economics and neuroeconomics have thus brought a new perspective to these models of consumer choice, trying to integrate insights about what people think and feel. The concepts of bounded rationality, of automatic processing, and of emotional processing are among the features of these theories that bring a new light on the possible interactions between context and economic behavior. The notion of bounded rationality was forged by Herbert Simon in the 1950s and refers to the fact that decision-makers use a limited amount of cognitive resources, information and time when making decisions. This notion departs from the neoclassical economics view of the rational decision-maker (where decision-makers consider all options before making a choice). In this view, the decision-makers will resort to heuristics— cognitive shortcuts or routines that allow them to make decisions at a lower cognitive cost. Heuristics rely, in part, on cues from the decision-makers’ environment, making the context of decision a central element for the study of economic behaviors. The economic approach of consumer behavior is the acknowledgement that most decisions—especially regarding food—result from automatic processes of decisions. Therefore, the importance given to the controlled processes in decision making up to now must be nuanced. This is especially critical with regard to informational contexts, the importance of which, in food purchasing decisions, may have been overestimated.

12.3

Evaluating eating behaviors within a meal context

Investigating human eating behavior is never straightforward, and several methodological approaches have been developed and employed with various strengths and limitations at both laboratory and free-living settings. The following section will briefly discuss the various methodologies commonly employed in both laboratory and freeliving environments for measuring appetite and eating behavior in humans, and study food choice, as well as measure product liking. Their relative benefits and limitations are also presented, opening the way to living labs.

12.3.1 Laboratory settings Whether the study of appetite and eating behavior, of perception and food decisions should be conducted under the well-controlled laboratory conditions, or within more realistic settings in free-living environments depends on the study hypotheses, research scope, and theoretical background of the study. To evaluate the benefits of either approach, two types of validity are considered: internal validity (or the ability of the experiment to provide valid data about the causal effect of a factor) and external validity (or the possibility of applying, or generalizing the results obtained in one experimental setting to another setting).

246

Context

12.3.1.1 Food perception and intake research in laboratory settings Given the several behavioral and environmental factors that could confound the results on satiety and food intake, studies exploring human appetite research are often conducted in a laboratory environment as described in this book by France Bellisle. This is further stimulated due to the requirements imposed by the European Food Safety Authority on satiety power claims of functional foods. Such controlled conditions provide much greater precision, accuracy, and control on the measurements of appetite and energy intake compared with the free-living natural settings (Gibbons, Finlayson, Dalton, Caudwell, & Blundell, 2014). Laboratory studies allow the isolation of specific factors, and thus determine their effect on appetite and energy intake free from external influences usually present in natural surroundings (Blundell et al., 2009). A mechanistic insight on the physiological parameters that regulate appetite is possible by collection of biological samples and metabolic measurements (Blundell et al., 2010). Finally, laboratory studies provide a strong internal validity, enabling causal effects to be drawn among the variables of interest. A widely used experimental design in a controlled laboratory setting is the “preload-test meal paradigm,” in which a fixed amount of food or drink (preload) of well-determined composition and structure is consumed. After a pre-determined time interval, an ad libitum test meal is provided, and food intake is measured. The satiating power of the preload food is monitored using visual analogue scales (VAS), a type of subjective assessment of appetite ratings (fullness, hunger, desire to eat, and prospective consumption), and the amount of energy consumed at the following meal (Blundell et al., 2010). While the appetite questionnaires are sensitive to experimental manipulations, and can predict initiation of a meal and food intake, they cannot accurately predict the size of the meal (Mattes, 1990). Thus, direct and precise measure of food intake can be a more useful and relevant way to assess satiety. Covert and automatic measure of food intake over time is feasible with the use of a universal eating monitor, a set of concealed weighing scales connected to a computer (Kissileff, Klingsberg, & Van Itallie, 1980). This automated method allows the measuring of minute by minute alterations in human food intake and appetite ratings (hunger, fullness, and palatability) taken at fixed intervals. Even more precise measures of recording the development of satiation and the eating rate or style of individuals, completing energy intake measures, is possible with the use of the Mandometer (Ioakimidis, Zandian, Bergh, & S€ odersten, 2009). In addition, a mechanistic insight in appetite regulation could be provided by the measurement of appetite-related peptides and physiological measurements (Blundell et al., 2010). This type of study design implies that all procedures are carefully monitored and applied in a consistent and standard manner between the researcher and any participants, following good laboratory practice, and are written in formal standard operating procedure documents (Gibbons et al., 2014). However, one of the major limitations is the external validity, because laboratories reflect an unnatural environment in which study participants typically ingest food (Blundell et al., 2009), thus limiting the generalizability of the data (Meiselman, 1992). Furthermore, factors such as provision of food portions served in excess and

Studying natural meals: What are the benefits of the living lab approach?

247

available free of cost could promote overconsumption (Rolls, Morris, & Roe, 2002). Given the impact of the hedonic system on appetite and energy intake, the difference between “liking” and “wanting” could confound the results of food intake, as individuals may like the study food, but they may not want to eat it at that occasion (Finlayson et al., 2007). Finally, a recent review showed that increased participant awareness of observation in laboratory studies (i.e., energy intake is monitored) may alter their eating behavior to meet social norms and reduce food intake (Robinson, Hardman, Halford, & Jones, 2015).

12.3.1.2 Food decision research in laboratory settings Similar considerations have been made regarding the study of consumers’ decisions. Because of the relative recency of experimental methods in Economics, the validity of experimental data in this field is still highly debated. While internal validity has been the main focus of concern in establishing the principles of economic experiments, the issue of generalizability and transferability of experimental data is highly controverted. Internal validity in Economic research relies on two leading principles: control and incentives. Control over factors of interest is the condition allowing hypothesis testing and ensuring replicability. To achieve this level of replicability, it is widely accepted that concessions must be made to the realism of the task. In the words of Camerer (2015): “Experiments contribute especially diagnostic evidence by virtue of extraordinary control over independent variables (internal validity). The price of this control is a necessary sacrifice in obtrusive measurement. There is an offsetting bonus of near-perfect replicability” (Camerer, 2015, p. 252). The robustness of the data collected is then ensured through the use of incentive-compatible evaluation tasks—procedures that have been designed specifically to provide incentives to respondents to act according to their true preferences, without any bias induced by the experimental task. This often involves experimental tasks that are binding—the responses of a participant will affect his/her final outcome in the experiment (often their remuneration). The incentive compatibility of the mechanisms used in experiments is demonstrated theoretically. This is the case, for example, in the so-called “Vickrey auction” (second price sealed bid auction) (Vickrey, 1961), which is often used to elicit willingness-to-pay for a food product in the laboratory (see for example (Lusk, 2003; Noussair, Robin, & Ruffieux, 2004; Saulais & Ruffieux, 2012).

12.3.2 Free-living settings Free-living field studies is an alternative approach, applicable to large-scale study research questions of eating behavior that cannot be addressed in the laboratory setting, which are extremely valuable for providing data from real people in real eating environments (Meiselman, 1992). The key strength is the high ecological relevance. However, the poor control over the varying external factors and experimental conditions is the most significant weakness of such studies. In addition, measures of the endpoints of interest, such as food intake in free-living individuals, is imprecise and prone to bias given the methodology

248

Context

commonly employed (some sort of self-report or recall of their dietary intake) (Goldberg & Black, 1998). This could be attributed to the heterogeneity in real life measurements and the lack of an established set of methodologies, procedures, and analyses in these free-living studies, contrary to the validated, reliable, and widely used methodology and experimental procedures used in the laboratory (Blundell et al., 2010; Wansink, 2009). Regarding the study of food decisions, the recent developments in behavioral economics, and especially the acknowledgement of the role of heuristics—using cues from the environment—in decision-making, have challenged the view of laboratory settings as the gold standard. In particular, the role played by context in decisionmaking has been underlined as a potential limit to the incentive compatibility of the tasks, leading some authors to advocate for the use of field experiments, which provide a realistic context of decision-making (Harrison & List, 2003; Harrison & List, 2004).

12.3.3 Living lab settings The Living Lab approach considers the contextual cues used in decision-making and eating behavior studies, in an attempt to bring better control to the factors of interest in the study, while keeping the task and pool of subjects natural. The principle of a Living Lab is to collect consumer data in a real usage experience with some control of various environmental conditions. In the field of consumer food and beverage studies (Giboreau, 2017), a Living Lab is designed to replicate actual consumption situations in a selected range of interests according to the purpose of the study. It could be a family dining room or a restaurant. For example, when the project evaluates children’s preference for food, the Living Lab can be designed like a school canteen, meaning that the evaluation occurs in a collective setting with a fixed menu, where food choices and intakes are recorded throughout the meal (Morizet, Depezay, Combris, Picard, & Giboreau, 2012). It must include an appropriate variation of provided food and drinks, but also the standardization (or contrast) of environmental criteria, such as the atmospherics, the food presentation, and the other guests involved. The Living Lab offers control of several contextual factors, including physical ones (e.g., served quantities), and some social ones (e.g., staff interaction). It also allows a recruitment process to ensure the selection of the target group related to the ongoing study. According the European Network of Living Lab organization (Enol http://www. openlivinglabs.eu/), a Living Lab is a user-centered, open innovation ecosystem based on five key elements: -

Active user involvement (in the innovation process), for example, in a food study, participants are individuals consuming the food in a meal situation. Real-life setting (i.e., testing new artifacts “in the wild”). For example, in an eating-out context, participants are paying for their meal. Multi-stakeholder participation (e.g., service providers, institutional actors, professional or residential end users): in a school canteen, the school staff prepares and serves the food to the kids.

Studying natural meals: What are the benefits of the living lab approach?

-

-

249

A multi-method approach (e.g., ethnography, psychology, sociology, strategic management, engineering). In food Living Labs, questionnaires are often combined with the measurement of consumed quantities. Co-creation (i.e., iterations of design cycles). Results of experiments are shared with designers and managers in order to allow future adaptation of the studied product/service.

In order to ensure the minimum impact of the research team, distant observation means are favored in Living Lab approaches, mainly using video recording, with participants being informed prior to the capture. This implies little use of declarative methods, although researchers may ask participants to fill out a short questionnaire, through a smartphone, tablet, or paper format, preferably at the end of the behavioral data collection. Considering the constraint of keeping the situation as natural as possible, Living Lab approaches are not appropriate for all research questions, especially when many samples need to be tested, or when long questionnaires need to be completed. However, it is particularly relevant to study consumers’ choice and preference in real-life situations. It can also be used for appetite research, especially when a mechanistic insight is required, facilitating the inclusion of biological sampling protocols (see case study A, which follows).

12.3.4 Comparison of experimental approaches in the case of food research The differences between approaches in the case of a food evaluation experiment are summarized in Table 12.1, adapted from (Harrison & List, 2004). While free-living experiments provide a necessarily more ecological environment of decision, this does not equate to greater external validity or generalizability (Camerer, 2015). Several authors advocate for a complementary, rather than a substitutive approach of the laboratory and field experiments in the study of consumer behavior and decisions (Camerer, 2015; Fiore, Harrison, Hughes, & Rutstrom, 2009), arguing that generalizability issues are not specific to laboratory approaches, as both field and laboratory are specific contexts (Guala, 2008). Additionnally, while the experimental context in the field may be more realistic, and therefore ecologically valid, this does not mean that these results obtained in a specific context are automatically valid in other contexts. Some authors therefore emphasize the need to re-center the debate regarding the generalizability of experimental measures on the qualitative, rather than the quantitative validity of data (Kessler & Vesterlund, 2015). Complementary to laboratory and free-living approaches, new research platforms such as experimental restaurants have been constructed, such as the Institut Paul Bocuse Food and Hospitality Research Centre (IPB) in Lyon, France. These types of restaurants provide a controlled environment that is ecologically valid from a consumer point of view. The following case studies were conducted at the experimental restaurant of the Institut Paul Bocuse, a platform labeled ENoLL (European Network of Living Labs). At the Institut Paul Bocuse, a laboratory is also available for light biomedical sampling in collaboration with a medical team from Lyon’s hospital.

250

Context

Table 12.1 Characteristics of experimental approaches in the case of food behavior research Laboratory

Freeliving

Living lab

Context of evaluation

Laboratory

Real eating environment, specific to the product evaluated

Approach of the “context” issue

As little context as possible

Context cues as close to the situation of consumption as possible

Subject pool

Subjects recruited for the experiment, sometimes students

Subjects recruited for the experiment or restaurant customers

Type of product evaluated

Generally, separate products

Products in the context of a whole meal, dishes

Involvement of subjects in valuation task, recruitment and remuneration

Conscious: the subjects know they are in an experiment and are recruited for it Participants could be paid

Not conscious: the participants do not know they are in an experiment and come with the primary purpose to have lunch, not to take part in a study Participants often pay for the primary service they were seeking (e.g., a meal)

Nature of the task that subjects have to perform

(Mostly) “Unnatural,” often unrelated to the consumption of the product: auctions on food products, liking scales, questionnaires

“Natural”: choices of food, consumption of given quantities of food

12.4

Living lab case studies

12.4.1 Study A. The effect of meal frequency on food intake in natural environment Studies that examine the effects of specific dietary intervention on the regulation of appetite and food intake are typically performed in a laboratory or clinical environment. Those studies usually provide an ad libitum single course meal, where a large portion of the test meal (e.g., pasta or pizza) is offered to the subjects, and only the total amount eaten is measured (Allirot et al., 2012). Also, the use of medical equipment (e.g., intravenous cannula or calorimetric dispositive) makes them prone to biases related to the study of eating behaviors in a clinical setting. Given the

Studying natural meals: What are the benefits of the living lab approach?

251

complexity of appetite and food intake measurement, research studies on food intake should be “multidisciplinary in nature,” incorporating behavioral and metabolic approaches. Allirot et al. aimed to bridge the gap between the laboratory and the real world in a series of studies integrating behavioral measurements (subjective appetite ratings, food intake, choices, and eating rhythms) in physiological approaches (appetite-related biomarkers and metabolic measures) by separating temporarily and geographically those two aspects (Allirot et al., 2012; Allirot et al., 2013; Allirot et al., 2014). They developed and validated an ecological buffet test meal design. It is an original methodology in appetite and food intake research by duplicating the same protocol in two different research sites: a clinical center specialized in metabolic research in nutrition for an in-depth physiological exploration, and Institut Paul Bocuse experimental restaurant for reproducing the realistic nature of a test meal and an ecological meal situation. Contrary to most buffet-type meal designs, which use a cold buffet, and are conducted in the classical laboratory, the men ate and sat in front of a small dining table at the experimental restaurant, and they had to get up and serve themselves as in an actual buffet. The buffet contained a range of cold and hot food items, as in a typical French meal. Although each person ate individually to control for the effect of the social aspects of eating, this design was as close to a typical eating situation as possible in terms of food choices and eating environment. Using this integrated methodology, combining behavioral and metabolic measurements in two specific and adequate places, they examine the effects of increased smaller and more frequent meals on short term appetite, biomarkers, eating behavior during the subsequent ad libitum test meal, and metabolic orientations. Twenty normal weight men consumed a breakfast in one eating episode at 0 min (BF1), or in four isoenergetic eating episodes at 0, 60, 120, and 180 min (BF4), followed by an ecological ad libitum buffet meal at 240 min in two sessions at the IPB experimental restaurant. They also participated in two more sessions consuming the same two breakfasts, BF1 and BF4, in the Clinical Centre, where biomarkers of appetite, glucose, insulin, and indirect calorimetry measures were taken. Appetite ratings were assessed throughout the experiment, and energy intake for the whole buffet meal and for each quarter of the breakfast meal were recorded at the experimental restaurant (Allirot et al., 2013). The same study design was applied in seventeen obese men in a follow-up study (Allirot et al., 2014). Results showed that appetite was reduced after consumption of smaller and more frequent meals in normal weight, but not in obese men, because there were no differences in total energy intake at the subsequent ad libitum buffet test meal. There were no differences in metabolic results between the normal weight and obese groups, with increased frequency of meals decreasing energy expenditure via diet induced thermogenesis. Those studies are the first step in demonstrating that more integrated research for behavioral explorations in a controlled, but ecological, eating environment dedicated to food and catering is feasible, and to provide a useful insight in the interpretation of the results in terms of external validity. Even more realistic eating situations with inclusion of social aspects of eating seem prudent.

252

Context

12.4.2 Study B. The impact of the ordering process on the valuation of coffee at the restaurant Studies on preferences for variety reveal a potential conflict from the point of view of the consumer: decision-makers tend to be variety-seeking (they look for and appreciate variety because it allows them to increase their chances to obtain the option that most closely matches their preferences) (Van Trijp & Steenkamp, 1992), and, at the same time, choice-avoidant (faced with too many options, they tend to be confused and opt for heuristics that simplify their decisions) (Iyengar & Lepper, 2000). In real food choice situations, such as when a consumer orders food and beverages in a restaurant, some characteristics of the options may be unknown to the consumers. This is the case, for example, of the origin of the coffee served in most restaurants in France, as the origin is a characteristic that is rarely disclosed for this type of product. Such a situation of incomplete information may lead consumers to under- or overvalue the product, if informed consumers would value coffee origins differently. Based on this consideration, producers of coffee of valued origins may want the restaurants to provide more information on the products’ origins. However, this implies that consumers who are made aware of coffee origins have a known preference between origins. In the opposite case, introducing a variety of quality-differentiated products may decrease, rather than increase, the overall value of the product range due to a “choice overload” phenomenon. As the tradeoff between variety-seeking and choice avoidance depends on the context, a field approach is necessary to ensure the realism of the task. Living Lab approaches allow this realism, but also the creation of controlled and comparable experimental contexts. In this study, two ranges of four coffee products, differentiated either by an unfamiliar characteristic [origin (O)] or by a more familiar one [preparation method (PM)] were introduced in a Living Lab restaurant setting, where usually only one coffee option is available. A menu card presenting these four options was provided to consumers. Then, depending on the experimental condition, participants had to place their order following one of three processes: (1) consumers had to make a choice of coffee within the products’ range on the menu card (“active choice” presentation), or (2) the choice was made for consumers randomly among the options of the range (“random” presentation), or (3) the choice came from a recommendation from a specialist (“recommendation” presentation). For each consumer, willingness to pay for one coffee from the menu was elicited for two of these three choice presentation scenarios, using an experimental auction procedure (see Becker, Degroot, & Marschak, 1964 for the BDM mechanism), which required an adaptation of the procedure of implementation of an experimental auction in a real choice situation (Lusk & Fox, 2003). Then, one of the two scenarios was drawn for each consumer, and a real purchase was then made. The experiments were conducted at the Living Lab for 15 sessions (5 sessions per treatment). In total, 231 French subjects took part in the experiment: 75 for treatment 1, 86 for 2, and 70 for 3. The experiment took place at the end of the meal. The results show that the average willingness to pay for a coffee of the range per preparation (PM) was higher than in the range by origins (O), regardless of the choice task. This

Studying natural meals: What are the benefits of the living lab approach?

253

indicates that variety-seeking behaviors depend on the characteristic of differentiation. The range for which consumers had a greater familiarity with the characteristics of the products (PM) had a higher value compared with (O). Choice avoidance was observed, but not to its full extent. In the case of low familiarity (O), there was a premium when the choice was delegated to a qualified person (“recommendation”) compared with the “active choice” scenario, and the contribution of a recommendation brings the overall value of the coffee of the range by origins to the height of the range by preparation. However, consumers’ value active choice more than random choice, regardless of the type of differentiation characteristic, indicating that consumers value choice avoidance only when the possibility of not choosing is delegated to an agent who is more knowledgeable. Overall, consumers placed more value on product ranges that are composed according to characteristics for which they have clear preferences. However, regardless of the range, consumers do not see the added value of not having to choose from the variety if it is done by chance. On the other hand, the recommendation by a third person is highly valued in the case of the range by origins, for which there is indecision. These results have direct implications for foodservice professionals concerned with how to design optimal menus (how many dish or beverage options should there be on a menu card to maximize customer satisfaction?). With this methodology, it is possible to gain insight into the value of variety on a menu card for restaurant consumers. In that type of study, a Living Lab approach provides a clear benefit, as decisions are made using cues from the way the task was conducted in order to make a decision. Therefore, a realistic setting, while maintaining control over the task, ensures both that the heuristics at play are the ones that would be used in a natural restaurant environment, and that the factors of the task remain constant for the study. This allows to researchers strengthen the transferability of the results to other restaurant environments.

12.5

Future trends

Whereas eating can be studied in a controlled lab, natural dining is hardy reproducible, and some compromises are needed to follow a research procedure in real-life meals. One such approach is to conduct experiments in free-living settings, that is, to acquire data in real dining environments and collect detailed information on the conditions to help with interpretation. A series of experimental studies conducted out of the lab and out of the Living Lab are now available, showing the interest in free living environments for studying categories of consumers and contexts demanding high specificity of meals, such as kindergarten, nursing homes, and hospitals (Pouyet, Cuvelier, Benattar, & Giboreau, 2015). A second approach is to use a Living Lab where the eating situation occurs in a real dining occasion, but where contextual factors are designed and controlled by the research team (Allirot et al., 2014; Cliceri, Petit, Garrel, Monteleone, & Giboreau, 2017; Fernandez, Bensafi, Rouby, & Giboreau, 2013). A third approach takes place in laboratories, which use recreated contexts (contextualized laboratories), or even virtual reality (Fiore et al., 2009) to integrate these

254

Context

contextual elements into the task. Scenario-based methods present consumers with a scenario and request them to predict their food intake under various manipulated conditions. Although such studies have lower ecological validity relative to actual or reported food intake studies, they are inexpensive and can generate useful information in methodological studies or where there is a special interest in direct comparison of contextual influences ( Jaeger & Porcherot, 2017). The use of written and pictorial scenarios is a common approach for evoking consumption context for productfocused consumer tests, and has also been used to measure hedonics, acceptance, and intention or rejection of purchases and emotional responses (Sinesio et al., 2018). However, the cognitive effort required to imagine the situation may detract attention away from the products and analytical tasks, thus more immersive approaches using auditory and visual stimulation (immersive virtual reality), as less cognitively demanded, have been used by some researchers (Bangcuyo et al., 2015; Sester et al., 2013). A number of methodologies related to product-focused investigations taking into account the consumption context and ecological validity have been summarized and discussed in a recent review by Jaeger and Porcherot (2017). As the authors stated, “Considering that ecological validity is not a simple, nor a well-defined, construct, asking whether or not a study is ecologically valid is probably too simplistic. Asking instead when ecological validity is needed and how much, and accepting that the answers will be highly study specific, may be a better way forward” ( Jaeger & Porcherot, 2017).

References Addessi, E., Galloway, A. T., Visalberghi, E., & Birch, L. L. (2005). Specific social influences on the acceptance of novel foods in 2–5-year-old children. Appetite, 45, 264–271. https:// doi.org/10.1016/j.appet.2005.07.007. Allirot, X., Saulais, L., Disse, E., Roth, H., Cazal, C., & Laville, M. (2012). Validation of a buffet meal design in an experimental restaurant. Appetite, 58, 889–897. https://doi.org/ 10.1016/j.appet.2012.02.011. Allirot, X., Saulais, L., Seyssel, K., Graeppi-Dulac, J., Roth, H., Charrie, A., et al. (2013). An isocaloric increase of eating episodes in the morning contributes to decrease energy intake at lunch in lean men. Physiology and Behavior, 110–111, 169–178. https://doi.org/ 10.1016/j.physbeh.2013.01.009. Allirot, X., Seyssel, K., Saulais, L., Roth, H., Charrie, A., Drai, J., et al. (2014). Effects of a breakfast spread out over time on the food intake at lunch and the hormonal responses in obese men. Physiology and Behavior, 127, 37–44. https://doi.org/10.1016/j. physbeh.2014.01.004. Bangcuyo, R. G., Smith, K. J., Zumach, J. L., Pierce, A. M., Guttman, G. A., & Simons, C. T. (2015). The use of immersive technologies to improve consumer testing: The role of ecological validity, context and engagement in evaluating coffee. Food Quality and Preference, 41, 84–95. https://doi.org/10.1016/J.FOODQUAL.2014.11.017. Becker, G. M., Degroot, M. H., & Marschak, J. (1964). Measuring utility by a single-response sequential method. Behavioral Science, 9, 226–232. https://doi.org/10.1002/bs.3830090304. Benelam, B. (2009). Satiation, satiety and their effects on eating behaviour. Nutrition Bulletin, 34, 126–173. https://doi.org/10.1111/j.1467-3010.2009.01753.x.

Studying natural meals: What are the benefits of the living lab approach?

255

Blundell, J., De Graaf, C., Hulshof, T., Jebb, S., Livingstone, B., Lluch, A., et al. (2010). Appetite control: Methodological aspects of the evaluation of foods. Obesity Reviews, 11, 251–270. https://doi.org/10.1111/j.1467-789X.2010.00714.x. Blundell, J., de Graaf, K., Finlayson, G., Halford, J. C. G., Hetherington, M., & King, N. (2009). Measuring food intake, hunger, satiety and satiation in the laboratory j QUT ePrints. In D. B. Allison & M. L. Baskin (Eds.), Handbook of assessment methods for eating behaviors and weight-related problems (pp. 283–325): Sage Publications. Blundell, J., Rogers, P., & Hill, A. (1987). Evaluating the satiating power of foods: implications for acceptance and consumption. In J. Colms, D. A. Booth, R. M. Pangborn, & O. Raunhardt (Eds.), Food acceptance and nutrition (pp. 205–219). . Blundell, J. E. (2017). The contribution of behavioural science to nutrition: Appetite control. Nutrition Bulletin, 42, 236–245. https://doi.org/10.1111/nbu.12279. Brunstrom, J. M., Collingwood, J., & Rogers, P. J. (2010). Perceived volume, expected satiation, and the energy content of self-selected meals. Appetite, 55, 25–29. https://doi.org/ 10.1016/j.appet.2010.03.005. Brunstrom, J. M., & Rogers, P. J. (2009). How many calories are on our plate? Expected fullness, not liking, determines meal-size selection. Obesity, 17, 1884–1890. https://doi.org/ 10.1038/oby.2009.201. Brunstrom, J. M., Shakeshaft, N. G., & Scott-Samuel, N. E. (2008). Measuring “expected satiety” in a range of common foods using a method of constant stimuli. Appetite, 51, 604–614. https://doi.org/10.1016/j.appet.2008.04.017. Camerer, C., Loewenstein, G., & Prelec, D. (2005). Neuroeconomics: How neuroscience can inform economics. Journal of Economic Literature, 43, 9–64. https://doi.org/10.1257/ 0022051053737843. Camerer, C. F. (2015). The promise and success of lab-field generalizability in experimental economics: A critical reply to levitt and list. In G. R. Frechette & A. Schotter (Eds.), Handbook of experimental economic methodology (pp. 249–295). New York: Oxford University Press. https://doi.org/10.2139/ssrn.1977749. Cardello, A., Bell, R., & Kramer, F. (1996). Attitudes of consumers toward military and other institutional foods. Food Quality and Preference, 7, 7–20. Cardello, A. V. (1994). Consumer expectations and their role in food acceptance. In D. Thompson & H. Macfie (Eds.), Measurement of food preferences (pp. 253–297). Glasgow: Blackie Academic and Professional. https://doi.org/10.1007/978-1-4615-2171-6_10. Cliceri, D., Petit, E., Garrel, C., Monteleone, E., & Giboreau, A. (2017). Effect of glass shape on subjective and behavioral consumer responses in a real-life context of drinking consumption. Food Quality and Preference, 60, 19–30. https://doi.org/10.1016/j.foodqual.2017.09.008. Cruwys, T., Bevelander, K. E., & Hermans, R. C. J. (2015). Social modeling of eating: A review of when and why social influence affects food intake and choice. Appetite, 86, 3–18. https://doi.org/10.1016/j.appet.2014.08.035. de Andrade, J. C., Giongo, C., de Barcellos, M. D., & Deliza, R. (2017). Parameters under different evoked contexts. Meat Science, 130, 30–37. https://doi.org/10.1016/j. meatsci.2017.04.003. de Castro, J. M. (1994). Family and friends produce greater social facilitation of food intake than other companions. Physiology and Behavior, 56, 445–455. de Castro, J. M. (2010). The control of food intake of free-living humans: Putting the pieces back together. Physiology and Behavior, 100, 446–453. https://doi.org/10.1016/j. physbeh.2010.04.028.

256

Context

de Castro, J. M., & de Castro, E. S. (1989). Spontaneous meal patterns of humans: Influence of the presence of other people. American Journal of Clinical Nutrition, 50, 237–247. https:// doi.org/10.1093/ajcn/50.2.237. de Graaf, C., Cardello, A. V., Matthew Kramer, F., Lesher, L. L., Meiselman, H. L., & Schutz, H. G. (2005). A comparison between liking ratings obtained under laboratory and field conditions: The role of choice. Appetite, 44, 15–22. https://doi.org/10.1016/J. APPET.2003.06.002. Divert, C., Laghmaoui, R., Crema, C., Issanchou, S., Van Wymelbeke, V., & Sulmont-Rosse, C. (2015). Improving meal context in nursing homes. Impact of four strategies on food intake and meal pleasure. Appetite, 84, 139–147. https://doi.org/10.1016/j.appet.2014.09.027. Edwards, J. S. A., Meiselman, H. L., Edwards, A., & Lesher, L. (2003). The influence of eating location on the acceptability of identically prepared foods. Food Quality and Preference, 14, 647–652. Emilien, C., & Hollis, J. H. (2017). A brief review of salient factors influencing adult eating behaviour. Nutrition Research Reviews, 30, 233–246. https://doi.org/10.1017/ S0954422417000099. Fernandez, P., Bensafi, M., Rouby, C., & Giboreau, A. (2013). Does olfactory specific satiety take place in a natural setting? Appetite, 60, 1–4. https://doi.org/10.1016/j. appet.2012.10.006. Finlayson, G., King, N., & Blundell, J. E. (2007). Liking vs. wanting food: Importance for human appetite control and weight regulation. Neuroscience and Biobehavioral Reviews, 31, 987–1002. https://doi.org/10.1016/j.neubiorev.2007.03.004. Fiore, S., Harrison, G., Hughes, C., & Rutstrom, E. E. (2009). Virtual experiments and environmental policy. Journal of Environmental Economics and Management, 57, 65–86. Gibbons, C., Finlayson, G., Dalton, M., Caudwell, P., & Blundell, J. E. (2014). Metabolic phenotyping guidelines: Studying eating behaviour in humans. Journal of Endocrinology, 222, G1–G12. https://doi.org/10.1530/JOE-14-0020. Giboreau, A. (2017). Sensory and consumer research in culinary approaches to food. Current Opinion in Food Science, 15, 87–92. https://doi.org/10.1016/j.cofs.2017.06.001. Goldberg, G. R., & Black, A. E. (1998). Assessment of the validity of reported energy intakes – € Review and recent developments. Naringsforskning, 42, 6–9. https://doi.org/10.3402/fnr. v42i0.1760. Grunert, K. (1997). What’s in a steak? A cross-cultural study on the quality perception of beef. Food Quality and Preference, 8, 157–174. Grunert, K. G. (1995). Food quality: A means-end perspective. Food Quality and Preference, 6, 171–176. Grunert, K. G., Bredahl, L., & Brunsø, K. (2004). Consumer perception of meat quality and implications for product development in the meat sector—A review. Meat Science, 66, 259–272. Guala, F. (2008). Experimentation in economics. In U. M€aki (Ed.), Elsevier handbook of the philosophy of science, vol. 13: Philosophy of economics (pp. 1–48). Harrison, G., & List, J. (2003). What constitutes a field experiment in economics? Work. Pap. Available from: PDF/ListHarrison_JEL_2004_WP-3952934400/ListHarrison_JEL_2004_WP.pdf. Harrison, G. W., & List, J. A. (2004). Field experiments. Journal of Economic Literature, 42, 1009–1055. Higgs, S. (2005). Memory and its role in appetite regulation. Physiology and Behavior. https:// doi.org/10.1016/j.physbeh.2005.04.003. Ioakimidis, I., Zandian, M., Bergh, C., & S€odersten, P. (2009). A method for the control of eating rate: A potential intervention in eating disorders. Behavior Research Methods, 41, 755–760. https://doi.org/10.3758/BRM.41.3.755.

Studying natural meals: What are the benefits of the living lab approach?

257

Iyengar, S. S., & Lepper, M. R. (2000). When choice is demotivating: Can one desire too much of a good thing? Journal of Personality and Social Psychology, 79, 995–1006. https://doi. org/10.1037/0022-3514.79.6.995. Jaeger, S. R., & Porcherot, C. (2017). Consumption context in consumer research: Methodological perspectives. Current Opininon in Food Science, 15, 30–37. https://doi.org/10.1016/j. cofs.2017.05.001. Kantono, K., Hamid, N., Shepherd, D., Lin, Y. H. T., Yakuncheva, S., Yoo, M. J. Y., et al. (2016). The influence of auditory and visual stimuli on the pleasantness of chocolate gelati. Food Quality and Preference, 53, 9–18. https://doi.org/10.1016/J.FOODQUAL.2016.05.008. Kessler, J. B., & Vesterlund, L. (2015). The external validity of laboratory experiments: The misleading emphasis on quantitative effects. In G. R. Frechette & A. Schotter (Eds.), Hanbook of economics (pp. 391–406): Oxford Press University. Kissileff, H. R., Klingsberg, G., & Van Itallie, T. B. (1980). Universal eating monitor for continuous recording of solid or liquid consumption in man. American Journal of PhysiologyRegulatory, Integrative and Comparative Physiology, 238, R14–R22. https://doi.org/ 10.1152/ajpregu.1980.238.1.R14. Korzen, S., & Lassen, J. (2010). Meat in context. On the relation between perceptions and contexts. Appetite, 54, 274–281. https://doi.org/10.1016/j.appet.2009.11.011. K€ oster, E. P. (2009). Diversity in the determinants of food choice: A psychological perspective. Food Quality and Preference, 20, 70–82. https://doi.org/10.1016/J.FOODQUAL. 2007.11.002. Lusk, J. L. (2003). Using experimental auctions for marketing applications: A discussion. Journal of Agricultural and Applied Economics, 35, 349–360. Lusk, J. L., & Fox, J. A. (2003). Value elicitation in retail and laboratory environments. Economics Letters, 79, 27–34 [Available from: PDF/LuskFox_EconLetters_20030778481409/LuskFox_EconLetters_2003.pdf]. Mattes, R. (1990). Hunger ratings are not a valid proxy measure of reported food intake in humans. Appetite, 15, 103–113. Meiselman, H. L. (1992). Obstacles to studying real people eating real meals in real situations. Appetite, 19, 84–86. https://doi.org/10.1016/0195-6663(92)90243-Y. Meiselman, H. L. (2006). The role of context in food choice, food acceptance and food consumption. In R. Shepherd & M. Raats (Eds.), The psychology of food choice (pp. 179– 199). Wallingford: CABI. https://doi.org/10.1079/9780851990323.0179. Meiselman, H. L., Johnson, J. L., Reeve, W., & Crouch, J. E. (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35, 231–237. https:// doi.org/10.1006/appe.2000.0360. Morizet, D., Depezay, L., Combris, P., Picard, D., & Giboreau, A. (2012). Effect of labeling on new vegetable dish acceptance in preadolescent children. Appetite, 59, 399–402. https:// doihttps://doi.org/10.1016/j.appet.2012.05.030. Noussair, C., Robin, S., & Ruffieux, B. (2004). A comparison of hedonic rating and demandrevealing auctions. Food Quality and Preference, 15, 393–402. Pouyet, V., Cuvelier, G., Benattar, L., & Giboreau, A. (2015). Influence of flavour enhancement on food liking and consumption in older adults with poor, moderate or high cognitive status. Food Quality and Preference, 44, 119–129. https://doi.org/10.1016/J. FOODQUAL.2015.04.014. Quartier, K., Vanrie, J., & Van Cleempoel, K. (2014). As real as it gets: What role does lighting have on consumer’s perception of atmosphere, emotions and behaviour? Journal of Environmental Psychology, 39, 32–39. https://doi.org/10.1016/J.JENVP.2014.04.005. Redden, J. P., Mann, T., Vickers, Z., Mykerezi, E., Reicks, M., & Elsbernd, S. (2015). Serving first in isolation increases vegetable intake among elementary schoolchildren. PLoS One, 10, e0121283. https://doi.org/10.1371/journal.pone.0121283.

258

Context

Robinson, E., Hardman, C. A., Halford, J. C. G., & Jones, A. (2015). Eating under observation: A systematic review and meta-analysis of the effect that heightened awareness of observation has on laboratory measured energy intake. American Journal of Clinical Nutrition, 102, 324–337. https://doi.org/10.3945/ajcn.115.111195. Rolls, B. J., Morris, E. L., & Roe, L. S. (2002). Portion size of food affects energy intake in normal-weight and overweight men and women. American Journal of Clinical Nutrition, 76, 1207–1213. Rozin, P. (1998). Towards a psychology of food choice. In Towards a psychology of food choice: Institut Danone. Rozin, P., & Tuorila, H. (1993). Simultaneous and temporal contextual influences on food acceptance. Food Quality and Preference, 4, 11–20. https://doi.org/10.1016/0950-3293 (93)90309-T. Saulais, L., & Ruffieux, B. (2012). A field experiment to design healthier foods: Consumer valuation of butter production processes. Food Quality and Preference, 26, 178–187. https:// doi.org/10.1016/j.foodqual.2012.04.011. Sester, C., Deroy, O., Sutan, A., Galia, F., Desmarchelier, J.-F., Valentin, D., et al. (2013). “Having a drink in a bar”: An immersive approach to explore the effects of context on drink choice. Food Quality and Preference, 28, 23–31. https://doi.org/10.1016/J. FOODQUAL.2012.07.006. Silva, A. P., Jager, G., Voss, H.-P., van Zyl, H., Hogg, T., Pintado, M., et al. (2017). What’s in a name? The effect of congruent and incongruent product names on liking and emotions when consuming beer or non-alcoholic beer in a bar. Food Quality and Preference, 55, 58–66. https://doi.org/10.1016/j.foodqual.2016.08.008. Sinesio, F., Saba, A., Peparaio, M., Saggia Civitelli, E., Paoletti, F., & Moneta, E. (2018). Capturing consumer perception of vegetable freshness in a simulated real-life taste situation. Food Research International, 105, 764–771. https://doi.org/10.1016/j. foodres.2017.11.073. Stroebele, N., & De Castro, J. M. (2004). Effect of ambience on food intake and food choice. Nutrition, 20, 821–838. https://doi.org/10.1016/j.nut.2004.05.012. Thaler, R. H. (1980). Toward a positive theory of consumer choice. Journal of Economic Behavior & Organization, 1, 39–60. https://doi.org/10.1016/0167-2681(80)90051-7. van Kleef, E., Otten, K., & van Trijp, H. C. (2012). Healthy snacks at the checkout counter: A lab and field study on the impact of shelf arrangement and assortment structure on consumer choices. BMC Public Health, 12(1), 1072. https://doi.org/10.1186/1471-2458-12-1072. Van Trijp, H. C., & Steenkamp, J.-B. E. M. (1992). Consumers’ variety seeking tendency with respect to foods: Measurement and managerial implications. European Reviews of Agricultural Economics, 19, 181–195. https://doi.org/10.1093/erae/19.2.181. Vickrey, W. (1961). Counterspeculation, auctions, and competitive sealed tenders. Journal of Finance, 16, 8–37. Visschers, V. H. M., & Siegrist, M. (2015). Does better for the environment mean less tasty? Offering more climate-friendly meals is good for the environment and customer satisfaction. Appetite, 95, 475–483. https://doi.org/10.1016/j.appet.2015.08.013. Wansink, B. (2009). Measuring food intake in field studies. In D. B. Allison & M. L. Baskin (Eds.), Handbook of assessment methods for eating behaviors and weight related problems (pp. 327–345). Los Angeles: Sage Publishing. Wansink, B. (2010). Mindless eating: Why we eat more than we think. Bantam Books. Weber, A. J., King, S. C., & Meiselman, H. L. (2004). Effects of social interaction, physical environment and food choice freedom on consumption in a meal-testing environment. Appetite, 42, 115–118. https://doi.org/10.1016/j.appet.2003.10.001.

The effects of environment on product design and evaluation: Meals in context, institutional foodservice

13

John S.A. Edwards, Heather J. Hartwell, Sarah Price Faculty of Management, Bournemouth University, Dorset, United Kingdom

13.1

Introduction

The foodservice sector is dynamic, multifaceted, and one of the world’s largest and fastest growing industries, with a worldwide turnover of $2661.3 billion (Euromonitor International, 2017) (Fig. 13.1). In the United States, restaurant sales have risen from $42.8 billion in 1970 to $379.0 billion in 2000,with sales of approximately $798.7 billion in 2017. In the United States, the restaurant industry has 14.7 million employees working in >1 million outlets (National Restaurant Association, 2017). However, the Asia Pacific region has the largest and most dynamic foodservice market, and accounts for most of the current sector evolution, showing an average annual growth forecast of 7.5% for the 2017–26 period. The Middle East and Africa was the second fastest emergent region between 2006 and 16, with average annual growth of 7.4%, and is also expected to see development for 2017–26 at 7.3% per year. Elsewhere, China is expected to see strong growth, caused in part by rapid urbanization impacting consumers looking increasingly for convenience in everyday meals. The foodservice industry is comprised of a number of eating out establishments, which can be classified in a variety of ways. A general classification is the public sector and the private sector; or the profit, and not for profit sector, although the exact differences are often difficult to determine, as there is invariably a lot of overlap. For example, hospitals, schools, and universities can be found in both the public and private sectors; even so, a generalized schematic categorization, designed to illustrate the diversity of the foodservice industry, is given in Fig. 13.2.

13.2

Purpose

The purpose of this chapter is to: l

Identify and describe what exactly is understood by foodservice and institutional foodservice;

Context. https://doi.org/10.1016/B978-0-12-814495-4.00013-1 Copyright © 2019 Elsevier Inc. All rights reserved.

260

Context

Fig. 13.1 Worldwide turnover of foodservice sectors. Source: Euromonitor International, (2017). Consumer foodservice global industry overview, October 2017. Available from: http://www.portal.euromonitor.com/portal/analysis/tab (Accessed 16 October 2017).

l

l

Identify the characteristics of institutional foodservice and what sets it apart from other foodservice operations; Consider and evaluate how these characteristics, that is, the context, influence, and need to be considered in product design.

13.3

What is foodservice?

In order to understand the context of institutional food service, it is first necessary to be clear on semantics, as the terminology, while often used synonymously, actually has different meanings (and spellings) in various parts of the English-speaking world; these differences have been previously enunciated (Edwards & Overstreet, 2009). In practice, many countries use the all-embracing term “hospitality,” others prefer the more specific terms “foodservice” (USA), “food service” (UK), and “catering,” the latter having totally different connotations on each side of the Atlantic. “Foodservice” and “food service” only differ in spelling, with the suggestion that the two words were combined to avoid any confusion with the pure service of food. In the United States, “catering” tends to be used to denote the actual provision of food (a catered event); with “foodservice” being reserved for the industry. In the United Kingdom, the term “catering industry” is used to embrace the industry, and also the provision of food; although the term “foodservice” is becoming more and more prevalent.

Categorisation of the foodservice secotors Profit sector Commercial sector Restaurants etc Room service Bars

Public sector Cost sector Subsidised sector Institutional/welfare sector

Hotels In-patients Out-patients

Patients Formal Bistros Speciality

Informal

Restaurants

Hospitals

Ethnic

Medical Others

Staff Visitors

Cafes, coffee shops, sandwich bars, etc Restaurants

Public houses, wine bars etc

Bars

Roadside/motorway

Old people's homes

Social services

Day care centres Meals on wheels

Restaurant On-board

Buffet car Trolley

Rail

Restaurants etc

In-flight Terminal Cruises Ferries Restaurants etc Burgers etc Fried chicken Fish & chips Pizzas Pasta

Bars

Students

Universities

Station

Transport

Staff

Education

Day Residential

Schools

Air

Sea

Clubs

The effects of environment on product design and evaluation

Private sector

Prisoners

Prisons

Fast food & franchising

Staff Visitors Police

Public services

Others Public restaurants etc Staff feeding

In-store

Navy

Take outs Home delivery Home meal replacement Restaurants Cafeterias Kiosks

Fire Ambulance

Miscellaneous

Leisure and theme parks

Armed forces

Army Air Force Marines Industrial

Other employee feeding

Nonindustrial Contract vs in-house

261

Fig. 13.2 The diversity of the foodservice industry. Source: Adapted from Edwards, J.S.A., & Overstreet, K. (2009). What is foodservice? Journal of Foodservice, 20, 1–3.

262

Context

Another area is the term “Contract Catering,” which has a variety of other names, including “Food Service Management.” For the purpose of this chapter, therefore, the term “foodservice” has been adopted throughout to refer to the industry and the provision of food; and “contractors/contract catering/contracting out” to refer to situations in which a company’s and organization’s main purpose is to provide foodservice and meals to various third-party organizations.

13.4

What is institutional foodservice?

As might be expected, various definitions exist. For example, a popular definition might be: entities that provide meals at institutions including schools, colleges and universities, and hospitals, as well as correctional facilities, public and private cafeterias, nursing homes, and day-care and senior centers (Conor, 2014). Another example (Warde and Martens, 2000, p 35) suggests also that the provision is on-site and restricted by criteria of membership, rather than the ability to pay, although, of course, payment is often made. However, this type of definition does not really encapsulate the nature and characteristics of institutional food service where, for example, a similar style of foodservice can be found in office complexes and factories; but even in these organizations, the style of service might vary from a “self-service canteen” to a “directors’ dining room.” Furthermore, semi-commercial restaurants serving hospital visitors, and commercial restaurants in shopping malls and department stores may also have a similar style of service. Adding to the confusion are the various terms that are used to describe what ostensibly are similar operations, as shown in Table 13.1. Perhaps, therefore, institutional foodservice might be better categorized, first by the nature and type of business it serves (Fig. 13.3), and second, by the inherent characteristics present in the operation (Fig. 13.4).

Table 13.1 Examples of names used to describe institutional dining areas Type of worker

Possible terminology

Industrial workers Office workers and Hospital staff Directors

(Work) canteen (Staff ) restaurant or (Staff ) cafeteria Directors’ dining room or Executives’ dining room Refectory or Cafeteria Dining room

Students Care home Military Officers (army, air force) Officers (navy) Soldiers etc.

Officers’ mess Ward room Dining room, Dining hall or Cookhouse or Mess hall

The effects of environment on product design and evaluation

263

Fig. 13.3 Alternative Categorization of Institutional Foodservice. Note: Many of these institutions, for example, schools, universities, hospitals and care homes, could be either publically or privately funded.

13.5

A brief history and development of institutional foodservice

Europe, at the turn of the 19th century, was characterized primarily as an agricultural society. In 1801, in England, for example, only one-fifth of the population of 8.9 million were town dwellers; four-fifths were working the land (Burnett, 1979, p. 15). Meals at home were very meager, while the midday meal on the land would have consisted primarily of bread with some cheese, tea, or perhaps small (weak) beer. By 1851, the populations were evenly matched, and by 1901, they were completely reversed as workers moved to the towns, to be near factories and mines at the start of the industrial revolution, and needed to be fed at work. The feeding arrangements tended to vary and, although the standards of living might have improved for some, the working man, who could not go home for dinner, usually took cold food to work; with bread, cheese, meat or dripping, and pie with potatoes, being the most common (Burnett, 1979, p. 190). Canteens to feed workers, where provided by employers, made little progress before the first World War, where it was estimated that there were barely 100 in existence (Burnett, 2004, p. 110). There were, however, exceptions; “enlightened” employers, such as the chocolate and cocoa manufacturers, Messrs. Cadbury, Fry, and Rowntree; and soap manufacturers, Messrs. Lever Brothers, provided a range

264

Context

“ “

”

”

Fig. 13.4 Inherent characteristics of institutional foodservice in the United Kingdom (workplace canteens, welfare and commercial/semi-commercial establishment).

of benefits, including subsidized meals for their employees (Curtis-Bennett, 1949, p. 186). The outbreak of the first World War saw a number of initiatives, both voluntary and encouraged by the Government, which resulted in the number of work canteens rising from 100 to 1000, producing more than a million meals per day (Burnett, 2004, p. 181). The outbreak of the second World War produced a number of changes aimed at stabilizing the food supply situation. In 1940, for example, the Factory Canteens Order required by law that every factory employing >250 workers provided a dining room where “wholesome meals at reasonable prices” could be obtained. This resulted in the number of canteens rising from 7530 in April 1942 to 11,635 in November of the same year. At the time, in the order of 4.5 million meals were served per week, which had risen to 50 million by 1945 (Curtis-Bennett, 1949, p. 249). One of the biggest influences, worldwide, was the United Nations, Hot Springs conference in 1943, where it was laid down, that as a general principle, all industrial workers should be given the opportunity to obtain a third of their daily calories at their midday meal at work (Curtis-Bennett, 1949, p. 269).

The effects of environment on product design and evaluation

265

After the second World War, at least 25,000 “canteens” existed, and in the United Kingdom were the principal eating out occasion. Most canteen users regarded this as the main meal of the day (Burnett, 2004, p. 303). Today, the provision of a place to eat, often subsidized, can be regarded in the United Kingdom as an integral part of a total remuneration package, and a high standard of meal and dining experience are expected.

13.6

Types and categories of institutional foodservice

13.6.1 Industrial (work canteens) and office (staff restaurants) Often referred to as “Business and Industry” by contract caterers, providing the workforce with meals is generally seen as being advantageous. Employers have especially recognized the effects meals can have on the productivity of their employees, and have taken more responsibility by offering meals at the workplace (Jørgensen et al., 2010). It is here where employees spend a large amount of time, and consume a high proportion of their overall dietary intake, with a large number of workers taking one or more meals at work ( Jørgensen et al., 2010), either in a workplace canteen, invariably a self-service cafeteria, or “out of hours,” by vending machines (Nyberg & Olsen, 2010). A self-service cafeteria is suitable and capable of serving a varying number of customers, in different situations, in a limited amount of time and with fewer staff, hence also becoming popular in the commercial/semi-commercial sectors, and can be found in complexes including shopping malls; travel terminals, such as airports and railway stations; motorway service areas; and within larger stores and department stores. There are differences in the availability and use of workplace canteens within Europe. It is suggested (Lund, Kjærnes, & Holm, 2017) that in Scandinavia, eating out patterns have changed little in the past 15 years, and eating out is not a frequent part of everyday eating. Most eating takes place at home, with most of the eating out taking place at work or at school. Finland, for example, has a long tradition of providing food at work, and with meals based on the national dietary recommendations, where the food habits of the population have improved ( Jørgensen et al., 2010; Raulio, Roos, & Pr€att€al€a, 2010). In addition, meals in Scandinavian countries are usually subsidized and, therefore, they are not perceived to be too expensive, as is the case in the United Kingdom (Pridgeon & Whitehead, 2013; Raulio et al., 2010). Furthermore, in Denmark, some canteens offer ready meals that can be taken home for consumption in the evening in order to meet the demands of their time-constrained customers (Quintiliani, Poulsen, & Sorensen, 2010). In Germany, there is a difference between the prevalence and use of meals at work, which historically stems from the divide between East and West Germany. In East Germany, it was very common to eat the midday meal at a state-run workplace canteen, and this has continued to be the case in comparison with West Germany, where, although canteens are available, packed lunches, and more recently, other

266

Context

opportunities to buy food in cafes dominate (Heinzelmann, 2008). Nevertheless, as in other European countries, food offered at work is generally subsidized, and therefore a reasonably priced alternative to packed lunches or take-away foods (Heinzelmann, 2008).

13.6.2 Schools and universities The provision of meals at schools obviously varies, but in most countries, the emphasis has tended to move away from ensuring an adequate nutrient intake to one that addresses rising childhood obesity. On the other hand, students are increasingly demanding a choice, hence meal provision tends to be through a self-selected cafeteria where operators attempt to balance the provision of healthy eating with the demands of their young customers. Once at university, meal provision, usually subsidized, tends to focus on and reflect what is found outside a university environment, akin perhaps to that found on a high street, for example, pizzas and chips (french fries).

13.6.3 Welfare (hospitals, and elderly) The welfare sector is comprised of a number of establishments, the major ones being hospitals, care for the elderly, in either residential homes, day centers, or in the community.

13.6.4 Hospitals Hospitals provide a wide range of foodservice activities, including feeding patients, both in-patients and day patients, hospital staff, and visitors; each requiring a different type and style of foodservice.

13.6.4.1 Hospital in-patients In-patients tend to be able to order their meals in advance (24 h) from a fixed, cyclical menu with, perhaps, some choice, which are then in the United Kingdom served ready plated to patients in bed. More recently, hospitals have been experimenting with a restaurant style multi-choice menu from which patients can order the type of meals that suit them.

13.6.4.2 Hospital staff, day patients, and visitors Day patients, staff, and visitors are invariably provided with a number of eating opportunities, including self-service cafeterias, serving both hot and cold food, cafes operated by the hospital foodservice operators, contractors, voluntary organizations, or branded outlets.

The effects of environment on product design and evaluation

267

13.6.5 The elderly The worldwide population is both growing and becoming increasingly older, and by 2050 the number of EU citizens aged 65 and over is expected to grow by 70%; and the number of people aged over 80 by 170% (Eurostat, 2017). Hence, the challenges are to meet the higher demand for healthcare, adapt health systems to the needs of an aging population, and ensure that they receive adequate nutrition while living in the community. This is often via a “meals-on-wheels” type service at home, or meals in a “community center” prior to the elderly moving into a residential home where meals would be provided for them. “Meals on wheels” are provided in a number of countries, for example Australia, the United States, and the United Kingdom, where they are invariably supplied through a commercial operator. Often, these meals are frozen, and then regenerated by the recipient in their home, but this does demand facilities such as a microwave or oven, and the ability to follow instructions. The Home Meal Freezer Scheme (HMFS) has a number of advantages over conventional daily hot meal delivery. Meals can be selected from a range held in the freezer, and can be eaten as and when the individual is hungry. As they are not held “hot” they do not suffer any “physical” damage, deterioration in their appearance, or loss of aesthetic appeal. The nutrient profile associated with overlong storage does not suffer, but equally important, the possibility of the meal being stored at an inappropriate temperature is also reduced. The biggest criticism, however, is that as they are delivered weekly rather than daily, there is a lack of social contact, which, for isolated, vulnerable older people, is important (Edwards, Smith, Edwards, & Hartwell, 1999). Of the dependent populations in Europe, 15% receive formal care in institutions, 27% receive formal care at home, and 56% receive informal or no care (European Social Network, “Services for older people in Europe”, 2008).

13.6.6 Prisons It is estimated that, as of October 2015, the worldwide prison population was in excess of 10.35 million, and on average, the United Nations estimates that the national prison population level is 144 inmates per 100,000 in population (Walmsley, 2017). Nations have different approaches as to how those incarcerated should be fed; these range from a very basic diet that often does not meet the nutritional requirements of those incarcerated, to a much more benevolent approach. Even in first-world countries, it is suggested that that some prisons fail to meet the nutritional recommendations (Prison Voice Washington, 2017). Often, in the United Kingdom, prison foodservice is comprised of a meal tray served directly to a prisoner in his or her cell; there being little or no choice. In other countries (Edwards, Hartwell, & Schafheitle, 2009), inmates are provided with three meals per day, and are able to choose their meal in advance, which is then collected from a service counter and consumed either in a dining room or taken back to the cell. It is often pointed out, when considering the feeding of prisoners, that the punishment is being incarcerated for the period of the sentence, not the food which is served, hence meals should be of similar quality found in other institutional foodservice settings.

268

Context

13.6.7 Military In barracks, on a base or on a ship, meals tend to be either a self-service style cafeteria or restaurant style service, the provisions increasingly being provided by contractors—even in operational theaters. Where these meals cannot be provided, such as on operations, meals are usually provided using packaged operation rations. Twenty years of testing in the field has consistently revealed that food intake is inadequate when packaged military rations are fed as the sole source of food. Food intake is much lower, and there is a loss of body weight. Conversely, when these rations are fed to students or military personnel for periods ranging from 3 to 42 days in a cafeteria-like setting, food intake is comparable to levels of a control group provided with freshly prepared food. Under these conditions, body weight is maintained (Hirsch, Kramer, & Meiselman, 2004). The feeding environment is therefore clearly important, however, the potential variables are myriad. The effort associated with consuming a meal in the field, and the nature of the social environment, are two contextual variables that have been examined in both environments, and the parallel effects they exert on food intake underscores their potential importance for an understanding of food intake in the real world (Hirsch et al., 2004).

13.7

Consumer attitudes and expectations of institutional foodservice, issues, and challenges

13.7.1 The menu The menu offered in institutional foodservice has historically been associated with and characterized by a very limited food choice. In most cases, the menu tends to operate on a fixed-term cycle, for example, 3 weeks, where the menu repeats itself. This is cost effective in that the operator can purchase foods in advance, staff and workloads can be preplanned; although some flexibility is also needed to avoid menu fatigue. This can also be an advantage as diners will know what the daily offering is, and could choose to make alternative arrangements if they didn’t like the menu. Menu selection takes place at the point of service as opposed at the table, giving diners the advantage of actually seeing what is being offered. Against this, when the menu is offered at the table, the diner has to rely on the description of the dish or any photographs, if supplied. The importance of dish description has been shown (Wansink, Painter, & Ittersum, 2001) where a description can influence not only the choice of food and then acceptability, but also other aspects, such as, perception of the establishment and value for money. It has also been shown that “attractive” names can increase vegetable consumption in schools (Wansink, Just, Payne, & Klinger, 2012). Menus are invariably limited in their physical size, and therefore limited in the amount of information that can be offered. This could be overcome with the use, for example, of a barcode or quick-response (QR) code, which could provide information such as menu description, photograph, nutritional content, and allergy advice. Such a program is currently being developed for the European Union for use on a smartphone (Foodsmart, 2017).

The effects of environment on product design and evaluation

269

13.7.2 Food choice A lot of academic research has been conducted into food choice, but interestingly, at work, Thomas, Ribera, Senye-Mir, and Eves (2016) established that the most important factors influencing choice were: cost, time, taste, ability to satisfy hunger, and the food environment. Cost was the most important factor, because many participants thought that healthy foods were more expensive. Time to eat and the food environment were also important because work schedules, long queues (lines), and the availability of poor nutritional choices meant that participants found it difficult to eat a healthy lunch at work. Special dietary requirements, food allergies, intolerance, and restricted diets were influential factors for women in this study. Health was a consideration, but only for normal weight participants. For many participants, providing information about calorie content and reducing cost would facilitate healthy choices when food was available for purchase at work. It has been claimed that a lack of transparency within the food chain hinders consumers when making rational food choices (Holle, 2013). In order to aid consumer choice, policy intervention has led to the introduction of information provision such as mandatory nutrition labeling (Lusk & Marette, 2012). Even though nutritional labeling aims to inform and encourage better food choices, its impact on the food intake of healthier products has been limited (Westenhoefer, 2013). It is recognized, however, that there is a call for more information provision on the side of the consumer. Nevertheless, data that are communicated are often not understood as consumers struggle to process this, and have little understanding of concepts such as traceability (Van Rijswijk & Frewer, 2012). Consequently, information processing alongside habitual elements of food choice and eating need to be taken into account in order to understand how further food information can be provided in a meaningful way (Westenhoefer, 2013). Furthermore, Holle (2013) questions whether there is a duty of food producers to provide consumers with sufficient information in a meaningful way, or whether it is the duty of the consumer to become information literate and actively seek information; hence, this debate is lively and current. Influenced by food policy and governance, food information can be delivered to consumers in a dry and factual manner. However, providing it this way has been suggested to be overruled by consumers’ emotions and habits (Sunstein, 2013). Nutrition labeling aims to provide information in a simple way to enable informed and healthier food choices (Souiden, Abdelaziz, & Fauconnier, 2013). Simultaneously, nutrition labeling can further product knowledge and decrease research costs (Berning, Chouinard, McCluskey, Manning, & Sprott, 2010). In regard to the availability of processed food, the United Kingdom is one of the most developed European markets and therefore, nutritional labeling is highly prevalent (Hodgkins et al., 2012).

13.7.3 Consumer attitudes and expectations There is often a perception, arguably false, that institutional foodservice is of a lower standard than other eating out establishments. This view is supported by the work of authors such as Warde and Martens (2000, p 36), who suggest that:

270

Context

… such outlets rarely deliver highly valued eating experiences, although they do provide a major source of eating away from home.

Consumer expectations of inferior quality of food are often based on previous experiences, but accept this is due to time constraints and the convenience of eating onsite. The influence of convenience over other factors affecting food choice has previously been recognized, and plays an important role in the selection of food at work (Kamphuis, de Bekker-Grob, & van Lenthe, 2015). Furthermore, past issues in the food chain can influence consumer perceptions of the quality of food served. Although this influence is short-lived, it can lead to a temporary cessation of consumption of certain food groups, such as processed meats. Consequently, foods high in salt and saturated fats, such as chips and fried foods, are chosen based on the assumption that they are safe to eat, and additionally, will taste good (Price, Hartwell, Hemingway, & Chapleo, 2016). Consumers’ attitudes toward institutional foodservice in the workplace in the United Kingdom (n ¼ 561 aged 16 + both full-time and part-time employees) is summarized in Table 13.2.

13.7.4 Institutional stereotyping The strength of the (negative) views held by consumers toward institutional foodservice has been clearly demonstrated in a series of studies by Cardello, Bell, and Kramer (1996). When asked what they thought about institutional food, such Table 13.2 Attitudes toward UK workplace foodservice facilities

A packed lunch offers better value for money than a canteen meal Offering in-house foodservice facilities is a good way for employers to show they value their employees I prefer to leave the office/workplace at lunchtime I would use a workplace canteen if one was available A workplace canteen would improve communication within my company The quality of canteen food is not as good as other lunch time food outlets

Agree

Neither agree nor disagree

Disagree

%

%

%

63

33

4

48

43

9

41

45

14

39

41

20

31

49

20

23

65

12

Source: Mintel, (2017). Contract catering, 2017. Available from: http://academic.mintel.com/display/749847/# (Accessed 26 October 2017).

The effects of environment on product design and evaluation

271

as that found in school cafeterias, hospitals, military dining halls, on airlines, and so forth, respondents invariably held negative views on the quality and acceptability of that food, which in turn affected consumers’ actual food ratings. This phenomenon was termed “institutional stereotyping,” leading Cardello and colleagues to conclude, inter alia, that: l

l

l

l

l

consumers hold strong negative attitudes about both the quality and acceptability of institutional foods; the primary causes for these attitudes are poor variety, poor food presentation, and poor physical dining setting; when compared with other commercial food, institutional food was perceived as being much poorer in sensory, as opposed to non-sensory, characteristics, due in part to negative media exposure at an early age, although initial exposure to the food can modify or reinforce these attitudes; the association of a food with institutional foodservice will decrease its expected liking, although actual acceptance ratings of the food rarely match established expectations, and these expectations of acceptability can affect actual like/dislike for the food when eaten consumer expectations of acceptability can affect actual like/dislike for the food when eaten; lowering its acceptance when expectations are low and raising its acceptance when expectations are high.

It is further suggested that institutional foodservice may be better served by addressing the causes and potential solutions to poor consumer attitudes and expectations for institutional food (the context), rather than by continued efforts to improve the intrinsic quality of foods, which may already be quite high. Following this, a number of initiatives have been assessed to ascertain if they can alleviate the shortcomings identified by Cardello et al. (1996). In hospitals, one example is a foodservice concept “At Your Request”, which enables hospital patients to choose what they eat, and when they eat. In one study (Doorduijn, van Gameren, Vasse, & de Roos, 2015), patients were able to order food and drinks from a menu card, between 7 a.m. and 7 p.m., by placing a telephone call to trained operators in the nutrition call center. The operator then entered the order, which sent the data to the hospital kitchen. Kitchen staff prepared the order on a tray, which was then delivered to the patient within 45 min. Patients who required an energy- and protein supplement could still do so by choosing them from a special menu card. Results show that patients were more satisfied with the meal service than the traditional system with the meal service and were able to maintain their nutritional status and food intake.

13.7.5 The dining environment Once food has been chosen and served at the service counter, diners carry their own trays and food to self-selected tables. Diners are less likely to linger at the table, and once they have finished, they clear their own plates and trays to a central area, before leaving. Limited staff are usually available to keep tables clean and clear any abandoned crockery and cutlery. The dining environment, the context where food is consumed, is important in any eating out environment, and no less so in institutional foodservice. Unfortunately, this

272

Context

importance is often overlooked, or not fully appreciated. The dining environment in institutional foodservice tends to be less luxurious, and more functional in appearance and operation, with aspects such as decor, including flooring, tending to be hardwearing, and easier to clean, and keep clean, with a minimum number of staff. Notwithstanding, the context of the dining room experience is crucial to the enjoyment of the meal; a factor that invariably is recognized by “forward thinking” foodservice operators, who give more emphasis to the dining room; the importance of which has been supported by academic research. The relative importance and influence of the dining room context, in relation to various types of foodservice operations, has been demonstrated in a number of academic studies. It has been shown, for example, that the context, including the meal context, physical and social environment, and choice can significantly impact the acceptability rating of food (King, Weber, Meiselman, & Nan, 2004; Webber, King, & Meiselman, 2004). This is further supported in other studies where identically prepared food was served to students in a university training restaurant, and in a student refectory in the United Kingdom (Meiselman, Johnson, Reeve, & Crouch, 2000). In each location, the dishes were identical (US Military field rations), and freely chosen from a “typical” menu. Once the meal was finished, diners were asked to rate acceptability on a 9-point hedonic scale (1 ¼ “Dislike Extremely” to 9 ¼ “Like Extremely”). Results are shown in Table 13.3. Gibbons and Henry (2003) prepared and served identically prepared meals to older people (mean age 74.3 years) in two different environments; a university training restaurant, and a staff canteen. The university restaurant has a 5-star appearance with silver service, while the canteen environment was less grand, with simpler decor and limited waiter service. Diners were served a two-course meal from a preset menu designed to be appealing to an older group in each environment. Both males and females consumed significantly more energy in the “improved” restaurant environment.

Table 13.3 Mean acceptability ratings of food service in different locations Refectory

Dish

Beef and Green Pepper Stew Beef Casserole Diced Potatoes Parboiled Rice Chocolate Mousse Apple and Cinnamon Overall

Restaurant

n

Mean

n

Mean

20 12 20 12 15 14 32

6.25 6.42 5.45 5.92 4.73 5.93 6.16

10 8 10 7 12 9 18

6.50 6.63 5.10 7.43 6.50 6.22 7.06

Source: Meiselman, H. L., Johnson, J.L., Reeve, W., & Crouch, J.E., (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35, 231–237.

The effects of environment on product design and evaluation

273

This idea was extended still further when again, identically prepared food was served in 10 locations representing different styles of foodservice establishments (Edwards, Meiselman, Edwards, & Lesher, 2003). Chicken a` la king and rice was produced centrally using the sous vide method, and distributed to each location under chilled conditions. At each location, the food was reheated using local facilities and served to customers using “local” customs and procedures. In all but two of the locations (freshman’s buffet and private part), customers freely chose the dishes, and once the food had been consumed, they were asked to rate the acceptability on a 9-point scale anchored around 1 ¼ “Dislike Extremely” and 9 ¼ “Like Extremely.” Results from this research are presented in Table 13.4. What is clear from this series of studies is that, although in each context there were diners who would usually be found in those locations, the results support the notion of institutional stereotyping. As the location and associated style of service “improves” or moves away from institutional foodservice, the acceptability of the food also improves.

13.7.6 The dining room in context Various terms are used to describe the context of the dining room experience, ranging from “facility based effects,” such as atmospherics (Kotler, 1973), environmental psychology, and servicescapes (Bitner, 1992). In addition, a number of authors, commencing with Campbell-Smith (1967) in Europe, and others, such as Hoffman and Bateson (1997), and Turley and Milliman (2000), have attempted to list or categorize what constitutes the dining room context or atmosphere. A summary of these

Table 13.4 Ratings of overall acceptability of Chicken a` la King and rice Overall acceptability Location/situation

Mean

Army Training Camp University Staff Refectory Private Boarding School Freshman’s Buffet Private Party Residential Home (elderly) Student Refectory Day Care Center (elderly) University 4-star Restaurant Hotel 4-star Restaurant

6.6 6.6 6.7 6.7 7.0 7.1 7.1 7.1 7.6 7.6

n* a a a a ab ab ab ab b b

43 36 88 83 77 43 33 33 19 32

Means a are significantly different (P .05) from b . *4 subjects did not rate overall acceptability. Source: Edwards, J.S.A., Meiselman, H.L., Edwards, A., & Lesher, L., (2003). The influence of eating location on the acceptability of identically prepared foods. Food Quality and Preference, 14, 647–652.

274

Context

Fig. 13.5 Contextual influences in an institutional dining room.

attributes is given in Fig. 13.5, although the relative importance of these variables in relation to each other is difficult to establish. A number of studies have been undertaken to assess the importance and effects of the dining room context; a brief resume of some is given in the following sections:

13.7.6.1 Time available The time available for a meal has an effect on the amount of food consumed; and the “lunch hour” for many workers is the exception rather than the rule. In Paris, 77% of French workers still, it seems, are able to take >30 min for lunch; compared with 27% in the United Kingdom, and 37 min for Europe as a whole (Eurest European, 2016). Tokyo has less, around 20 min, but on the other hand, in countries such as Mexico, 90 min, and Nairobi, Kenya, 2 h, do not seem to be uncommon (Financial Times, 2016).

13.7.6.2 Length of wait (queuing) Waiting or queuing is seen as being potentially central to the total customer experience for two basic reasons (Dawes & Rowley, 1996): l

l

It is often the first encounter that the customer has with the service provider. It is an identifiable and memorable part of the total experience, because there may be little else to occupy the customer.

Studies have quantified how queuing affects an individual’s perception of the food, and the food service system. Meiselman (1979) investigated the importance of a

The effects of environment on product design and evaluation

275

Table 13.5 The relationship between queuing time and stated importance Location

Length of wait in the queue (minutes)

Importance ranking of the wait

1 2 3 4 5 6 7

4.19 4.73 5.45 5.74 5.24 6.00 8.23

12 9 7 8 4.5 1 1

Note: Rankings are 1–14. Source: Meiselman, H.L., (1979). Determining consumer preference in institutional food service. In Livingston, G.E. (Ed.), Food service systems analysis, design and planning (pp. 127–153). London: Academic Press.

reported delay, and the stated importance of waiting in an institutional setting. Results show that the longer a person waits for a meal, the more “serious” the problem is rated, Table 13.5. Similarly, Edwards (1984) showed how making people wait actually affected the acceptability of the food selected. As individuals arrived at the entrance of a selfservice cafeteria, the first group was allowed to enter immediately. Subsequent groups were made to wait, with no explanation given as to the cause, for three, six, and nine minutes. Each group proceeded, as usual, to the service counter, and selected from eight main meal components, a meal of their own choice. They then found their own table, where they were given a questionnaire, and asked to record what they had selected and when finished, and to rate each food component using a nine-point hedonic. The experiment was stopped as soon as the full range of meal choices was not available. Results, shown in Table 13.6, clearly indicate, with one exception, how the longer individuals are made to wait, the lower the rating of the food.

Table 13.6 The effects of delay on the preference ratings of food Delay

Chicken

Roast potatoes

Boiled potatoes

Carrots

(minutes)

Mean

SD

Mean

SD

Mean

SD

Mean

SD

No delay 3 6 9

7.28 6.94 6.75 6.50 N ¼ 62

1.32 2.01 1.88 2.68

6.42 6.16 5.05 4.44 n ¼ 98

1.67 1.68 2.46 2.42

6.27 4.50 3.67 3.42 n ¼ 32

2.28 2.56 2.25 2.51

6.76 6.11 6.14 6.20 n ¼ 50

0.97 2.89 1.79 1.62

Notes: 1 ¼ Dislike extremely; 5 ¼ Neither, nor; 9 ¼ Like extremely. Source: Edwards, J.S.A., (1984). The effects of queuing on food preferences. International Journal of Hospitality Management, 3(2), 83–85.

276

Context

Table 13.7 Mean ratings for like/disliked and no music. Factor

Liked music

Disliked music

No music

Perceived duration of the wait in minutes Emotional evaluation of the service encountera Emotional response to the waita Approach behavior toward the service organizationa

7.02

5.94

4.93

4.70

4.29

3.33

3.71 3.87

2.92 3.22

2.95 2.87

a

Ratings 1–7; higher ratings imply a more positive evaluation. Source: Hui, M.K., Dube, L., & Chebat, J.-C., (1997). The impact of music on consumers’ reaction to waiting for services. Journal of Retailing, 73(1), 87–104.

On the other hand, the speed of service and time spent waiting in a queue in a cafeteria could be advantageous, and influence what is chosen. Spending longer in a queue can lead to higher nutrient values for the meals chosen, although it is not entirely clear why (Lieux & Manning, 1992). Hui, Dube, and Chebat (1997) investigated the effects of playing no music, and music that was both liked and disliked, on the perceived length of waiting time, along with the emotional response to the wait, an evaluation of the service encounter, and the approach behavior to the wait. Results are summarized in Table 13.7, and show that by playing music that is liked, customers perceive the wait as being longer, but are more positive to the emotional evaluation of the service environment, the response to the waiting time, and approach behavior to the service encounter.

13.7.6.3 Effort In a self-service cafeteria, the diner provides part of the labor, such as carrying food to the table, then, clearing once finished. The amount of labor provided by the consumer will diminish as the level of service, and price, rises, and where consumers provide part of the labor, the “amount of effort” that the diner has to expend has the ability to affect what items are chosen. Meiselman, Hedderley, Staddon, Pierson, and Symonds (1994) demonstrated this in two studies involving increased effort to obtain crisps (potato chips) and confectionery. In the first study in a university refectory (cafeteria), food could be selected from several service points, including two hot counters, a sandwich bar, and beverage counter. Items were individually priced, and once students had chosen and collected their meal, they paid for it at one of four cash points. Part of the way down (20 m) the dining room was another small snack bar with drinks and its own cash point. During the baseline study, confectionery items were located at each of the cash points. During the subsequent weeks, they were moved to the small snack bar, which required diners to walk and join another queue and pay at the separate cash point. Cashiers at the four points told diners where to find the confectionery. Students who took part were not aware of the true reasons for the study, and thought they were taking part in a

The effects of environment on product design and evaluation

277

Table 13.8 Average nutrient intake of students Energy

Week

Protein

Fat

CHO g

kcal

kj

g

g

691 723

2905 3035

23.4 26.1

31.2 33.2

84.4 84.9

3661 2881

24.5 21.7

43.0 34.1

103.2 77.9

All students Baseline Experimental

Confectionery-eating students Baseline Experimental

872 686

Source: Adapted from Meiselman, H.L., Hedderley, D., Staddon, S.L., Pierson, B.J., & Symonds, C.R., (1994). Effect of effort on meal selection and meal acceptability in a student cafeteria. Appetite, 23(1), 43–55.

general survey. Results, summarized in Table 13.8, illustrate how effort to obtain confectionery resulted in a significant reduction in the amounts consumed, but with a corresponding increase in the number of cooked desserts and fruit chosen. In a follow-up study, where other foods, such as crisps (potato chips) were included, the selection of crisps dropped dramatically when additional effort was involved. Students, in the main, chose to substitute with another starch dish Meiselman et al. (1994). A similar effect has also been shown with drinking. Engell, Kramer, Malafi, Salomon, and Lesher (1996) observed diners where water was available, ad libitum, either on the table, 20 ft, or 40 ft away. Where water was available on the table, people drank significantly more, 444  259 g, than when it was 20 ft away, 197  100 g, and 187  115 g, when 40 ft away. Consumption of water had no effects on the acceptability of food in any of the conditions. These studies show how eating and choice patterns change when diners are required to expend more effort in obtaining their food and drink. This, in itself, need not necessarily have negative connotations, and effort can be used to manipulate situations to consumers’ advantage. The consumption of salt (sodium chloride) needs, for health reasons, to be reduced, so instead of placing salt shakers on the table in a cafeteria, one strategy could involve placing them at the end of the serving line, where it has been shown to reduce consumption from 0.73 g to 0.53 g per user. Care needs to be taken with sachets of salt that have the opposite effect on consumption, increasing it to between 1.0 and 1.1 g per user (Greenfield, Maples, & Wills, 1983); a phenomenon that is also seen where more effort is needed to use the shaker (Greenfield, Smith, & Wills, 1984).

13.7.6.4 Eating alone or with others (social facilitation) Eating is generally regarded as a social activity, but in many instances, people actually eat alone, or at least, not in the company of others. In the hospital, or at a home for the elderly, people may be surrounded by others in similar circumstances, but for various reasons, may eat alone. Often, in such settings, diners do not choose whom they eat

278

Context

with, or they eat with strangers. The links between number of people present and the amount consumed can be mediated by other factors such as gender of consumers, level of familiarity, social status differences, power and subordination (Feunekes, Graaf, & Staveren, 1995). Studies conducted to determine how eating alone or in the company of others (social facilitation), have shown how they can influence both the amount and type of food consumed, and the time spent in consuming that food. Why social facilitation should be so important is far from clear, although it has been suggested that it might be either conscious or subconscious. It could, for example, be that the presence of others increases levels of arousal and drive; or provides cues as to appropriate or inappropriate behavior (Zajonc, 1965). Alternatively, it could be that when meals are eaten together, more food is provided, and individuals might be hungrier in the presence of others, the context (atmosphere) might be more sociable, the food might taste better, or the meal might last longer (Feunekes et al., 1995). In one of the earliest studies, adult diners (294 male and 245 female) were observed at midday and in the evening, in a variety of both formal dining rooms and fast food operations (Klesges, Bartsch, Norwood, Kautzman, & Haugrud, 1984). The amount of energy consumed was found to be determined by three main effects: gender, where overall, males consumed more than females (835 kcal vs. 716 kcal); the type of restaurant, where greater amounts were consumed in fast food restaurants than more formal dining settings (842 kcal vs. 710 kcal); and the company of others, where more food was consumed when eating in groups rather than alone (828 kcal vs. 742 kcal). What is perhaps important from this early study is that although the type of group, single sex or mixed sexes, did not affect the energy consumed, there was a significant difference in consumption in fast food restaurants where mixed sex groups consumed significantly more than those in single sex groups. Furthermore, females ate less than males in larger groups, but ate similar amounts as the males when eating in smaller groups (Klesges et al., 1984). Social facilitation is not confined to single or formal meals, but can be seen in a complete range of meals served. Strong positive and significant correlations have been shown between meal size, and the number of people present for all meals consumed during the breakfast, lunch, and dinner periods, when eaten in restaurants, at home, and elsewhere, consumed with or without alcohol, for snacks on their own, for meals on their own (de Castro, Brewer, Elmore, & Orozco, 1990). Eating with others has also been shown to influence the speed of food consumption (Rosenthal & McSweeney, 1979), and also, in part at least, and to increase the length of the meal, but not the rate of intake. This is independent of the subjective state of hunger, emotional states of elation and anxiety, and acts independent of the content of the stomach and premeal interval (de Castro, 1990). Not only is the number of people present important, their relationship to the person consuming the meal is also important. In comparison with other eating companions, it has been shown, irrespective of the time of the day, that meals eaten with a spouse and family are larger and eaten faster, while meals eaten with friends were larger but of a longer duration. Male companions had a greater impact for females, but not for males (de Castro, 1993).

The effects of environment on product design and evaluation

279

Table 13.9 The relationship between the number of people present and meal size and the amount consumed Number of people present

Increase in meal size (5%)

2 2 4 5 6 or more

28 41 53 53 71

Source: de Castro, J.M., & Brewer, E.M., (1992). The amount eaten in meals by humans is a power function of the number or people present. Physiology & Behavior, 51(1), 121–125.

The importance and robustness of the claim that the number of people present influences meal intake has also been demonstrated as being linear, described as a power function, clearly illustrating the orderliness and lawfulness of the phenomenon. Using data from a number of studies, the strength of the relationship has been shown (de Castro & Brewer, 1992) and summarized in Table 13.9. The amount consumed, however classified, that is, snacks, breakfast, midday and evening, increases when consumed in the presence of others. It is far from clear as to exactly why consumption should increase, but more recent research suggests that it is a function of the amount of time spent eating (Bell & Pliner, 2003). Meals consumed with others in convivial surroundings last longer, and as a result, more is consumed.

cor 13.7.6.5 De The effects of decor have received comparatively little academic attention, although in one seminal study (Bell, Meiselman, Pierson, & Reeve, 1994), the restaurant decor and ambience were manipulated in order to measure the effects. In a university training restaurant, an Italian menu was offered, and diners were asked to state, among other things, which country; Britain, Italy, France, or simply foreign, they thought the food came from. Over subsequent weeks, the same menu was again offered, but this time the Italian menu was complimented with the restaurant being decorated in an Italian theme, including the use of Italian flags and bunting. Results, where the restaurant had not been decorated, showed that diners were unsure as to which country the food came from. When the Italian decor was added, diners immediately recognized that the food was Italian. Nothing had changed in terms of food, but by manipulating the context, the decor, changes in perception had been brought about. Clearly, therefore, the decor, the context of the dining environment, was having much more of an influence than might at first have been thought. In addition, it is interesting to note that diners were unable to identify the ethnicity of the food, and were only able to ascertain the ethnicity of the food once further stimuli had been introduced; thus demonstrating, in part, internationalization of food. However, other studies in a care home, for example, have demonstrated that factors affecting the context of the meal, such as names of dishes and decor, had little impact

280

Context

on consumption; direct factors such as increased variety on the plate and condiments on the table had greater influence on improving residents’ satisfaction with their meals, and increasing the quantities of meat or vegetables consumed (Divert et al., 2014). It appears that more research is required in this area.

13.7.6.6 Table layout and seating Table and seating layout can be symbolic, and used to indicate status, define personal space, and to regulate privacy and interaction (Robson, 2002). Whenever possible, people choose seats around the edges of the room, and use items such as bags, and other personal effects, to help define and protect their territory and dissuade others from sitting nearby (Robson, 2002). The types of tables, the seating, and their overall configuration also have the ability to influence how long people spend in the dining room, and how much is spent (Thompson, 2003). The average dining duration of customers sitting at anchored seats did not differ significantly from those at unanchored seats, although customers seated at banquettes had a significantly longer dining duration than those seated at other types of tables. The average spend of diners seated in booths, that is, anchored seats, spent significantly more than those seated in other types of seats; although those seated next to the window tended to have a lower average spend (Kimes & Robson, 2004). Cultural differences can also be noted; North Americans prefer seats that are anchored by either a permanent or semi-permanent structure such as booths, columns, or planters; English diners prefer to sit at right angles to each other, whereas Swedish diners tend to sit face to face when interacting, although this, of course, may vary depending on the relationships between the individuals concerned (Robson, 2002). The importance of layout and seating is not confined to a cafeteria, and is just as important when feeding, for example, hospital patients. Malnutrition is prevalent in many hospitals worldwide, but it has been shown that by changing the seating/dining arrangement, patient food intake can be increased (Edwards & Hartwell, 2004). Patients who sat by the side of their beds ate more than those sitting in bed, and those who ate around a dining table in the company of others ate still more.

13.7.6.7 Background music Various music genres and the volume of that music have been studied, for example: the effects of music genre were studied in a cafeteria-style dining room (Feinstein, Hinkston, & Erdem, 2002), where it was shown that Italian music significantly affected the choice of ethnic entrees over non-ethnic entrees. Italian music was shown to positively affect the consumption, not only of Italian food, but also Mexican food. Surprisingly, Mexican music did not significantly affect the selection of Mexican dishes, perhaps because the music was of a similar tempo. Favorable effects have also been found where different pieces of music were played in a cafeteria, and where customers were asked to rate their liking for the atmosphere of the cafeteria, their willingness to return, and their liking for the music.

The effects of environment on product design and evaluation

281

Results show a positive correlation; the more the music was liked, the more customers liked the situation, the more they were attracted to the source of the music, and the more they wanted to return to that situation and (North & Hargreaves, 1996).

13.7.6.8 Odor The “odor” found in the dining room has the ability to affect perceptions of hunger and the amount of food actually consumed. Blackwell (1997) demonstrated this phenomenon by exposing diners to odors (grilled bacon and boiled cabbage), which has been established as being well liked or disliked. Diners were invited to an “all-day breakfast buffet,” where they could freely choose what they could eat and how much; they could return to the buffet as many times as they wished. On some days, the odor was in the dining area was neutral, that is, no odor; on other days, the odor was pleasant (grilled bacon), and on others, the odor was disliked (boiled cabbage). On days when diners were exposed to the odor that was liked rather than neutral or disliked, consumption of all items increased over the days when the odor was disliked. Clearly the pleasantness of the smell had a positive influence on the amount consumed.

13.8

Summary and conclusions

Worldwide, the foodservice industry is one of the largest sectors of many economies, and is one of the main employers, and is set for further expansion, primarily in the Asia Pacific Region; although the Middle East and Africa should also see growth. Institutional foodservice is a comparatively small part of this industry; notwithstanding, it forms an integral part of the total overall food consumption when eating out. While many individuals might have an understanding of what institutional foodservice is, defining the sector can be quite difficult. One possible solution is not to define it classically, as the public sector or private sector, or the people served, but by considering the physical characteristics of the operation; but even here are a number of anomalies. The primary characteristics of institutional foodservice have been outlined and addressed in this chapter. However, both institutional foodservice and eating out in general involve more than the simple ingestion of foods and nutrients, and include a myriad of other factors, all of which have the ability to influence the eating out experience. In essence, the context of institutional foodservice has the potential to be of equal, if not more importance than the food itself. The contextual variables associated with institutional foodservice have been identified in this chapter and, by way of illustration, a number, namely: the time available, length of wait, effort involved, eating alone or with others, decor, table layout, background music and odor have been considered. These characteristics tend to be viewed as single issues, and what has not been taken into account here, or in the literature, is the synergistic effects these might have; hence productive avenues for further research.

282

Context

What is important, when examining institutional foodservice, is to consider not only the food or the meal, but also the customer and the context, that is, the situation in which the meal will be chosen and consumed. Only then will it be possible to ensure that the foodservice provision is a total entity, or a meal experience, and not simply a refueling process.

References Bell, R., Meiselman, H. L., Pierson, B. J., & Reeve, W. G. (1994). Effects of adding an Italian theme to a restaurant on the perceived ethnicity, acceptability, and selection of foods. Appetite, 22(1), 11–24. Bell, R., & Pliner, P. L. (2003). Time to eat: The relationship between the number of people eating and meal duration in three lunch settings. Appetite, 41, 215–218. Berning, J. P., Chouinard, H. H., McCluskey, J. J., Manning, K. C., & Sprott, D. E. (2010). Identifying consumer preferences for nutrition information on grocery store shelf labels. Food Policy, 35(5), 429–436. Bitner, M. J. (1992). Servicescapes: The impact of physical surroundings on customers and employees. Journal of Marketing, 56, 57–71. Blackwell, L. R. (1997). The role of olfactory cues and their effects on food choice and acceptability. PhD Thesis Bournemouth University. Burnett, J. (1979). Plenty and want. A social history of diet in England from 1815 to the present day. London: Scolar Press. Burnett, J. (2004). England eats out. 1830 – present. Harlow: Pearson Education Ltd. Campbell-Smith, G. (1967). Marketing the meal experience. A fundamental approach. Surrey, England: University of Surrey. Cardello, A., Bell, R., & Kramer, F. M. (1996). Attitudes of consumers toward military and other institutional foods. Food Quality and Preference, 7(1), 7–20. Conor, D. S. (2014). Institutional food service. In P. B. Thompson & D. M. Kaplan (Eds.), Encyclopedia of food and agricultural ethics (pp. 1258–1263). The Netherlands: Springer. Curtis-Bennett, N. (1949). The food of the people being the history of industrial feeding. London: Faber and Faber. Dawes, J., & Rowley, J. (1996). The waiting experience: Towards service quality in the leisure industry. International Journal of Contemporary Hospitality Management, 8(1), 16–21. de Castro, J. M. (1990). Social facilitation of duration and size but not rate of spontaneous meal intake in humans. Physiology & Behavior, 47(6), 1129–1135. de Castro, J. M. (1993). The effects of spontaneous ingestion of particular food or beverages in the meal patterns and overall nutrient intake of humans. Physiology & Behavior, 56(6), 1133–1144. de Castro, J. M., & Brewer, E. M. (1992). The amount eaten in meals by humans is a power function of the number or people present. Physiology & Behavior, 51(1), 121–125. de Castro, J. M., Brewer, E. M., Elmore, D. K., & Orozco, S. (1990). Social facilitation of the spontaneous meal size of humans occurs regardless of time, place alcohol or snacks. Appetite, 15(2), 89–101. Divert, C., Laghmaoui, R., Crema, C., Issanchou, S., Van Wymelbeke, V., & Sulmont-Rosse, C. (2014). Improving meal context in nursing homes. Impact of four strategies on food intake and meal pleasure. Appetite, 84, 139–147. Doorduijn, A. S., van Gameren, Y., Vasse, E., & de Roos, N. E. (2015). At Your Request® room service dining improves patient satisfaction, maintains nutritional status, and offers opportunities to improve intake. Clinical Nutrition, 35(5), 1174–1180.

The effects of environment on product design and evaluation

283

Edwards, J. S. A. (1984). The effects of queuing on food preferences. International Journal of Hospitality Management, 3(2), 83–85. Edwards, J. S. A., & Hartwell, H. J. (2004). A comparison of energy intake between eating positions in a NHS hospital – A pilot study. Appetite, 43, 323–325. Edwards, J. S. A., Hartwell, H. J., & Schafheitle, J. (2009). Prison foodservice in England. Journal of Foodservice, 20(4), 157–166. Edwards, J. S. A., Meiselman, H. L., Edwards, A., & Lesher, L. (2003). The influence of eating location on the acceptability of identically prepared foods. Food Quality and Preference, 14, 647–652. Edwards, J. S. A., & Overstreet, K. (2009). What is foodservice? Journal of Foodservice, 20, 1–3. Edwards, J. S. A., Smith, A. P., Edwards, A., & Hartwell, H. (1999). Feeding older people in the community. Bournemouth, Centre for Culinary Research Report. Engell, D., Kramer, M., Malafi, T., Salomon, M., & Lesher, L. (1996). Effects of effort and social modeling on drinking in humans. Appetite, 26(2), 129–138. Eurest European. (2016). Eating at work report. Available from: https://www.slideshare.net/ GaurangPatel28/eurest-eating-at-work-report-2016. Accessed 29 November 2017. Euromonitor International. (2017). Consumer foodservice global industry overview. October 2017. Available from: http://www.portal.euromonitor.com/portal/analysis/tab. Accessed 16 October 2017. European Social Network. (2008). Services for older people in Europe. https://ec.europa.eu/ health/sites/health/files/mental_health/docs/services_older.pdf Accessed January 2018. Eurostat. (2017). People in the EU – statistics on an ageing society. Available from: https://ec. europa.eu/eurostat/statistics-explained/index.php/People_in_the_EU_%E2%80%93_sta tistics_on_an_ageing_society. Accessed 7 December 2017. Feinstein, A. H., Hinkston, T. S., & Erdem, M. (2002). Exploring the effects of music atmospherics on menu item selection. Journal of Foodservice Business Research, 5(4), 3–25. Feunekes, G. I. J., Graaf, C. d., & Staveren, W. A. v. (1995). Social facilitation of food intake is mediated by meal duration. Physiology & Behavior, 58(3), 551–558. Financial Times. (2016). A global anatomy of health and the workday lunch. Available from: https://www.ft.com/content/a1b8d81a-48f5-11e6-8d68-72e9211e86ab. Accessed 29 November 2017. Foodsmart. (2017). Shaping smarter consumer behaviour and food choice. Available from: https://microsites.bournemouth.ac.uk/foodsmart/. Accessed 28 November 2017. Gibbons, M. R. D., & Henry, C. J. K. (2003). The effect of eating environment on energy intake in the elderly. Proceedings of the Nutrition Society, 62, 93A. Greenfield, H., Maples, J., & Wills, R. B. H. (1983). Salting of food – A function of hole-size and location of shakers. Nature, 301, 331–332. Greenfield, H., Smith, A. M., & Wills, R. B. H. (1984). Influence of multi-holed shakers on salting of food. Human nutrition: Applied nutrition, 38A (pp. 199–201). Heinzelmann, U. (2008). Food culture in Germany. Westport: Greenwood Press. Hirsch, E. S. F., Kramer, M., & Meiselman, H. (2004). Effects of food attributes and feeding environment on acceptance, consumption and body weight: Lessons learned in a twentyyear program of military ration research: US Army research (Part 2). Appetite, 44(1), 33–45. Hodgkins, C., Barnett, J., Wasowicz-Kirylo, G., Stysko-Kunkowska, M., Gulcan, Y., Kustepeli, Y., et al. (2012). Understanding how consumers categorise nutritional labels: A consumer derived typology for front-of-pack nutrition labelling. Appetite, 59(3), 806–817. Hoffman, K. D., & Bateson, J. E. G. (1997). Essentials of services marketing. Fort Worth: The Drayen Press.

284

Context

Holle, M. (2013). Consumer model in change - is the consumer today ripe for state care?/ Verbraucherleitbild im Wandel - 1st der Verbraucher von heute reif f€ ur staatliche € Verbraucherschutz und Lebensmittelsicherheit, 8(4), 336–339. Betreuung? Journal fur Hui, M. K., Dube, L., & Chebat, J.-C. (1997). The impact of music on consumers’ reaction to waiting for services. Journal of Retailing, 73(1), 87–104. Jørgensen, M. S., Arsky, G. H., Brandhøj, M., Nyberg, M., Roos, E., & Mikkelsen, B. E. (2010). Eating at worksites in Nordic countries: National experiences and policy initiatives. International Journal of Workplace Health Management, 3(3), 197–2010. Kamphuis, C. B. M., de Bekker-Grob, E. W., & van Lenthe, F. J. (2015). Factors affecting food choices of older adults from high and low socioeconomic groups: A discrete choice experiment. American Journal of Clinical Nutrition, 101(4), 768–774. Kimes, S. E., & Robson, S. K. A. (2004). The impact of restaurant table characteristics on meal duration and spending. The Cornell Hotel and Restaurant Administration Quarterly, 45, 333–346. November. King, S. C., Weber, A. J., Meiselman, H. L., & Nan, L. V. (2004). The effect of meal situation, social interaction, physical environment and choice on food acceptability. Food Quality and Preference, 15, 645–653. Klesges, R. C., Bartsch, D., Norwood, J. D., Kautzman, D., & Haugrud, S. (1984). The effects of selected social and environmental variables in the eating behavior of adults in the natural environment. International Journal of Eating Disorders, 3(4), 35–41. Kotler, P. (1973). Atmospherics as a marketing tool. Journal of Retailing, 49(Winter), 48–64. Lieux, E. M., & Manning, C. K. (1992). Evening meals selected by college students; impact of the foodservice system. Journal of the American Dietetic Association, 92(5), 560–566. Lund, T. B., Kjærnes, U., & Holm, L. (2017). Eating out in four Nordic countries: National patterns and social stratification. Appetite, 119, 23–33. Lusk, J. L., & Marette, S. (2012). Can labeling and information policies harm consumers? Journal of Agricultural & Food Industrial Organization, 10(1), 1–15. Meiselman, H. L. (1979). Determining consumer preference in institutional food service. In Livingston, G.E. Food service systems analysis, design and planning (pp. 127–153). London: Academic Press. Meiselman, H. L., Hedderley, D., Staddon, S. L., Pierson, B. J., & Symonds, C. R. (1994). Effect of effort on meal selection and meal acceptability in a student cafeteria. Appetite, 23(1), 43–55. Meiselman, H. L., Johnson, J. L., Reeve, W., & Crouch, J. E. (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35, 231–237. National Restaurant Association. (2017). News and research, facts at a glance. Available from: http://www.restaurant.org/News-Research/Research/Facts-at-a-Glance. Accessed 20 October 2010. North, A. C., & Hargreaves, D. J. (1996). The effects of music on responses to a dining area. Journal of Environmental Psychology, 16, 55–64. Nyberg, M., & Olsen, T. D. (2010). Meals at work: Integrating social and architectural aspects. International Journal of Workplace Health Management, 3(3), 222–232. Price, S., Hartwell, H., Hemingway, A., & Chapleo, C. (2016). Workplace foodservice; perception of quality and trust. Appetite, 97, 169–175. Pridgeon, A., & Whitehead, K. (2013). A qualitative study to investigate the drivers and barriers to healthy eating in two public sector workplaces. Journal of Human Nutrition and Dietetics, 26(1), 85–95. Prison Voice Washington. (2017). Correcting Food Policy in Washington Prisons. Available from: http://prisonvoicewa.org/content/CorrectingFoodPolicy-2016-10-25.pdf. Accessed 9 November 2017.

The effects of environment on product design and evaluation

285

Quintiliani, L., Poulsen, S., & Sorensen, G. (2010). Healthy eating strategies in the workplace. International Journal Of Workplace Health Management, 3(3), 182–196. Raulio, S., Roos, E., & Pr€att€al€a, R. (2010). School and workplace meals promote healthy food habits. Public Health Nutrition, 13(6A), 987–992. Robson, S. K. A. (2002). A review of psychology and cultural effects on seating behaviour and application to food service settings. Journal of Foodservice Business Research, 59(2), 89–107. Rosenthal, B., & McSweeney, F. K. (1979). Modelling influences on eating behavior. Addictive Behavior, 4, 205–214. Souiden, N., Abdelaziz, F. B., & Fauconnier, A. (2013). Nutrition labelling: Employing consumer segmentation to enhance usefulness. Journal of Brand Management, 20(4), 267–282. Sunstein, C. R. (2013). Simpler: The future of government. New York, NY, US: Simon & Schuster. Thomas, E. L., Ribera, A. P., Senye-Mir, A., & Eves, F. F. (2016). Promoting healthy choices in workplace cafeterias: A qualitative study. Journal of Nutrition Education and Behavior, 48(2), 138–145. Thompson, G. M. (2003). Optimizing restaurant-table configurations: Specifying combinable tables. Cornell Hotel and Restaurant Administration Quarterly, 44, 53–60. February. Turley, L. W., & Milliman, R. E. (2000). Atmospheric effects on shopping behaviour: A review of the experimental evidence. Journal of Business Research, 49, 193–211. Van Rijswijk, W., & Frewer, L. J. (2012). Consumer needs and requirements for food and ingredient traceability information. International Journal of Consumer Studies, 36(3), 282–290. Walmsley, R. (2017). World prison population list (eleventh edition). Available at: http://www. prisonstudies.org/sites/default/files/resources/downloads/world_prison_population_list_ 11th_edition_0.pdf. Accessed 8 November 2017. Wansink, B., Just, D. R., Payne, C. R., & Klinger, M. Z. (2012). Attractive names sustain increased vegetable intake in schools. Preventive Medicine, 55(4), 330–332. Wansink, B., Painter, J., & Ittersum, K. V. (2001). Descriptive menu labels’ effects on sales. Cornell Hotel and Restaurant Administration Quarterly, 42(6), 68–72. Warde, A., & Martens, L. (2000). Eating out. Social differentiation, consumption and pleasure. Cambridge: Cambridge University Press. Webber, A. J., King, S. C., & Meiselman, H. L. (2004). Effects of social interaction, physical environment and food choice freedom on consumption in a meal-testing environment. Appetite, 42, 115–118. € Westenhoefer, J. (2013). Influencing eating behavior through food labeling? Journal fur Verbraucherschutz und Lebensmittelsicherheit, 8(4), 327–329. Zajonc, R. B. (1965). Social facilitation. Science, 149(3681), 269–274.

Further reading Cushman, & Wakefield. (2017). The global food and beverage market. Summer 2017. Available from: https://www.dtz.nl/media/449353/global-food-beverage-report_final.pdf. Accessed 30 November 2017. Mintel (2017). Contract catering, 2017. Available from: http://academic.mintel.com/display/ 749847/#. Accessed 26 October 2017.

This page intentionally left blank

The effect of context on children’s eating behavior

14

Monica Laureati, Ella Pagliarini Department of Food, Environmental and Nutritional Sciences (DeFENS), University of Milan, Milan, Italy

14.1

Introduction

Contextual factors may have a strong impact on food perception and liking, especially when children are involved in sensory and consumer testing. We would like to start this chapter by reminding readers of the definition of ecological validity reported in Chapter 1. “Ecological validity, in psychology, is a measure of how test performance predicts behaviors in real-world settings. Although test designs and findings in studies characterized by low ecological validity cannot be generalized to real-life situations, those characterized by high ecological validity can be. The usefulness of ecological validity as a concept, however, has been much debated, with some questioning the importance of psychological realism (that is, how much processes appearing in the experiment mirror those in everyday life).” The concept of ecological validity is essential when working with children as consumers, because during development, the individual shapes his/her personality, as well as food preferences, by interacting with the environment. It is evident that children are more easily influenced than adults are by extrinsic factors (e.g., the testing location and people present during the execution of the test). For this reason, it is especially important to consider the appropriate setting to conduct sensory testing with children in order to obtain ecologically valid results. Within this chapter, we will stress the importance of not considering context only as a “physical entity,” but rather as a multi-factor system, including the food and/or drinks consumed, and the social interaction with people. In this vein, according to review articles and empirical studies dealing with context effects on adults’ food preference and choice (King, Weber, Meiselman, & Lv, 2004; Meiselman, 2009), we have structured the chapter in three sections, each of which is dedicated to one of the main factors considered relevant to the food context (Fig. 14.1): (1) the meal, (2) the physical environment, and (3) the social environment. In this introductory section, it is wise to provide some indication about the use of the term “children,” which includes a somewhat wide age range. Based on the individual’s motor and cognitive skills, the ASTM guidelines (ASTM, 2013) categorizes children as infants (from birth to 18 months), toddlers (from 18 months to 3 years), preschoolers (from 3 to 5 years), early readers (5–8 years), pre-teens (8–12 years), and teenagers (12–18 years). It is well recognized that sensory and consumer science Context. https://doi.org/10.1016/B978-0-12-814495-4.00014-3 Copyright © 2019 Elsevier Inc. All rights reserved.

288

Context

The meal • Sensory properties • Portion and container size • Food variety • Appropriateness of the eating occasion

The physical environment

The social environment

• Appropriate location and setting (e.g., familiar environment) • Imaginative context and gamification

• People present at the experiment (e.g., modeling from peers, experimenter, parents, teachers)

Age and gender-related differences

Fig. 14.1 Schematic overview of contextual factors influencing children eating behavior.

procedures should be carefully used according to children’s developmental phase (Guinard, 2001; Laureati, Pagliarini, Gallina Toschi, & Monteleone, 2015), and that the perception of the food context and the factors connected to it may change considerably, depending on the child’s age. In this chapter, we review articles dealing with contextual factors that are most influential on children’s sensory and hedonic response, limiting the age range to 3–18 years. On one hand, we decided not to include infants and toddlers, because in this age range, the food context is limited to the parents and the family; while from 3 years old, children generally start attending kindergartens or recreational centers, and they start selecting the food they want to eat, and with whom. In other words, they start interacting with the food environment. Therefore, contextual factors that can contribute to shape their food preference and choice increase. On the other hand, although it is assumed that individuals aged 13–18 years start reasoning and performing sensory and consumer testing similarly to adults (Guinard, 2001; Laureati et al., 2015), we included teenagers in our review because in this period of life, social interaction is a highly influential factor when choosing what to eat and drink.

14.2

The meal

There are many factors associated with meal consumption that can influence children’s food acceptability and selection, including its sensory properties, the expectation that the meal can taste good or bad, the appropriateness of it for a certain period of the year or of the day, how it is served, and whether it is presented alone or in combination with other meal components (i.e., food variety). The influence of sensory properties on children’s food choice has been studied in a number of empirical and review articles (Laureati et al., 2015). It is widely known that children have an innate preference for sweet taste and an innate rejection for all that is bitter or sour (Mennella, 2014), and it is also widely recognized that texture is an important factor in children’s food acceptance and refusal (Szczesniak, 2002).

The effect of context on children’s eating behavior

289

Although these topics are interesting, and important for developing healthy and wellliked food by children, they are not the focus of this chapter. Therefore, in this section, we will consider the factors most strictly related to the food context, that is, how and when a meal is presented. This translates into three main topics: (1) the portion and container size, (2) the food variety, and (3) the appropriateness of the eating occasion. In recent years, food portions size has been under discussion as larger portions are considered one of the causes of the increase in food and energy intake worldwide. The increase in portions served during meals is, unfortunately, the result of changes in the eating environment, as well as in eating behavior during the past century. These changes are a big issue in view of the increasing obesity epidemic that is being observed among children and adults (WHO, 2014). Systematic studies on adults have shown that providing individuals with larger portions leads to substantial increases in energy intake for both amorphous (e.g., pasta with cheese) and unit foods (e.g., sandwiches), as well as beverages (Rolls, 2014). These effects are partially explained by size-related visual cues and inflated consumption norms (Mathias et al., 2012). The effect of portion size on food intake (the bigger the portion, the more is eaten) has been shown also in children of different ages. In general, findings from current research reveal that serving large portions of palatable foods can increase food intake by 25% to 60% among 5- to 9-year-old children, regardless of whether the portions are served directly by the child or by the mother (see Birch, Savage, & Fisher, 2015 for review). Recently, these effects have been used in an attempt to increase consumption of healthy and often disliked foods, such as fruits and vegetables. For example, Mathias et al. (2012) found that when portions were doubled, children increased their vegetable (cooked broccoli) and fruit (peaches) intake by 37% and 70%, respectively. Effects were not seen in children who disliked the items, a finding that is in agreement with other research (Kral, Kabay, Roe, & Rolls, 2010), and suggests that optimization of food formulations, instead of portion size, should be used as a strategy to improve children’s acceptance of healthy food. In fact, serving larger fruit and vegetable portions at meals may also have important negative side effects, including substantial plate waste (Birch et al., 2015) and a distorted perception of meal portions that may even contribute to the opposite effect of increasing food intake overall. In view of these important limitations, we recommend focusing on more sustainable principles, such as experiential learning, repeated exposure, and modeling and reward to guide children toward healthy eating behaviors (Laureati, Bergamaschi, & Pagliarini, 2014). Another strategy for modifying children’s food selection and consumption that can be considered more sustainable than increasing meal portion is based on the modulation of the plate/container size in which the food is served. There is a growing belief that the size of dishware influences how much people serve and consume during a single meal occasion. A recent experimental study demonstrated that adults perceived a given food portion to be smaller when the plate size and the amount of empty space on the plate were increased (Van Ittersum & Wansink, 2012). Similarly, DiSantis et al. (2013) found an effect of dishware size on children’s

290

Context

actual food serving and consumption, with larger plates and bowls increasing children’s self-served portions. Size-related visual cues are thought to be at the basis of the mechanisms underlying these effects. On one hand, larger dishware may indeed suggest inflated norms for consumption that translate into larger self-served portion sizes (Wansink, 2004). On the other hand, larger dishware may induce a size-contrast illusion, that is, the so-called Delboeuf illusion, and alter children’s visual perception of portion size. In other words, they perceive a similar amount of food to be smaller on a large plate or in a large container. A schematic representation of the Delboeuf illusion, which has been proposed by Van Ittersum and Wansink (2012) to explain dishware size effects on individuals’ eating behavior, is depicted in Fig. 14.2. Wansink and coworkers have also shown that the package or container size effect may affect the amount people consume, regardless of the sensory properties of food. In a study conducted in a movie theater, adult consumers were given a medium or a large

Fig. 14.2 The Delboeuf illusion: (A) two identical circles are perceived of different size when surrounded by a much larger and a slightly larger circle; (B) the same amount of food is perceived differently when served on a large vs small plate (B).

The effect of context on children’s eating behavior

291

pack of popcorn that was either fresh or stale. People ate respectively 45.3% and 33% more of fresh and stale popcorn when it was given to them in large containers, indicating that perceived taste and quality had little impact on how much popcorn was eaten, while the package size was the best predictor for intake (Wansink & Kim, 2005). Interestingly, Aerts and Smits (2017) have recently replicated this experiment with children (age 3–7). In a first experiment, they presented children with either regular or large cups of salted or sugared popcorn while watching a movie. In a second experiment, they used a less palatable food (baby carrots) or a more palatable one (cookies) served in either regular or large packages during breaks at school. In both experiments, they found that children ate more from large packages compared with regular packages, and they ate more of the sugared food compared with the less tasty one. Furthermore, the package size effect was stronger for sugared foods in both experiments, an indication that for children, both sensory properties and container size play an important role in food intake. In view of the strong effect of portion, dishware and package size on consumer eating behavior and health, further research is needed, especially on children, in order to better understand the mechanisms behind the relationship of larger size and higher consumption, and to implement effective and efficient interventions to promote healthy growth and development. Another important aspect related to how the meal can influence food preference and selection is whether it is presented alone or in combination with other components (i.e., food variety). Most of the meals consumed in various contexts, for example, in a school canteen or at home, include several components, and rarely is a food served alone. Sensory evaluation of food performed in a laboratory context, which requires products being evaluated monadically, is, of course, an exception, but this situation is considerably far from being representative of habitual meal consumption. Previous research has shown that individual food item preferences are not predictive of the preference for item combinations (King et al., 2004). Despite this, very little information is available about acceptability of a food or beverage product presented within a complete meal, especially in children (e.g., Caporale, Policastro, Tuorila, & Monteleone, 2009; Pagliarini, Gabbiadini, & Ratti, 2005; Pagliarini, Ratti, Balzaretti, & Dragoni, 2003). Increasing the number of components of a meal or changing its appearance by modifying the shape and/or the color of these components may increase the perceived complexity and variety of the meal. Variety within a meal is known to be one of the most powerful ways to increase energy intake, with larger amounts of food being consumed in meals characterized by high variety (Bergamaschi et al., 2016). Seeking a variety of foods is considered an adaptive trait to protect the organism from nutritional deficiencies (Rolls et al., 1981; Wadhera & Capaldi-Phillips, 2014). At the same time, the effect of food variety can be explained by the mechanism of sensory-specific satiety (SSS). When a variety of foods is available, there is the tendency to switch from one food to another because of the decrease in palatability of the eaten food compared with the uneaten one (Rolls et al., 1981; Rolls, Rowe, & Rolls, 1982).

292

Context

Recently, variety has been used to increase children’s acceptance and consumption of healthy, low-energy dense food, such as fruits and vegetables. Using real food, Roe, Meengs, Birch, and Rolls (2013) tested several familiar fruits and vegetables as snacks with pre-school children, and found that providing more variety increased the amount they actually chose and ate. Relatively little is known about what food variety actually means for children. In a recent study (Bergamaschi et al., 2016), we investigated the effect of two different types of variety on fruit and vegetable snack intake and acceptance in primary school children (age 9–11 years): “classical variety,” that is, serving of different foods and “perceived variety,” that is, serving of the same food in different shapes. For each set, three levels of variety in the servings were tested: low, medium, and high (Fig. 14.3). We found that perceived variety was more effective than classical variety in increasing children’s liking, but, contrary to our expectation, in the classical variety set, the highest intake was observed in the serving with the lowest variety (Fig. 14.4). We explained this outcome by the fact that higher levels of classical variety were obtained by adding items (e.g., cranberry and white cabbage) that were less liked and less frequently consumed by children, suggesting that children’s intake is more affected by acceptability and familiarity of the single stimulus included in the meal than by variety. Other studies have found that shape of food can affect children’s liking. For example, using real food, van Kleef, Vrijhof, Polet, Vingerhoeds, and de Wijk (2014) found that presenting fun-shaped whole wheat bread rolls almost doubled consumption of whole wheat bread in children aged 8–12 years. Due to the strong effect that visual cues can have on children’s eating behavior, we believe it is of crucial importance to deepen the study on how children’s food choice might be impacted by small contextual cues. As such, findings from current research have implications for policies and practices within school meal programs, and suggest that instead of requiring children to eat certain foods, improving the attractiveness of healthier food options could be a promising route.

Classic variety (different food items) Low

Medium

Apple slices Carrot sticks

Apple slices Carrot sticks White cabbage Cranberries

Perceived variety (same foods in different shapes) High

Apple slices Carrot sticks White cabbage Cranberries Almonds Plum

Low

Apple chunks Carrot chunks

Medium

Apple chunks Apple slices Carrot chunks Carrot sticks

High

Apple chunks Apple slices Apple triangles Carrot chunks Carrot slices Carrot sticks

Fig. 14.3 Stimuli used to represent different types of snack variety among primary school children (Bergamaschi et al., 2016).

The effect of context on children’s eating behavior

293

100 90 80 Intake (gr) %

70

a

60

c

cb

b

cb

med

high

low

med

b

50 40 30 20 10 0 low

CV

high

PV Variety

Fig. 14.4 Mean intake values (g) of classical variety (CV) and perceived variety (PV), and their variety levels (low, medium, high); (different letters denote significant differences for P < .05 between and within the different variety conditions (Bergamaschi et al., 2016).

Finally, we will consider the effect of eating occasion appropriateness on children’s food acceptance and selection. As already discussed, the appropriateness of a specific food/beverage for a specific eating context is probably one of the most powerful factors influencing the enjoyment of eating and drinking (K€oster, 2003; PiquerasFiszman & Jaeger, 2015). Bell and Meiselman (1995) reported that by simply changing the perception of the eating environment, the consumer can change what is deemed to be appropriate for that situation, and can substantially change her/his behavior. For example, in an elegant restaurant, it is usually considered less appropriate to order a hamburger, while this choice may be more appropriate in a fast food or informal eating setting. Although children are less influenced by these social norms, and probably would have no hesitation in ordering a pizza in a Michelin-starred restaurant, there is evidence that they master the concept of food-context. In a recent qualitative study, Waddingham, Shaw, Van Dam, and Bettiol (2018) showed that, among the reasons behind their food choice, primary school children reported examples of eating context (e.g., choosing ice-cream to cool down in a hot day or choosing a hot chocolate to warm up on a cold afternoon). The only other study on the effect of food-context appropriateness on the child’s acceptability and eating behavior of which we are aware is a dated article by Birch, Billman, and Salisbury Richards (1984). In this research, preschoolers were evaluated for changes in food acceptability within time of the day by providing, both in the morning and in the afternoon, familiar food appropriate for breakfast, for dinner, or for either mealtime. Despite their young age (3–4 years), most of the children showed significant preference shifts with time of the day, with breakfast items more preferred in the morning than the afternoon, while the reverse was observed for dinner items.

294

Context

The appropriateness of food-context is an important concept for children, as this association strongly depends on cultural background. Through repeated association of food and context, children learn what is appropriate and what is the most acceptable because it is familiar. Familiarity is known to be one of the most powerful determinants of children’s food acceptance and consumption. Based on the few data available in the literature, it seems that children, even the youngest ones, have already acquired information about mealtime appropriateness, and are reliable informants about factors that influence their food choice. However, data are scanty, and it would be interesting to deepen this topic in order to have a better insight on children’s perception of contextual factors’ appropriateness related to food consumption.

14.3

The physical environment

Previous research on adults has demonstrated that identical foods are perceived differently in different settings (King et al., 2004). For example, it is well-known that adult people expect food to be better and indeed rate them higher at home or at a restaurant versus a laboratory or an institutional cafeteria (Cardello, Bell, & Kramer, 1996; Meiselman, Johnson, Reeve, & Crouch, 2000). This consideration is valid also for children, for whom performing tests in laboratory settings is discouraged, as this context may be perceived as artificial by children, or they may even be intimidated by a formal and aseptic environment. If a laboratory setting is deemed necessary for the evaluation, then it is recommended that the test site is set to meet children’s needs. Tables may be lowered to suit the respondents, age appropriate pictures may be used to create a more relaxed and less sterile environment, and experimenters should avoid lab coats that can be perceived as overly authoritative (ASTM, 2013). For example, Kimmel, Sigman-Grant, and Guinard (1994) designed the testing room specifically for young children, providing round tables of different colors, and decorating the location with balloons and ribbons to create a relaxed atmosphere. In order to get more ecological results, it is advisable to conduct sensory and consumer testing with children in familiar environments, such as school, recreation centers, or home. However, in these settings, it is important to ensure that the evaluation is as rigorous as possible, for example, avoiding mutual influence among children, and performing the test in a quiet location, so the child can concentrate on the task (see also Section 14.4). In most European countries, children consume about half of their daily energy intake during school and afterschool activities (School lunch standards in Europe, 2012), thus school is an ideal place to influence children’s eating habits, and help them establish healthy dietary habits. Having lunch at school has an important educational function, because the meal implies a number of hidden significances, namely a physiological significance to learn to feed properly, a cultural significance to know different varieties and origins of foods, and a psychological significance to understand why a specific food product may evoke emotions (Pagliarini et al., 2005). However, previous research has shown that children are generally suspicious about food served at school, and report to prefer food eaten at home (Osowski, G€oranzon, & Fjellstr€om,

The effect of context on children’s eating behavior

295

2012). Despite negative attitudes toward institutional food, which may be classified as stereotypes (Cardello et al., 1996), it is important to understand how to improve food context in order to reduce food waste and increase consumption of healthy food. Hedonic responses to school menus have been collected in a number of studies. For example, Pagliarini et al. (2005, 2003) evaluated liking for individual components of 22 school menus supplied to Italian primary school children. Their data provided useful indications to school refectories about the most, as well as the least, favorite food formulations by children. Moreover, they found strong age-related differences in hedonic responses, with children becoming increasingly aware of their preferences and critical in their choices with growing age (i.e., 8–9 years). Similarly, Caporale et al. (2009) studied liking and consumption of school meals provided to Italian pre-schoolers (4–5 years), and found that food waste was highly predicted by hedonic ratings (Caporale et al., 2009). More recently, in view of the high rate of leftovers, Laureati, Cattaneo, Bergamaschi, Proserpio, and Pagliarini (2016) investigated how different fish recipes were liked by Italian primary school children. They found that children’s liking was strongly dependent on cooking methods, and that improving texture was an important aspect for increasing acceptance also among neophobic children. Probably the best and most comprehensive study performed to investigate the effect of contextual factors on children’s food acceptance is by Tuorila, Palmujoki, Kyt€ o, T€ ornwall, and Vehkalahti (2015). In this study, the authors evaluated how aspects related to the sensory properties of the meal served in the school canteen, as well as a series of variables more directly related to the food environment are perceived by Finnish students aged from 8 to 15 years. In this study, students rated their overall meal experience, as well as the appropriateness of the temperature, spiciness, and saltiness of a series of meals. Contextual factors that were evaluated were noise in the dining hall, progress of the line, timing of lunch break, and kindness of the dining staff. All these variables were taken together, along with individual factors such as age, gender, and food neophobia to build a predictive model of food acceptance. A general positive attitude toward the school menu, perceived hunger prior to lunch, and appropriate queue timing predicted acceptance positively; whereas food neophobia and being older were negative predictors. Sensory characteristics promoting the acceptance of the school meal were related to recognizable sensory qualities, such as bright and colorful; while the least liked dishes evoked fatty and cheese sensations. As reported by previous research (Laureati et al., 2016; Pagliarini et al., 2005), the authors found age-related differences in food context evaluation. When approaching the teenage years (from approximately 9 years to 12 years of age), students became critical about the quality of meals, attributing their complaints to missing qualities of main dishes, such as inappropriate serving temperature and low spiciness. At the same time, negative attitudes toward school foods appeared (Tuorila et al., 2015). Gender was found to influence food context evaluation as well. In the youngest students, both genders answered equally, whereas with increasing age, boys became more critical or more willing to state their opinion than girls. Age- and gender-related differences in contextual factor perception are important considerations, as during this period of life,

296

Context

children undergo major physical and psychological changes, so their expectations and their needs for food also change (Tuorila et al., 2015). In addition to the physical environment per se, imaginative context and gamification are also important when performing sensory and consumer testing with children. Presenting an engaging context to the children by making the experiment game-like, or developing a story-based method, can be of help to overcome the tedious nature of the task, and keep the children’s attention high. For example, Vennerød, Hersleth, Nicklaus, and Almli (2017) measured taste sensitivity in very young children (age 3–4 years) using a fairy tale in order to engage children, thus increasing their interest and participation rate. The analytical nature of the task (i.e., a paired comparison test requiring, for example, children to indicate the sweetest sample) was converted into a game-like task by asking the child to identify the cups of “magic water,” defined as tasting different from plain water, and represented by different magical characters (e.g., a fairy represented sweet taste, an elf sour taste, a mermaid salty taste). Game-like procedures and imaginary contexts are especially important when testing a consumer group with a short attention span, and who is not motivated by contribution to science and/or monetary rewards (Vennerød et al., 2017). Other examples of task and context gamification to trigger children’s curiosity and interest can be found in Brard and L^e (2016) and Tatlow-Golden, Hennessy, Dean, and Hollywood (2013). To conclude, the analysis of the literature concerning the effect of the physical context on children’s food liking and choice seems rather poor. Although there are numerous articles dedicated to the optimization of the sensory properties of school meals, there is only one study in which several variables related to the school food context are considered simultaneously. Future perspectives of study should be directed to better understand the child’s perception of the school food context as a whole, especially in view of the fact that the optimization of eating environments, as well as meals, may be useful to set up guidelines for institutional caterers, and may help to predict children’s perception of meals in a real eating context, thus maximizing consumption while decreasing waste.

14.4

The social environment

Food consumption has implications that go beyond merely providing nutrients and energy needed for survival, as eating and drinking are also deeply connected to social interaction (Cruwys, Bevelander, & Hermans, 2015). The great majority of meals are indeed consumed in the company of someone else, and this is often perceived as an enjoyable part of our cultural experience (Rozin, 2005). Therefore, it should not be surprising that a major determinant of human eating behavior is social modeling (see also Chapter 2 for further information on social influence). Social modeling emerges in the very first years of life, and remains stable across development (Cruwys et al., 2015). A clear example is given by young children who learn which foods are palatable by observing other people eating (Marty, Chambaron, Nicklaus, & Monnery-Patris, 2018).

The effect of context on children’s eating behavior

297

Social learning is indeed important for survival, as it plays an important role in guiding humans into what and how much to eat. Previous studies have shown that more food is consumed by individuals in a group than by individuals alone, the so-called social facilitation effect (Herman, 2015; Salvy, de la Haye, Bowker, & Hermans, 2012). In the same way, people eat less when they are in the company of individuals who eat less (Cruwys et al., 2015). Similar effects have been found also in young children who were observed to consume more food when eating in bigger, rather than smaller, groups of peers (Lumeng & Hillman, 2007). A different pattern has been observed in overweight children (6- to 10-year-olds), who ate much (31%) more when alone than when in a group of two overweight and two normal-weight children (Salvy, Coelho, Kieffer, & Epstein, 2007). This effect of “social suppression” of food selection (also observed in overweight adults) has been explained with overweight people’s concern about the negative inferences about them that might arise from overeating in public (Herman, 2015). In line with this assumption, Salvy, Howard, Read, and Mele (2009) found that overweight youths ate more when paired with other overweight youths than when paired with normal-weight youths. In many contexts, including the food context, modeling has been shown to be a very powerful process by which children of different ages acquire complex behaviors (Bandura, 1977). Children use food for many social purposes, for example, to construct their desired self-image, as a way to judge others, to emphasize friendships, and to live up to peer norms (Andersen et al., 2016). Models that have been shown to be effective in modulating children’s eating behavior include cartoon characters, peers, mothers, unfamiliar adults, and teachers (Laureati et al., 2014). In a study by Osowski et al. (2012) on children’s understanding of food context, children strived for social belonging to their peers in the canteens, and spoke of not liking to eat alone, probably because eating alone is seen as unusual and stigmatized. Children reported that at school it is important to eat with peers, but the presence of adults was also appreciated. In one of the first studies on the effect of social influence on children’s food preference, Duncker (1938) showed that target children exposed to a story hero with a strong preference for a bad-tasting food over a more pleasant one temporarily shifted their preference to the food preferred by the hero. Moreover, exposing target children to peers with different preferences produced a high percentage of choices of the peers’ preferred foods by the target children when asked about their favorite food in the presence of the peers (Duncker, 1938). Using a similar approach, Birch (1980) demonstrated that the same peer effect occurs even when the target children are asked to express their preference in the absence of their classmates, indicating that the child’s choice shift is not due to merely conforming to the majority’s choice, but rather to a change in preference that persists in the absence of direct peer influence. The fact that peers are important shapers of children’s eating behavior is not surprising, and some studies have even suggested that peers have more influence than adults on children’s food selection (Frazier, Gelman, Kaciroti, Russell, & Lumeng, 2012). Children do not wish to stand out from the crowd by making socially unacceptable food choices that may have social consequences, such as stigmatization and social exclusion. This is understandable because when growing up, children

298

Context

experience a period of uncertainty and, as a result, they try to seek acceptance by conforming to the norms of their peer group (Andersen et al., 2016). Moreover, peers and friends may be more influential during adolescence than during childhood, as social networks become increasingly important to motivations and behaviors, and seem to exert a stronger influence than parental norms (Salvy, Elmo, Nitecki, Kluczynski, & Roemmich, 2011; Salvy et al., 2012 for review). Recent evidence suggests that the effect of social modeling may be modulated by specific situations, and the food itself. For example, Andersen et al. (2016) showed that classmates influence children’s preference of a new type of school lunch, whereas preferences of familiar lunches (brought from home) were unaffected by peers. The authors speculated that this is probably because the lunch brought from home is an already negotiated system of preference; therefore, children do not need to look at their peers to decide if it is good or bad. Similar findings were reported by other studies (Greenhalgh et al., 2009; Hendy, 2002) in which it was found that children’s consumption of novel food increased after hearing positive statements from their peers. When an individual is faced with a novel food, social facilitation usually leads to a faster acceptance of the novel food (Visalberghi & Addessi, 2000), as it reduces the uncertainty of that specific eating situation, for example, ‘if a lot of people are doing this, it’s probably a wise thing to do’ (Cialdini, 2007). Therefore, social influences have been increasingly used to overcome food neophobia (i.e., the reluctance of trying novel food) in children. For example, children ate more of an unfamiliar food when an adult was eating it than when the food was merely offered (Addessi, Galloway, Visalberghi, & Birch, 2005). Similarly, enthusiastic teacher modeling was effective to encourage novel food acceptance in preschool children (Hendy & Raudenbush, 2000). Reward is another important principle related to social context that can be influential in modifying children’s eating behavior. A clear example is given by sweets, which are considered highly palatable, not only because they are closely linked with satisfaction of physiological needs, but also because they are frequently used by parents as rewards and treats, and are usually offered to children in positive socialaffective contexts such as parties and holiday celebrations (Birch, 1980). Reward and peer modeling have been explored quite extensively in past decades, in combination with other principles (e.g., repeated exposure), in order to develop strategies to change children’s eating behavior and guide them toward a healthy direction. For example, in a recent study by our group (Laureati et al., 2014), we investigated the effectiveness of a multicomponent school-based intervention based on peer modeling, reward, and repeated exposure to increase liking of fruits and vegetables and reduce food neophobia in a large cohort of children (age range 6–11 years). Children in the experimental group watched motivational videos consisting of episodes featuring the heroic “Food Dudes” who were a group of 12- to 13-year-old teenagers who were seen to eat and enjoy a variety of fruits and vegetables and to encourage all other children to do the same. The choice of using teenagers as models derived from previous evidence that children are more likely to imitate the behavior of individuals of the same age or slightly older than themselves or who they like or admire, especially if the behavior is rewarded (Lowe, Horne, Tapper, Bowdery, & Egerton, 2004). With this approach, we found a significant and long lasting effect of the intervention in reducing food neophobia and increasing liking of fruits, and partially, vegetables (Fig. 14.5A and B).

The effect of context on children’s eating behavior

299

25

Control

Food neophobia score

Experimental

20

15 Preintervention (t0)

Intervention (t1)

Follow-up (t2)

Phases

(A)

Fruit control Fruit experimental Vegetable control Vegetable experimental

7

6

Liking score

5

4

3

2

1

(B)

t0'

t0" Preintervention

t1

t2

Intervention

Follow-up

Fig. 14.5 Food neophobia (A) and liking scores for fruit and vegetables (B) registered in the study by Laureati et al. (2014) for experimental and control group of children at preintervention, intervention phase (t1) and follow-up (t2) ().

300

Context

14.4.1 Age- and gender-related differences in social learning Age- as well as gender-related differences are important to consider when studying how social mechanisms interact with children’s eating behavior. Research studies indicate that young children (preschoolers or younger) are more susceptible than the older ones to social learning (see, for example, Birch, 1980; Laureati et al., 2014), and this is why it is often suggested to start as early as possible with school-based educational programs in order to maximize health benefits. Moreover, a qualitative study by Osowski et al. (2012) reported gender stereotypes when interviewing and observing children eating in the school canteen. Girls and boys sometimes sat together in the canteen, but it became more common for them to sit separately with growing age. The reason that children reported for such behavior was that boys and girls like to talk about different things. Thus, it is essential to reflect on what effect gender and age interactions could have on children’s perception of food situations. Gender-related issues were also addressed in the study by Salvy et al. (2011), who found that adolescent females, but not males, consumed healthy foods more in the presence of their friends than in the company of their mothers. Conversely, female adolescents consumed less energy from unhealthy snacks in the presence of their friends than in the presence of their mothers. It is largely known that adolescence is a period during which gender differences in terms of body image and dieting concerns are likely to appear, and because dietary concerns are often associated with an increased desire to be popular and accepted, female adolescents may be more inclined to adjust their intake to convey an image of healthy eating in front of their peers and friends (Salvy et al., 2011).

14.4.2 Effect of social environment on children’s performance during sensory and consumer testing Due to the evident effect of social context on eating behavior, it is important to consider the people present during sensory and consumer testing with children. As previously mentioned, we recommend performing testing with children in familiar environments such as schools, recreational centers, or home, as it is crucial to ensure that the child feels comfortable in the test environment. Each of these contexts has potential influencers that have to be carefully taken into consideration. If the test is performed at home, or if they are present at the experiment, parents may occasionally interfere with the child’s responses. For example, in a study by Kimmel et al. (1994), some children looked to the parent for approval or attempted to show off. When the presence of the parents cannot be avoided (e.g., young children), having the parents at the perimeter of the room could be less distracting. The presence of the parents can be avoided if children are tested at school. However, in this case, peers and teachers can influence the effectiveness of the test. In order to obtain individual responses and avoid the bias coming with peers’ interaction, it can be of help to separate children who are friends, especially when the test is performed in a group setting (ASTM, 2013).

The effect of context on children’s eating behavior

301

Finally, the effect of the experimenters should not be underestimated. Especially when children have to be tested individually (e.g., children under reading age), it is preferable that they are tested in the presence of a familiar adult, as they may feel uncomfortable with unknown people. Our experience suggests that if the experimenters has to test the child, a warm-up meeting can be of help to introduce the child to the experimenter and establish child’s involvement. It may also happen that, for the need of social acceptance, children can give an answer only because they believe it is the one the experimenter is requiring. In this sense, children may overestimate the liking of food that is usually not very much appreciated (e.g., vegetables) because they want to be indulgent with the evaluator. This fits well with unpublished results from our laboratory demonstrating that, keeping the testing location constant, and the stimulus and the procedure, the administration of the task by the experimenter produced significantly higher hedonic ratings for a fish formulation than when the task was administered by their teacher (Laureati & Pagliarini, 2018). In these cases, doubleblind testing should be conducted to ensure that the children are not involuntarily guided in their answers by the experimenter (Vennerød et al., 2017).

14.5

Conclusions and future perspectives of study

Three main contextual factors have been identified as being highly influential in children’s food decision-making: the meal, the physical environment, and the social environment. These factors are important to consider when performing sensory and consumer testing with children. The few studies available on this topic have highlighted the importance of taking into consideration age- and gender-related differences when studying the contextual factors on eating behavior, because children tend to become more critical about context and sensory properties when they grow up, and perception of food and factors related to the context may be different in girls and boys. As a conclusive remark, reviewing the literature on the effect of contextual factors on children’s food preference and choice has highlighted a lack of information. Therefore, further research is needed on this topic. Studying the contextual factors related to food consumption among children could provide a deeper understanding of their thoughts and expectations about food and food settings. Moreover, it could help in finding solutions to increase the acceptance of target products (e.g., fruits and vegetables), thus contributing to improvement of children’s health, and to reducing food waste, which is a critical issue, especially in institutional contexts, such as school.

References Addessi, E., Galloway, A. T., Visalberghi, E., & Birch, L. L. (2005). Specific social influences on the acceptance of novel foods in 2–5-year-old children. Appetite, 45(3), 264–271. Aerts, G., & Smits, T. (2017). The package size effect: How package size affects young children’s consumption of snacks differing in sweetness. Food Quality and Preference, 60, 72–80.

302

Context

Andersen, S. S., Vassard, D., Havn, L. N., Damsgaard, C. T., Biltoft-Jensen, A., & Holm, L. (2016). Measuring the impact of classmates on children’s liking of school meals. Food Quality and Preference, 52, 87–95. ASTM 2299. (2013). Standard guide for sensory evaluation of products by children and minors. American society for testing and materials. West Conshohocken, US: ASTM International. Bandura, A. (1977). Social learning theory. New York: General Learning Press. Bell, R., & Meiselman, H. (1995). The role of eating environments in determining food choice. In D. W. Marshall (Ed.), Food choice and the consumer (pp. 292–312). Bergamaschi, V., Olsen, A., Laureati, M., Zangenberg, S., Pagliarini, E., & Bredie, W. L. P. (2016). Variety in snack servings as determinant for acceptance in school-children. Appetite, 96, 628–635. Birch, L. L. (1980). Effects of peer models’ food choices and eating behaviors on preschoolers’ food preferences. Child Development, 51(2), 489. Birch, L. L., Billman, J., & Salisbury Richards, S. (1984). Time of the day influences food acceptability. Appetite, 5, 109–116. Birch, L. L., Savage, J. S., & Fisher, J. O. (2015). Right sizing prevention. Food portion size effects on children’s eating and weight. Appetite, 88, 11–16. Brard, M., & L^e, S. (2016). The ideal pair method, an alternative to the ideal profile method based on pairwise comparisons: Application to a panel of children. Journal of Sensory Studies, 31, 306–313. Caporale, G., Policastro, S., Tuorila, H., & Monteleone, E. (2009). Hedonic ratings and consumption of school lunch among preschool children. Food Quality and Preference, 20, 482–489. Cardello, A. V., Bell, R., & Kramer, F. M. (1996). Attitudes of consumers toward military and other institutional foods. Food Quality and Preference, 7, 7–20. Cialdini, R. B. (2007). Descriptive social norms as underappreciated sources of social control. Psychometrika, 72, 263–268. Cruwys, T., Bevelander, K. E., & Hermans, R. C. J. (2015). Social modeling of eating: A review of when and why social influence affects food intake and choice. Appetite, 86, 3–18. DiSantis, K. I., Birch, L. L., Davey, A., Serrano, E. L., Zhang, J., Bruton, Y., et al. (2013). Plate size and children’s appetite. Effects of larger dishware on self-served portions and intake. Pediatrics, 131, e1451–e1458. Duncker, K. (1938). Experimental modification of children’s food preferences through social suggestion. Journal of Abnormal Social Psychology, 33, 489–507. Frazier, B. N., Gelman, S. A., Kaciroti, N., Russell, J. W., & Lumeng, J. C. (2012). I’ll have what she’s having: The impact of model characteristics on children’s food choices. Developmental Science, 15(1), 87–98. Greenhalgh, J., Dowey, A. J., Horne, P. J., Fergus Lowe, C., Griffiths, J. H., & Whitaker, C. J. (2009). Positive- and negative peer modelling effects on young children’s consumption of novel blue foods. Appetite, 52, 646–653. Guinard, J. X. (2001). Sensory and consumer testing with children. Trends in Food Science and Technology, 11, 273–283. Hendy, H. M. (2002). Effectiveness of trained peer models to encourage food acceptance in preschool children. Appetite, 39, 217–225. Hendy, H. M., & Raudenbush, B. (2000). Effectiveness of teacher modeling to encourage food acceptance in preschool children. Appetite, 34(1), 61–76. Herman, C. P. (2015). The social facilitation of eating. A review. Appetite, 86, 61–73. Kimmel, S., Sigman-Grant, M. J., & Guinard, J.-X. (1994). Sensory testing with young children. Food Technology, 48, 92–99.

The effect of context on children’s eating behavior

303

King, S. C., Weber, A. J., Meiselman, H. L., & Lv, N. (2004). The effect of meal situation, social interaction, physical environment and choice on food acceptability. Food Quality and Preference, 15(7–8), 645–653. K€ oster, E. P. (2003). The psychology of food choice: Some often encountered fallacies. Food Quality and Preference, 14, 359–373. Kral, T. V., Kabay, A. C., Roe, L. S., & Rolls, B. J. (2010). Effects of doubling the portion size of fruit and vegetable side dishes on children’s intake at a meal. Obesity (Silver Spring), 18 (3), 521–527. Laureati, M., Bergamaschi, V., & Pagliarini, E. (2014). School-based intervention with children: Peer-modeling, reward and repeated exposure reduce food neophobia and increase liking of fruits and vegetables. Appetite, 83, 26–32. Laureati, M., Cattaneo, C., Bergamaschi, V., Proserpio, C., & Pagliarini, E. (2016). School children preferences for fish formulations: The impact of child and parental food neophobia. Journal of Sensory Studies, 31(5), 408–415. Laureati, M., & Pagliarini, E. (2018). New developments in sensory and consumer research with children. In: G. Ares & P. Varela (Eds.), vol. 2. Methods in consumer research: Alternative approaches and special applications (pp. 322–355): Woodhead Publishing. Laureati, M., Pagliarini, E., Gallina Toschi, T., & Monteleone, E. (2015). Research challenges and methods to study food preferences in school-aged children: A review of the last 15 years. Food Quality & Preference, 46, 92–102. Lowe, C. F., Horne, P. J., Tapper, K., Bowdery, M., & Egerton, C. (2004). Effects of a peer modelling and rewards-based intervention to increase fruit and vegetable consumption in children. European Journal of Clinical Nutrition, 58, 510–522. Lumeng, J. C., & Hillman, K. H. (2007). Eating in larger groups increases food consumption. Archives of Disease in Childhood, 92, 384–387. Marty, L., Chambaron, S., Nicklaus, S., & Monnery-Patris, S. (2018). Learned pleasure from eating: An opportunity to promote healthy eating in children? Appetite, 120, 265–274. Mathias, K. C., Rolls, B. J., Birch, L. L., Kral, T. V., Hanna, E. L., Davey, A., et al. (2012). Serving larger portions of fruits and vegetables together at dinner promotes intake of both foods among young children. Journal of the Academy of Nutrition and Dietetics, 112, 266–270. Meiselman, H. L. (Ed.), (2009). Meals in science and practice. Interdisciplinary research and business applications. Cambridge: Woodhead. Meiselman, H. L., Johnson, J. L., Reeve, W., & Crouch, J. E. (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35, 231–237. Mennella, J. A. (2014). Ontogeny of taste preferences: Basic biology and implications for health. American Journal of Clinical Nutrition, 99, 704S–711S. Osowski, C. P., G€oranzon, H., & Fjellstr€om, C. (2012). Children’s understanding of food and meals in the foodscape at school. International Journal of Consumer Studies, 36 (1), 54–60. Pagliarini, E., Gabbiadini, N., & Ratti, S. (2005). Consumer testing with children on food combinations for school lunch. Food Quality and Preference, 16, 131–138. Pagliarini, E., Ratti, S., Balzaretti, C., & Dragoni, I. (2003). Evaluation of a hedonic scaling method for measuring the acceptability of school lunches by children. Italian Journal of Food Science, 15(2), 215–224. Piqueras-Fiszman, B., & Jaeger, S. (2015). The effect of product–context appropriateness on emotion associations in evoked eating occasions. Food Quality and Preference, 40, 49–60.

304

Context

Roe, L. S., Meengs, J. S., Birch, L. L., & Rolls, B. J. (2013). Serving a variety of vegetables and fruit as a snack increased intake in preschool children. American Journal of Clinical Nutrition, 98, 693–699. Rolls, B. J. (2014). What is the role of portion control in weight management? International Journal of Obesity, 38, S1–S8. Rolls, B. J., Rowe, E. A., & Rolls, E. T. (1982). How sensory properties of foods affect human feeding behavior. Physiology & Behavior, 29(3), 409–417. Rolls, B. J., Rowe, E. A., Rolls, E. T., Kingston, B., Megson, A., & Gunary, R. (1981). Variety in a meal enhances food intake in man. Physiology & Behavior, 26(2), 215–221. Rozin, P. (2005). The meaning of food in our lives. A cross-cultural perspective on eating and well-being. Journal of Nutrition Education and Behavior, 37, S107–S112. Salvy, S.-J., Coelho, J. S., Kieffer, E., & Epstein, L. H. (2007). Effects of social contexts on overweight and normal-weight children’s food intake. Physiology & Behavior, 92, 840–846. Salvy, S. J., de la Haye, K., Bowker, J. C., & Hermans, R. C. J. (2012). Influence of peers and friends on children’s and adolescents’ eating and activity behaviors. Physiology & Behavior, 106, 369–378. Salvy, S. J., Elmo, A., Nitecki, L. A., Kluczynski, M. A., & Roemmich, J. M. (2011). Influence of parents and friends on children’s and adolescents’ food intake and food selection. American Journal of Clinical Nutrition, 93, 87–92. Salvy, S.-J., Howard, M., Read, M., & Mele, E. (2009). The presence of friends increases food intake in youth. American Journal of Clinical Nutrition, 90, 282–287. School lunch standards in Europe (2012). Food Today, No. 83. European Food Information Council.http://www.eufic.org/page/en/page/FTARCHIVE. Szczesniak, A. S. (2002). Texture is a sensory property. Food Quality and Preference, 13, 215–225. Tatlow-Golden, M., Hennessy, E., Dean, M., & Hollywood, L. (2013). ‘Big, strong and healthy’. Young children’s identification of food and drink that contribute to healthy growth. Appetite, 71, 163–170. Tuorila, H., Palmujoki, I., Kyt€ o, E., T€ornwall, O., & Vehkalahti, K. (2015). School meal acceptance depends on the dish, student, and context. Food Quality and Preference, 46, 126–136. Van Ittersum, K., & Wansink, B. (2012). Plate size and color suggestibility: The Delboeuf Illusion’s bias on serving and eating behavior. Journal of Consumer Research, 39(2), 215–228. van Kleef, E., Vrijhof, M., Polet, I. A., Vingerhoeds, M. H., & de Wijk, R. A. (2014). Nudging children towards whole wheat bread: A field experiment on the influence of fun bread roll shape on breakfast consumption. BMC Public Health, 14, 906. Vennerød, F. F. F., Hersleth, M., Nicklaus, S., & Almli, V. L. (2017). The magic water test. An affective paired comparison approach to evaluate taste sensitivity in pre-schoolers. Food Quality and Preference, 58, 61–70. Visalberghi, E., & Addessi, E. (2000). Seeing group members eating a familiar food enhances the acceptance of novel foods in capuchin monkeys. Animal Behaviour, 60(1), 69–76. Waddingham, S., Shaw, K., Van Dam, P., & Bettiol, S. (2018). What motivates their food choice? Children are key informants. Appetite, 120, 514–522. Wadhera, D., & Capaldi-Phillips, E. D. (2014). A review of visual cues associated with food on food acceptance and consumption. Eating Behaviors, 15, 132–143.

The effect of context on children’s eating behavior

305

Wansink, B. (2004). Environmental factors that increase the food intake and consumption volume of unknowing consumers. Annual Review of Nutrition, 24, 455–479. Wansink, B., & Kim, J. (2005). Bad popcorn in big buckets. Portion size can influence intake as much as taste. Journal of Nutrition Education and Behavior, 37, 242–245. WHO (2014). Limiting portion sizes to reduce the risk of childhood overweight and obesity. E-library of evidence for nutrition actions (eLENA). Retrieved from http://www.who. int/elena/bbc/portion_childhood_obesity/en/. Accessed 7 December 2017.

This page intentionally left blank

Food combinations and food and beverage combinations in meals

15

Jacob Lahne Department of Food Science & Technology, Virginia Tech, Blacksburg, VA, United States

15.1

Introduction

The concept of “food pairing” is widespread throughout popular culture. The idea that combinations of foods and/or beverages go “better” together than separately is a common theme in published recipes and cookbooks. For example, the New York Times regularly has their wine critic suggest pairings for published recipes (see for example Tanis, 2018), and texts that attempt to advise the public on pairings regularly win prestigious culinary awards (Dornenburg & Page, 2006). According to one of the earliest published scientific studies on food pairings, “a person may only slightly like each of two foods by themselves, but he may like their combination very much,” and thus “preference for the combination [of foods] might not be predictable from any sum or weighted average of the preferences [for the individual foods]” (Eindhoven & Peryam, 1959, p. 379). This definition has changed strikingly little over the past 60 years. For example, Harrington posits that “a pairing selection can transform sensory components… this transformation can have a positive impact, a neutral impact, or, in some cases, a negative impact” (2005, p. 104). Food pairings are very basic “context effects” (Chapter 1 this volume) in that a consumer’s perception of one food in a pair changes dramatically because of the presence of the second in the eating context. Recently, Traynor, Burke, O’Sullivan, Hannon, and Barry-Ryan (2013) wrote that “flavour [sic] pairing (or food pairing) is the coupling of flavours to produce an experience that is more appreciated than either of the two flavours alone” (p. 570). In fact, this basic definition—which we might as well gloss as a food pairing being a nonadditive increase (different than would be predicted from adding together liking for the two individual foods) in affective liking for the components when eaten together—is so well-established that very recent papers on the subject no longer bother to define what a food pairing is, and dive right into hypothesized mechanisms for food pairing (e.g., Eschevins, Giboreau, Allard, & Dacremont, 2017; Galmarini, Loiseau, Debreyer, Visalli, & Schlich, 2017; Galmarini, Loiseau, Visalli, & Schlich, 2016; Paulsen, Rognsa˚, & Hersleth, 2015). The essential idea of food pairing appears to be one not merely of peaceful coexistence, but of synergy: foods that pair well together create a super-additive combination (more liked than expected) when eaten together. Of course, here “foods” is a broad term that has to include beverages, as a great deal of the research into pairings Context. https://doi.org/10.1016/B978-0-12-814495-4.00015-5 Copyright © 2019 Elsevier Inc. All rights reserved.

308

Context

focuses on the pairing of foods and beverages. Furthermore, there’s no good reason that two foods should act differently than a food and a beverage when paired. Within the context of a meal, the potential for food pairings to occur is huge. Across cultures, meals as a unit consist of more than one dish (Carroll, 2013; Visser, 1991), and usually involve primary dishes, condiments, and beverages (Phan, this volume; Rozin, 2000). Furthermore, the dishes and condiments (and beverages) are rarely “pure” substances—and, in reality, the idea of purity is probably not useful (see also Shotwell, 2016; Tsing, 2015), as in the case of foods only a very few pure chemical substances, such as salt, are actually consumed—so, within a meal context, it is probably also important to consider the pairings that are occurring within individual dishes, at the level of ingredients, and even sometimes between chemical compounds. Thus, a single eating occasion can present a multitude of occasions for pairings to occur— or not. How is one to measure the goodness of a pairing? A very simple measure would be to check for a non-linear additivity in liking between two food items or ingredients, but, perhaps because food liking is so variable and culture-bound, other definitions have been proposed. In a previous review of some of the scales that have been developed to measure pairing (Lahne, 2018), it was found that these measures are largely based on principles from either common sense (e.g., King & Cliff, 2005), or from received expert wisdom (e.g., Donadini, Spigno, Fumi, & Pastori, 2008; Harrington, 2005; Nygren, Gustafsson, & Johansson, 2002), and have attempted to operationalize an intuitive understanding that the elements of a pairing should be “balanced,” or that neither of the two elements should “dominate” the other. In general, the concept of balance has not done a good job of predicting the non-additive liking expected from food pairings (Lahne, 2018), and indeed is not well-supported in empirical considerations of good pairings. For example, monosodium glutamate is widely accepted to be an ingredient that enhances flavors—for example, creates a non-additive increase in liking—but also disappears into the dish: it is not a “balanced” pairing. Newer attempted measures have been “harmony” and “complexity,” but these have not been verified psychometrically as usable, reliable, and valid concepts, especially as applied to foods (Devellis, 2016). In everyday language they are so semantically related to positive affect that it is difficult to imagine them being disengaged from liking by an untrained subject. Indeed, how would you train the subject? We do not have a good lexical or physical reference for a “good pairing” that can be presented as we would a reference in DA training. For example, while oysters and Muscadet are widely considered a paradigmatic example of good pairing, it is easy to imagine that a subject inexperienced with these foods might not experience this, because the novelty of, for example, oysters would overwhelm the ability of the subject to notice any pairing. Therefore, this chapter is comprised of a critical review of the existing literature on food pairings in the sensory- and consumer-science literature, and some thoughts on designing more productive research into this fascinating phenomenon. First, I will put forward some basic challenges in researching food pairings. Then, I will review the definitions (or lack thereof ) necessary to formulate scientific hypotheses about food pairings. I will then discuss some of the operationalized theories of how food pairing

Food combinations and food and beverage combinations in meals

309

“works,” with concepts such as “dominance versus harmony” and commonsense principles of wine and food pairing. As a contrast, I will examine some specific cases in which a chemical basis is hypothesized and tested for food pairing, which will lead naturally into a discussion of the currently popular, but counter-indicated Flavor Pairing Theory. Finally, I will point to some trends in current research within food science that show promise, and briefly discuss some research from outside sensory and consumer science that points to a broader, cultural definition of food pairing. I will then try to synthesize these findings into recommendations for better research into food pairing, as well as to nicely package some reminders about the lack of universality that must—by definition—accompany this type of research.

15.2

Challenges in food pairing research

There are several fundamental challenges in food pairing research. An obvious challenge is what might be termed the combinatorial problem: with the very large number of potential foods, it is impossible merely to test a subset of foods and declare principles of food pairing. Even with a large sample of foods, it is impossible to conclude that one has covered the range of possible combinations. Specifically, experimental research, in particular, is hampered by this problem, as large sets of foods to pair means that the researcher is tasked with examining a growing set of pairs that quickly become unmanageable. Conversely, if the researcher chooses a convenient set of foods to examine, he or she is often limited, or at least guided, by attempting to follow or to violate cultural norms (e.g., Wang, 2017), which means that the results of these experimental designs are inevitably confounded by the researcher’s cultural context. Another key challenge is definitional, as discussed in the introduction: What constitutes a “food” in a food pairing test? This goes beyond the problem of food versus beverage: Is a food a base ingredient, such as flour, butter, or salt (e.g., Ahn, Ahnert, Bagrow, & Baraba´si, 2011)? Must food pairing operate on complete or finished dishes to “count” for the meal context? Is it possible to discuss pairings on the molecular level (e.g., The Food Pairing Company, 2017)? All of these have been considered as base units for investigating pairing, but this confusion points to a deeper question: Does food pairing research need to investigate what makes ingredients fit together into a recipe? Is the study of food pairing in meals an extension of culinary research? In this case, structured literature reviews of culinary handbooks and cookbooks will be necessary to establish an empirical guide to what ingredients go together (for a fascinating, if probably not useful guide to historical thinking in this arena, see Rietz, 1961), and to relate this pairing structure to cuisines and food cultures; and to some degree, so-called Big Data approaches may provide the opportunity for new insights in this arena (e.g., Ahn et al., 2011; Jain, Rakhi, & Bagler, 2015). However, these researchers would do well to consider the extensive work conducted by food-studies scholars such as Elizabeth Rozin (2000), who have examined and theorized cuisine primaries to some degree. An advantage that scholars in food studies and culinary studies have in this arena is an extensive reliance on qualitative and emic approaches to understanding food

310

Context

behavior in general, and pairing in particular. In fact, very few food-pairing studies in sensory and consumer science have asked the subjects what makes food pair well. An over-reliance on regulatory ideals of experimental and behavioral psychology—as described by K€ oster (2003, 2009)—has generally led sensory scientists to distrust their subjects and prefer to control, rather than to inquire; this tendency is counterproductive in the case of food-pairing (and in much research into the context of the meal), as the preceding discussion demonstrates the general fog surrounding the definitions of food pairing in the first place. There is a dearth of qualitative studies that attempt to define what makes foods pair well, or what the experience of well-paired foods is. This is a possible area of research. Similarly, I am not aware of studies that have used a mixed-methods approach, combining either qualitative or consumer research to identify a good pairing, and then using a more exacting sensory method such as DA to analyze the elements of that pair and their combination to identify what, exactly, is happening on a sensory level. Instead, good pairs are usually sourced from existing literature, from experimenter ideation, or from expert consultation, and then evaluated without further consideration by the researcher on whether these putative pairs are good representatives of the phenomenon, and can be the basis of generalization. Another challenge facing studies of food pairings in meals is the culture-bound nature of cuisine. It is quite unrealistic to believe that there is some universal set of “best” tastes, and in fact, even the commonly received wisdom about universal basic taste preferences—salty and sweet being affectively positive, sour and bitter being negative—are contradicted by studies that employ culturally diverse populations (Moskowitz, Kumaraiah, Sharma, Jacobs, & Sharma, 1975). Food science, of course, is dominated by Western scholars who, no matter how well intentioned, bring their own assumptions about how food works to their experimental design. This tendency is exacerbated when scientists rely on experts or reference sources they find convincing—another acculturated decision—for sourcing food pairs to test. There is a large body of research that shows that food preferences are anything but universal (e.g., Cardello, 1995; Meiselman, 2000; Meiselman, 2017; Mela, 2001; Moskowitz et al., 1975). That food pairings are often studied in “model” systems such as cheese or chocolate, and wine or beer, both demonstrate this cultural specificity (it is quite unlikely that these would be the first pairs that Asian or African subjects would suggest), and is a real barrier to any possible generalization about results. Is it really likely that the rules governing cheese and wine pairing derived from these studies can be generalized to understanding the quite distinct pairings common to Japanese (Otsuka, 2000) or Chinese (Newman, 2000) meals, much less the quite-understudied and diverse meal patterns characteristic of African cuisines (Rozin, 2000)?

15.3

What is a food pairing?

According to our summary definition, food pairings occur when a pair of foods (or ingredients) combine to produce a more positive affective response than would be predicted from the individual components (Eindhoven & Peryam, 1959;

Food combinations and food and beverage combinations in meals

311

Harrington, 2005; King & Cliff, 2005; Traynor et al., 2013). In this simple definition, a food pairing occurs when there is a non-additive increase in liking when two foods are consumed together. This definition is operationalizable through affective sensory testing (Lawless & Heymann, 2010) and appealing in its simplicity. However, it is clearly incomplete insofar as it locates the synergy solely in the two foods being combined: as detailed by Spence and Piqueras-Fiszman (2014), and of course explained in the Introduction to this volume, the meal context itself is at least as important for the sensory impression of the foods. Elements of the consumption experience as diverse as the subjects’ (the diners’) cultural and social backgrounds, the meal context, and the inclusion of the “unexpected” or the “surprise” (see also Mielby & Frøst, 2010) can all influence the overall affective valence of single food items, to say nothing of combinations. For a very simple example, Spence and Piqueras-Fiszman (2014, p. 190) document the non-additive synergy of monosodium glutamate and benzaldehyde for Japanese subjects, but not for American subjects, with a convincing cultural explanation: umeboshi, or salted plums, are a common savory ingredient in Japan, and combine both almond (benzaldehyde) and savory (monosodium glutamate) percepts, whereas the combination is very rare or unheard of in Western cuisine. Therefore, the synergy definition describes what might differ in the diner’s experience upon encountering a food pairing, but does little to help point to a definite cause in the foods being paired. Attributing the synergy merely to the molecular makeup or the material co-occurrence of two foods is inadequate. An empirical approach could determine which foods are preferentially consumed together, and then try to derive what is common between the observed common food pairs. For example, Eindhoven and Peryam (1959) surveyed expected (not actual) liking for menu items, and combinations of menu items in a sample of US army trainees, and looked for pairs in which expected liking was superadditive; they found super-additivity among items that the dominant American culture usually serves together, such as baked ham and candied sweet potatoes. In a similar population of US armed-service personnel, Moskowitz and Klarman (1977) asked respondents to rate expected compatibility for theoretical meals from mains (meat), sides (mostly potatoes), and vegetables, with similar results. These studies demonstrate that cultural norms determine a baseline for pairing success: the dominant American main meal tends to consist of a “meat and two veg” combination (Carroll, 2013), reflecting the Northern US preferences for meals with a single, meat-based protein source and bland, starchy sides (Dupuis, 2015). This norm is reflected in the expectation of pairing (liking super-additivity) for meals that follow the established cultural patterns. However, it does little to explain the sensory mechanism of these pairings, which, if norms indeed are the driving force, is probably based in (cultural) memory and associative effects (see for example Sutton, 2001). Similar, but different patterns exist in other cultures (see Rozin, 2000), and survey research would presumably find support for the super-additivity of these common associations. But what of the unexpected? If pairing is solely a cultural phenomenon, we would expect unexpected combinations to be universally negative, which is of course

312

Context

demonstrably not the case. The unexpected—at least to some degree—is often the basis of culinary success and acclaim, although of course individual responses are moderated by personal attributes such as neophobia. For example, in collaboration with professional chefs, Mielby and Frøst (2010) created 11 novel dishes served to diners in a restaurant setting and evaluated how likeable, challenging, and surprising they were. Overall, while diners preferred those dishes that were surprising takes on classical combinations (confirming the studies cited herein), different verbal presentations were able to moderate this effect, and create links between surprise and liking. Spence and Piqueras-Fiszman (2014) provide a wealth of examples from their own research on the importance of novel combinations to the acceptance of restaurant dishes. In experimenting with some of the predictions of Flavor Pairing Theory (see the following), Perkel (2012) reports that, while some predicted pairings were unpleasant, other novel combinations were indeed very well liked. Sometimes, caviar with dessert is exactly the right combination at the right time (de Klepper, 2011)! Of course, humans are not cultural dupes; taste preferences are not solely predetermined by society (Hennion, 2007), no matter how influential norms are (Bourdieu, 1984). Some culinary experimentations succeed, are popularized, and so gastronomic cultures develop and change, rather than remaining static (Rozin, 2000). Somehow, successful food pairings take advantage of both surprise and tradition to create new, but not uncomfortable sensations (Mielby & Frøst, 2010). These reports focus largely on the pairing of ingredients within a dish, or of multiple cooked components of a meal eaten together. But most research into the phenomenon of super-additive liking resulting from combinations focus on pairings of foods and beverages, usually, but not always, alcoholic. This is, again, in large part because of cultural norms: in Western culinary culture, which is still dominated by the influence of classical French cuisine, wines are the accompaniment par excellence for foods (Harrington, 2007), and vice versa. While Bastian et al. (this volume) address the pairing of food and beverages specifically, there is no real reason that food pairing generally and food and wine pairing specifically are distinct as sensory phenomena (Lahne, 2018). Most beverage pairings are similarly based in culturally determined traditions (Harrington, 2005; Harrington, 2007; King & Cliff, 2005; Paulsen et al., 2015), and on the whole, researchers have found, using various metrics, that consumers often, but do not always, agree with expert recommendations (Bastian, Payne, Perrenoud, Joscelyne, & Johnson, 2009; Donadini, Fumi, & Lambri, 2013; Harrington & Seo, 2015). Thus, research into pairings between foods and beverages and research into pairings between two foods points to the same fundamental question: if cultural, culinary norms contribute so heavily to the perception of a good pairing for consumers, how can fundamental research be conducted that speaks to some universal principle of pairing? Is it possible to divine a set of principles that will predict pairings outside of the general rules of cuisine that govern different cooking cultures?

Food combinations and food and beverage combinations in meals

15.4

313

Balance, harmony, complexity… pairing?

Recognizing this difficulty, several researchers have attempted to codify what makes two foods go well together. One popular theory is that the intensity of flavors, regardless of their subjective quality, should be approximately equal in two foods that will pair well together (Harrington, 2007, King & Cliff, 2005, Paulsen et al., 2015): this is usually called “balance.” In the literature, this is usually conceived of as an inverted U-shaped “dominance” relationship, in which the best pairing occurs when neither of the two foods “dominates” the other (Bastian et al., 2009; Bastian, Collins, & Johnson, 2010; Donadini et al., 2008; Donadini et al., 2013; Donadini & Fumi, 2014; Donadini, Fumi, & Lambri, 2012; Donadini, Fumi, & Newby-Clark, 2015; Harrington, 2005; Harrington, 2007; Harrington & Hammond, 2005; Harrington, Mccarthy, & Gozzi, 2010; Harrington & Seo, 2015; King & Cliff, 2005). What exactly “dominance” means and why it is problematic for pairing is unclear: once again, the example of MSG is perhaps informative, as MSG tends to enhance the intensity, and often hedonic acceptance, of foods while being entirely “dominated” by the flavors it is enhancing (Lawless & Heymann, 2010). Consider the case of palate cleansers used in sensory evaluation—these are required to not provoke negative interactions: Lucak and Delwiche (2009) found that water and unsalted crackers, the most intensity- and flavor-neutral stimuli they considered, had the broadest universal applicability as palate cleansers. The intuitive appeal of balanced intensity to predict pairing seems to be potentially misleading. Eschevins et al. (2017) claim that dominance falls within a “blending” paradigm, in which the two foods’ flavors combine more or less seamlessly into a new percept. This approach may be a better fit for examples such as that of MSG, as discussed herein. They relabel this blending/dominance theory as one of “harmony” between the two foods (for an earlier and somewhat different use of “harmony,” see Cerretani, Biasini, Bonoli-Carbognin, & Bendini, 2007). The purpose of this relabeling is to introduce a second, orthogonal construct labeled as “complexity,” which they claim leads to the U-shaped relationship found in pairings: levels of complexity that are too high or too low lead to decreased liking of stimuli. This is based on an analogy to abstract visual stimuli, however (see also Paulsen et al., 2015). Thus, an alternative, two-dimensional paradigm of “harmony” and “complexity” is suggested to predict the goodness of pairing. However, once again, these are poorly defined psychometric constructs: it is not at all clear that harmony is cognitively separable from overall liking, nor is the U-shaped relationship predicted between liking and complexity that the theory claims evident from the available data (see Fig. 2, Eschevins et al., 2017, p. 5). A third paradigm for predicting food pairings is that of aromatic similarity. While Eschevins et al. (2017) argue that aromatic similarity is more or less synonymous with their “harmony/complexity” paradigm, this is far from clear in the literature. Aromatic/flavor similarity is usually cited as important for pairings by culinary experts such as sommeliers and chefs (Bastian et al., 2009; Dornenburg & Page, 2006;

314

Context

Eschevins et al., 2017; Harrington, 2005; Harrington, 2007); it is certainly in line with ideas of harmony, but whether it speaks to either balance or complexity is less clear. Aromatic similarity also points toward a flavor-chemistry explanation for food pairing: aromatic similarity must derive from the aromas of foods, which we generally understand to be derived from the volatile chemicals found in the food (Lawless & Heymann, 2010).

15.5

When chemistry dictates pairing

A particular case of negative food pairing (subadditive liking) connected to aromatic incompatibility has been closely examined for a chemical basis: the poor pairing between red (and sometimes white) wines and seafood-based foods. In general, it is now fairly widely accepted that the interaction of polyunsaturated fatty acids found mainly in seafood can be rapidly oxidized in-mouth into particularly unpleasant volatile compounds associated with “fishy,” “beany,” and “painty” aromas. Culprit compounds identified in wines are ferrous iron, which is found in relative abundance in red wines, as compared with white wines and sakes (Tamura et al., 2009); and sulfur dioxide, which is used in wine production, but not sake production (Fujita et al., 2010). Thus, the theory goes, poor wine-seafood pairings occur because of inappropriate chemical interactions. However, it is important to draw attention to the cultural context that is implicit in this work—not to invalidate the conclusions, but to once again highlight that food pairings cannot be detached from their originating cultural context. Both papers are by Japanese research groups, and presumably the researchers bring with them their expectations for pairs that should and should not work well together. Thus, if Tamura et al. (2009) are correct, pinot noir—a red wine with relatively abundant ferrous iron—should not be an apparent good pair with salmon, a fatty fish rich in labile, polyunsaturated fatty acids (Harrington, 2007). Similarly, if chemistry solely governs pairings, the data from Fujita et al. (2010) should indicate a difference in pairing quality with fish between so-called “natural” white wines and those made with additions of sulfur dioxide, which, to my knowledge, has not been documented. At the same time, it is clear that in many cases red wine, which contains more ferrous iron, does not pair well with fish, and both papers point toward rapid in-mouth lipid oxidation as a source of the poor pairing. How can one separate the contributions of cultural expectation and of chemical interaction to explain food pairing?

15.6

Flavor pairing theory

If the chemical compositions of foods are the ultimate basis for their flavors, then perhaps not only the key to bad pairings such as fish and red wine, but to good pairings, such as salmon and pinot noir, lies in the volatile aroma compounds that characterize different foods. A recent, popular theory called the Flavor Pairing Theory was first advanced by chef Heston Blumenthal, inspired by commentary by flavorist Franc¸ois

Food combinations and food and beverage combinations in meals

315

Benzi (de Klepper, 2011; Kort, Nijssen, van Ingen-Visscher, & Donders, 2010), and posits that good pairing between foods (defined here as ingredients, incidentally) can be predicted by the sharing of volatile aroma compounds (Møller, 2013; The Food Pairing Company, 2017). Thus, if two foods share more volatile aroma compounds, they should make a better pair. The Flavor Pairing theory provides a testable hypothesis for food pairing: take a set of foods, combine them pairwise, and then see whether the number of shared aroma compounds is correlated with consumer liking of the pairs. Unfortunately, the few sensory studies that have tested the Flavor Pairing theory have failed to support it. In a small study, when pairs of pureed foods were combined, those pairs that shared more flavor compounds than an average pair of foods were no more likely to produce increased liking than those which were at or below average (Kort et al., 2010). More recently, foods chosen specifically for their shared aroma compounds without a basis in the culinary repertoire of Western culture—banana combined with bacon, basmati rice, or extra-virgin olive oil—did not reliably deliver increased liking when paired (Traynor et al., 2013). Even pairings directly predicted by the Food Pairing Company—the commercial entity that has found consultancy successes based on their promulgation of the Flavor Pairing Theory—often fail to hit the mark (Perkel, 2012). As a demonstration of network-based statistical analyses, several research scholars have attempted to test whether published recipes follow the predictions of Flavor Pairing Theory. In the first study of this type, Ahn et al. (2011) found that while European and North American recipes did seem to include ingredients that shared more volatile aroma compounds than would be expected on average, East Asian (Korean, Chinese, Southeast Asian) and Central and South American recipes, in fact, tended to call for ingredients that shared fewer than expected volatile compounds. A more recent study of South Asian culinary traditions using a similar model demonstrated that, with varying degrees, cuisines on the Indian subcontinent also called for fewer than expected shared aroma compounds ( Jain et al., 2015). Because food preferences—and therefore preferences for pairs of foods—are almost entirely cultural—aside from inborn aversions to bitterness and preference for sweetness (Herz, 2005; Rozin & Vollmecke, 1986)—this is not a surprising result: for the Flavor Pairing Theory to hold, there would have to exist invariant “best pairings” (de Klepper, 2011) for each food. It would also, in many ways, imply the death of culinary innovation, as a “best pairing” means that other, suboptimal pairings, should be avoided. Finally, it would imply that in general, dishes should be able to be enhanced by adding further ingredients that share volatile aroma compounds, whereas in general, adding further ingredients to a dish is no guarantee of improved flavor. Incidentally, the failure of Flavor Pairing Theory to predict food pairing also implies that complexity probably is not a dependable factor in universally predicting (negative) food pairing (e.g., Eschevins et al., 2017; Paulsen et al., 2015). At least two billion people seem to prefer increased aromatic complexity in their recipes: as described in two of the papers attempting computational evaluations of Flavor Pairing Theory, recipes from South Asia ( Jain et al., 2015) and East Asia

316

Context

(Ahn et al., 2011) are more likely to include ingredients that do not share volatile compounds than would be predicted by chance. As Paul Rozin noted pithily more than 30 years ago, warning against basing conclusions too heavily on a literature concentrated on North American and West European subjects, these phenomena are “not restricted to India, but even if [they] were, there are more people alive in India today than in all of North and South America” (Rozin & Vollmecke, 1986, p. 435).

15.7

New and old ideas: Alternative conceptions of food pairing

Recently, food-pairing research within the discipline of sensory evaluation has started to take an encouraging turn toward a somewhat more emic approach to understanding food pairings. By this I mean allowing the subjects of the research—who are, after all, experiencing the pairings—some degrees of freedom to explain their flavor experiences to the researchers. In general, these approaches depart from affective studies that, by necessity, a priori operationalize theories such as balance, harmony, or complexity, and require subjects to conform to these theories, and instead source reports of the interactions of two or more foodstuffs from subjects themselves. One approach that seems to offer the potential of increasingly capturing subjective experience is the use of Descriptive Analysis (DA; Heymann, King, & Hopfer, 2014) in food pairing research, which, in general, requires the subjects to give voice to their experiences. Several DA studies (Madrigal-Galan & Heymann, 2006; Nygren et al., 2002) have provided some key insight into the interaction of the paradigmatic cheese and wine pairing: particularly, that the cheese helps to ameliorate potentially unpleasant flavor characteristics of the wine. Note that this insight is not possible from the affective conception of food pairing, and required allowing the subjects of these studies to express their specific experiences—that is, giving voice to the “why” of food pairing. More recently, the adoption of Temporal Dominance of Sensation (TDS), which gives a more nuanced picture of specific eating occasions on the basis of the “mouthful,” has allowed detailed insight into, again, wine and cheese pairing (but see also and earlier Dinnella, Masi, Zoboli, & Monteleone, 2012). Researchers have overall confirmed that cheese seems to help “smooth out” the wine experience € om, 2017), but that, interestingly, there (Galmarini et al., 2016; Nygren, Nilsen, & Ostr€ is not necessarily a corresponding effect of wine on cheese (Galmarini et al., 2017), even though the contrasting flavors of the wine might, for example, be expected to initiate some sort of effect such as release from suppressing adaptation to the cheese flavor (Lawless & Heymann, 2010). While perhaps these specific results cannot generalize beyond wine and cheese, these applications of novel sensory-evaluation methodologies offer a rigorous approach for studying the “why” of food pairings once they have been identified. But how to identify food pairings in the first place, if concepts of balance, harmony, and Flavor Pairing are not the promising theory tools we once hoped they would be?

Food combinations and food and beverage combinations in meals

317

Here I would suggest that food and consumer scientists look to the methods of social scientists and humanists to generate better hypotheses. For example, to pick up a previous thread, perhaps pinot noir and salmon go well together, not just because they are good to eat together, but because they are good to think together (Trubek, 2008): the two products are both produced in the US, in the Pacific Northwest, and their shared geography forms a conceptual unity that may help to lead the consumer to a pleasant pairing experience. To my knowledge, none of the sensory studies cited here have explored the role of conceptual unity in encouraging perceptions of food pairing: might the super-additivity we seek to observe stem not just from the physical and sensorial attributes of the foods, but from the individuals’ engagement with expectations and cultural norms surrounding the food products? There is, however, a large and well-elaborated food-studies literature emerging from anthropology, philosophy, and sociology that looks at many of these issues (for a non-exhaustive introduction, see Adapon, 2008, Besky, 2013, Howes, 2003, Korsmeyer, 2002, Lahne & Trubek, 2014, Rozin, 2000, Sutton, 2010, Trubek, 2008). As sensory scientists seek to understand the impact of food attributes that stem from non-physical food properties, we need to engage with this body of scholarship or risk missing important aspects of how our subjects actually encounter their foods in their everyday lives (Chabrol & Muchnik, 2011; Hennion, 2015; Teil, 2011; Teil & Hennion, 2004). Furthermore, it is clear that the study of food pairings, like many aspects of eating behavior, needs to begin outside the sensory laboratory. The control of subjects that is so fundamental to our experimental precision is also antithetical to the eating experiences we wish to understand (K€ oster, 2009; Lahne, 2018). We need to work to develop partnerships with both observational researchers such as those noted herein, and with culinary researchers and professionals who can help us bring that observation into the context of the meal (Giboreau, 2017). By combining more ecologically valid meal-based research (Lahne, Pepino, & Zellner, 2016; Lahne & Zellner, 2015; Nygren et al., 2017) with new and sensitive methods such as TDS (Galmarini et al., 2016; Galmarini et al., 2017), we can create mixed-methods research that both develops emic theories of “why” food pairings happen, and then tests these theories, all in more ecologically appropriate circumstances.

15.8

Conclusion: A “manifesto” for researching food pairing

If there is one clear direction for food-pairing research that comes from this chapter, it is that we need new, flexible, and necessarily limited theories for both defining and explaining food pairings. Current theories—that food pairing is only super-additive liking, that food pairings come from the physiochemical or intrinsic sensory properties of foods, and that food pairings are culturally invariant—are clearly insufficient to explain the diversity of pairings both at different scales—molecular, ingredient, between foods, between foods and beverages, between different courses in a meal—and across different meal contexts—temporal, social, and cultural. Rather than

318

Context

continuing to test small theories that have generally not provided adequate explanatory power, we need to rethink our approaches to food pairing, and accept new research objectives and paradigms. We can and should employ more qualitative and mixed-methods research to provide better “why” explanations for food pairings, and be more open—as individuals and as a discipline—to theories that locate perceptual experiences—such as food pairings—not only in chemosensory and neurological phenomena, but in psychological and sociological origins. Our increasing sophistication with experimental methodologies such as TDS needs to be accompanied by an increasingly sophisticated set of theories about human behaviors and interactions. Only by combining these disciplinary competencies with new insights from other disciplines can we hope to produce satisfying explanations of phenomena—such as food pairings—that find their natural home in the rich contexts of everyday human experiences.

References Adapon, J. (2008). Culinary art and anthropology. Oxford and New York: Berg. Ahn, Y.-Y., Ahnert, S. E., Bagrow, J. P., & Baraba´si, A.-L. (2011). Flavor network and the principles of food pairing. Scientific Reports, 1. Bastian, S., Payne, C., Perrenoud, B., Joscelyne, V., & Johnson, T. (2009). Comparisons between Australian consumers’ and industry experts’ perceptions of ideal wine and cheese combinations. Australian Journal of Grape and Wine Research, 15, 175–184. Bastian, S. E. P., Collins, C., & Johnson, T. E. (2010). Understanding consumer preferences for shiraz wine and Cheddar cheese pairings. Food Quality and Preference, 21, 668–678. Besky, S. (2013). The Darjeeling distinction: Labor and justice on fair-trade tea plantations in India. Univ of California Press. Bourdieu, P. (1984). Distinction: A social critique of the judgment of taste. Cambridge, MA: Harvard University Press. Cardello, A. V. (1995). Food quality: Relativity, context and consumer expectations. Food Quality and Preference, 6, 163–170. Carroll, A. (2013). Three squares: The invention of the American meal. New York: Basic Books. Cerretani, L., Biasini, G., Bonoli-Carbognin, M., & Bendini, A. (2007). Harmony of virgin olive oil and food pairing: A methodological proposal. Journal of Sensory Studies, 22, 403–416. Chabrol, D., & Muchnik, J. (2011). Consumer skills contribute to maintaining and diffusing heritage food products. Anthropology of Food [Online], 8. Available: http://aof.revues. org/index6847.html. Accessed 5 July 2011. de Klepper, M. (2011). Food pairing theory: A European fad. Gastronomica: The Journal of Critical Food Studies, 11, 55–58. Devellis, R. F. (2016). Scale development: Theory and applications. Thousand Oaks, CA: SAGE. Dinnella, C., Masi, C., Zoboli, G., & Monteleone, E. (2012). Sensory functionality of extra-virgin olive oil in vegetable foods assessed by temporal dominance of sensations and descriptive analysis. Food Quality and Preference, 26, 141–150. Donadini, G., & Fumi, M. D. (2014). An investigation on the appropriateness of chocolate to match tea and coffee. Food Research International, 63, 464–476.

Food combinations and food and beverage combinations in meals

319

Donadini, G., Fumi, M. D., & Lambri, M. (2012). The hedonic response to chocolate and beverage pairing: A preliminary study. Food Research International, 48, 703–711. Donadini, G., Fumi, M. D., & Lambri, M. (2013). A preliminary study investigating consumer preference for cheese and beer pairings. Food Quality and Preference, 30, 217–228. Donadini, G., Fumi, M. D., & Newby-Clark, I. (2015). An investigation of matches of bottom fermented red beers with cheeses. Food Research International, 67, 376–389. Donadini, G., Spigno, G., Fumi, M. D., & Pastori, R. (2008). Evaluation of ideal everyday Italian food and beer pairings with regular consumers and food and beverage experts. Journal of the Institute of Brewing, 114, 329–342. Dornenburg, A., & Page, K. (2006). What to drink with what you eat. New York: Bulfinch Press. Dupuis, E. M. (2015). Dangerous digestion: The politics of American dietary advice. Oakland: University of California Press. Eindhoven, J., & Peryam, D. (1959). Measurement of preferences for food combinations. Food Technology, 13, 379–382. Eschevins, A., Giboreau, A., Allard, T., & Dacremont, C. (2017). The role of aromatic similarity in food and beverage pairing. Food Quality and Preference. Fujita, A., Isogai, A., Endo, M., Utsunomiya, H., Nakano, S., & Iwata, H. (2010). Effects of sulfur dioxide on formation of fishy off-odor and undesirable taste in wine consumed with seafood. Journal of Agricultural and Food Chemistry, 58, 4414–4420. Galmarini, M. V., Loiseau, A.-L., Debreyer, D., Visalli, M., & Schlich, P. (2017). Use of multiintake temporal dominance of sensations (TDS) to evaluate the influence of wine on cheese perception. Journal of Food Science, 82, 2669–2678. Galmarini, M. V., Loiseau, A.-L., Visalli, M., & Schlich, P. (2016). Use of multi-intake temporal dominance of sensations (TDS) to evaluate the influence of cheese on wine perception. Journal of Food Science, 81, S2566–S2577. Giboreau, A. (2017). Sensory and consumer research in culinary approaches to food. Current Opinion in Food Science, 15, 87–92. Harrington, R. J. (2005). The wine and food pairing process: Using culinary and sensory perspectives. Journal of Culinary Science & Technology, 4, 101–112. Harrington, R. J. (2007). Food and wine pairing: A sensory experience. John Wiley & Sons. Harrington, R. J., & Hammond, R. (2005). The direct effects of wine and cheese characteristics on perceived match. Journal of Foodservice Business Research, 8, 37–54. Harrington, R. J., Mccarthy, M., & Gozzi, M. (2010). Perceived match of wine and cheese and the impact of additional food elements: A preliminary study. Journal of Foodservice Business Research, 13, 311–330. Harrington, R. J., & Seo, H.-S. (2015). The impact of liking of wine and food items on perceptions of wine–food pairing. Journal of Foodservice Business Research, 18, 489–501. Hennion, A. (2007). Those things that hold us together: Taste and sociology. Cultural Sociology, 1, 97–114. Hennion, A. (2015). Paying attention: What is tasting wine about? In A. B. Antal, M. Hutter, & D. Stark (Eds.), Moments of valuation: Exploring sites of dissonance. Oxford: Oxford University Press. Herz, R. S. (2005). Odor-associative learning and emotion: Effects on perception and behavior. Chemical Senses, 30, i250–i251. Heymann, H., King, E. S., & Hopfer, H. (2014). Classical descriptive analysis. In P. Varela & G. Ares (Eds.), Novel techniques in sensory characterization and consumer profiling. Boca Raton: CRC Press.

320

Context

Howes, D. (2003). Sensual relations: Engaging the senses in culture and social theory. Ann Arbor: University of Michigan Press. Jain, A., Rakhi, N., & Bagler, G. (2015). Analysis of food pairing in regional cuisines of India. PLoS One, 10, e0139539. King, M., & Cliff, M. (2005). Evaluation of ideal wine and cheese pairs using a deviation-fromideal scale with food and wine experts. Journal of Food Quality, 28, 245–256. Korsmeyer, C. (2002). Making sense of taste: Food and philosophy. Cornell University Press. Kort, M., Nijssen, B., van Ingen-Visscher, K., & Donders, J. (2010). Food pairing from the perspective of the ‘volatile compounds in food’database. In: Expression of Multidisciplinary Flavour Science: Proceedings of the 12th Weurman Symposium, Interlaken, Switzerland, Institut of Chemistry and Biological Chemistry, pp. 589–592. K€ oster, E. P. (2003). The psychology of food choice: Some often encountered fallacies. Food Quality and Preference, 14, 359–373. K€ oster, E. P. (2009). Diversity in the determinants of food choice: A psychological perspective. Food Quality and Preference, 20, 70–82. Lahne, J. (2018). Evaluation of meals and food pairing. In G. Ares & P. V. Tomasco (Eds.), Methods in consumer research. Duxford: Woodhead. Lahne, J., Pepino, R., & Zellner, D. (2016). You’ll spoil your dinner: Attenuating hedonic contrast in meals through cuisine mismatch. Food Quality and Preference, 56, 101–106. Lahne, J., & Trubek, A. (2014). “A little information excites us.” Consumer sensory experience of Vermont artisan cheese as active practice. Appetite, 78, 129–138. Lahne, J., & Zellner, D. A. (2015). The great is the enemy of the good: Hedonic contrast in a coursed meal. Food Quality and Preference, 45, 70–74. Lawless, H. T., & Heymann, H. (2010). Sensory evaluation of food: Principles and practices. New York: Springer. Lucak, C., & Delwiche, J. (2009). Efficacy of various palate cleansers with representative foods. Chemosensory Perception, 2, 32–39. Madrigal-Galan, B., & Heymann, H. (2006). Sensory effects of consuming cheese prior to evaluating red wine flavor. American Journal of Enology and Viticulture, 57, 12–22. Meiselman, H. L. (2000). Dimensions of the meal: The science, culture, business and art of eating. Gaithersburg: Aspen Publishers. Meiselman, H. L. (2017). An historical perspective on variety in United States dining based on menus. Appetite, 118, 174–179. Mela, D. J. (2001). Why do we like what we like? Journal of the Science of Food and Agriculture, 81, 10–16. Mielby, L. H., & Frøst, M. B. (2010). Expectations and surprise in a molecular gastronomic meal. Food Quality and Preference, 21, 213–224. Møller, P. (2013). Gastrophysics in the brain and body. Flavour, 2, 1–4. Moskowitz, H., & Klarman, L. (1977). Food compatibilities and menu planning. Canadian Institute of Food Science and Technology Journal, 10, 257–264. Moskowitz, H. W., Kumaraiah, V., Sharma, K. N., Jacobs, H. L., & Sharma, S. D. (1975). Cross-cultural differences in simple taste preferences. Science, 190, 1217–1218. Newman, J. (2000). Chinese meals. In H. L. Meiselman (Ed.), Dimensions of the meal. Gaithersburg: Aspen. Nygren, I. T., Gustafsson, I. B., & Johansson, L. (2002). Perceived flavour changes in white wine after tasting blue mould cheese. Food service Technology, 2, 163–171. ˚ . (2017). Dynamic changes of taste experiences in wine € om, A Nygren, T., Nilsen, A. N., & Ostr€ and cheese combinations. Journal of Wine Research, 28, 105–122.

Food combinations and food and beverage combinations in meals

321

Otsuka, S. (2000). Japanese meals. In H. L. Meiselman (Ed.), Dimensions of the meal. Gaithersburg: Aspen. Paulsen, M. T., Rognsa˚, G. H., & Hersleth, M. (2015). Consumer perception of food–beverage pairings: The influence of unity in variety and balance. International Journal of Gastronomy and Food Science, 2, 83–92. Perkel, J. (2012). The new molecular gastronomy, or, a gustatory tour of network analysis. BioTechniques, 53, 19–22. Rietz, C. A. (1961). A guide to the selection, combination, and cooking of foods. Westport, Conn: AVI Pub. Co. Rozin, E. (2000). The role of flavor in the meal and the culture. In H. L. Meiselman (Ed.), Dimensions of the meal. Gaithersburg: Aspen. Rozin, P., & Vollmecke, T. A. (1986). Food Likes and Dislikes. Annual Review of Nutrition, 6, 433–456. Shotwell, A. (2016). Against purity. Minneapolis: University of Minnesota Press. Spence, C., & Piqueras-Fiszman, B. (2014). The perfect meal: the multisensory science of food and dining. Chichester, West Sussex, UK: John Wiley & Sons Inc. Sutton, D. E. (2001). Remembrance of repasts: An anthropology of food and memory. New York: Berg. Sutton, D. E. (2010). Food and the senses. Annual Review of Anthropology, 39, 209–223. Tamura, T., Taniguchi, K., Suzuki, Y., Okubo, T., Takata, R., & Konno, T. (2009). Iron is an essential cause of fishy aftertaste formation in wine and seafood pairing. Journal of Agricultural and Food Chemistry, 57, 8550–8556. Tanis, D. (2018). Meat and potatoes made magical. The New York Times, February 9. Teil, G. (2011). No such thing as terroir? Science, Technology, & Human Values, 37, 478–505. Teil, G., & Hennion, A. (2004). Discovering quality or performing taste? A sociology of the amateur. In M. Harvey, A. Mcmeekin, & A. Warde (Eds.), Qualities of food. Manchester and New York: Manchester University Press. The Food Pairing Company (2017). Foodpairing [Online]. Available: http://www.foodpairing. com/en/science-behind. Accessed 20 December 2017. Traynor, M. P., Burke, R., O’Sullivan, M. G., Hannon, J. A., & Barry-Ryan, C. (2013). Sensory and chemical interactions of food pairings (basmati rice, bacon and extra virgin olive oil) with banana. Food Research International, 54, 569–577. Trubek, A. (2008). The taste of place: A cultural journey into terroir. Berkeley: University of California Press. Tsing, A. L. (2015). The mushroom at the end of the world. Princeton: Princeton University Press. Visser, M. (1991). The rituals of dinner: The origins, evolution, eccentricities, and meaning of table manners. New York: Penguin. Wang, S.-T. (2017). When Chinese cuisine meets western wine. International Journal of Gastronomy and Food Science, 7, 32–40.

This page intentionally left blank

Virtual reality and immersive approaches to contextual food testing

16

Christina Hartmann, Michael Siegrist Department of Health Science and Technology, Consumer Behavior, ETH Zurich, Z€urich, Switzerland

16.1

Introduction

Virtual reality (VR) provides fascinating possibilities, not only for gaming, entertainment, education, and training, but also for research (Fox, Arena, & Bailenson, 2009). In VR, a user experiences and interacts with a computer-generated virtual environment ( Jerald, 2015). The degree to which the user interacts with the objects in the VR environment differs across applications. In a fully immersive virtual environment, the user can walk around in the virtual environment and explore it, grasp objects, and possibly receive feedback for different senses. A user exposed to VR will be more or less immersed in a simulated scene. The intensity of the experienced immersion depends on the fidelity with which the virtual environment is represented. In other words, how closely the reactions of the VR system mimic the reactions of the real environment is crucial for the experienced immersion. The goal of this chapter is to summarize existing studies that have used different forms of VR (e.g., video walls, head mounted display) in the domain of consumer research. One important aspect we will focus on is how valid the observed behavior in VR is. Throughout this chapter, we refer to articles that were identified via a systematic literature search that was conducted in January 2018 in the Web of Science Core collection database. The keywords included “virtual reality,” or “immersive or augmented reality,” and “food,” or “eat,” or “drink.” Articles based on reference lists and abstracts from conferences were screened for eligible research as well. The design and development of research in contextual food evaluation and consumer behavior by means of VR applications is in its early stages, and only a few studies have been published thus far. Based on the found literature, four main application fields of any kind of VR in the context of food behavior research were identified: sensory evaluation of food in context; food shopping behavior; influence of environmental cues on eating behavior; and treatment of eating disorders in controlled environments. Some studies have focused on the applicability and validity of the technological application in the food context, while others have tested the products’ sensory appeal in different environments. Comparatively, more research has been conducted with the focus on eating disorders and three-dimensional immersive technologies for the therapy and Context. https://doi.org/10.1016/B978-0-12-814495-4.00016-7 Copyright © 2019 Elsevier Inc. All rights reserved.

324

Context

treatment of eating disorders in clinical populations. However, in the following, these studies are only of marginal interest. A very limited number of studies have used a fully immersive VR technique in which the person wears a head-mounted display (HMD), and is thus surrounded by the artificial environment and can interact with the environment by, for example, grasping something or walking around (e.g., Schnack, Wright, & Holdershaw, 2018; Siegrist et al., in press; Ung, Menozzi, Hartmann, & Siegrist, 2018). Most studies were based on some kind of virtual environment whereby the participants sit in front of a video wall or PC screen that displays two- or three-dimensional (2D or 3D) pictures of the environment; also called mixed reality. In addition to the visual stimuli, some studies also used olfactory and/or auditory input in order to make the simulation more immersive and realistic (e.g., Bangcuyo et al., 2015; Hathaway & Simons, 2017).

16.2

A new research paradigm

Researchers estimated that adults make >200 food decisions every day in diverse environments (Wansink & Sobal, 2007). It is almost impossible to observe consumers’ behavior in their natural habitats within these diverse environments. Not only would it require significant financial resources, but it is also difficult to control contextual factors, which might influence the observed outcome variables. Standardization in experimental research is an important aspect to ensure that the observed effect can be attributed to systematically manipulated factors, rather than to unobserved variables in the environment that might even underlie variability from participant to participant. Thus, many consumer behavior researchers focus on internal validity of their experiment by conducting standardized lab experiments, which are necessary to identify cause and effect relationships. Whether the results of such studies are generalizable to other contexts and populations, and are therefore externally valid, is often unclear. To ban standardized lab conditions and to conduct experiments in real-life settings only is accompanied with a loss of internal validity, however. In nutrition physiology research, the animal model―the mouse model in particular―is the gold standard for research. The test animals live and undergo treatment in highly controlled and rudimentary environments in order to have no confounding influence in the experiment that might have an effect on the outcome. In product development, including sensory testing under lab conditions, it is the traditional sensory booth that is supposed to suppress environmental cues and external distractors that might influence the evaluation of the product. However, such testing systems and environments lack ecological validity and do not necessarily reflect pre-purchase product evaluation. Ecological validity is especially important when the results from the lab will be generalized to real-life settings ( Jaeger & Porcherot, 2017). Contextual cues of a typical consumption environment might influence product evaluation in a positive or negative manner. VR and immersive environments pose the potential to overcome such limitations of traditional test systems by achieving both; that is, control over environmental cues that are equal for all study participants

Virtual reality and immersive approaches to contextual food testing

325

and, simultaneously, higher ecological validity than in a lab test. Full immersive environments and VR can be used to test consumers’ purchases and food decisions in different environments and environmental settings that are difficult to realize in real life. Even though these artificial environments are not a complete simulation of the real world, they enable the same study participant to be exposed to multiple environments where food decisions take place, such as a buffet, the supermarket, the kitchen, or the restaurant. There is no dependence on the availability of certain foods or influences of product variability. VR and immersive environments have a lot of advantages, but in order to establish a new paradigm for consumer behavior research, it is of central importance to determine whether people in the artificial environment behave the same way as they do in real life.

16.3

Food selection behavior

16.3.1 Food buffet In a recent study, the present authors examined whether VR can be used as a research paradigm to examine people’s food choice behavior, and how strongly people’s VR behavior correlates with their behavior in a real-world setting (Ung et al., 2018). For the real-world setting, a buffet was set up on a table in an otherwise empty room, and it was based on three replica food items, that is, carrots, pasta, and chicken. The use of replica foods has some advantages compared with real foods, for example, no variability in appearance (Bucher, van Der Horst, & Siegrist, 2012). The external validity of using replica food items for real food choices has been demonstrated in previous studies (Bucher et al., 2012). The VR buffet was composed based on the same three food items, and the appearance of the room was similar to the room in the real-life setting (Fig. 16.1).

Fig. 16.1 The experimental setting: (left) The experimental setting of the fake food buffet; (right) The experimental setting of the virtual food buffet.

326

Context

The VR system consisted of an HMD and a purpose-built hand-tracking system, both connected to a desktop computer. A detailed description of the system can be found in Ung et al. (2018). The hand-tracking system enabled the researcher to record the position and orientation of the right hand, and allowed the participants to perform the same scooping task as in the real-world fake food buffet setting. The same study participants served themselves a meal from the fake food buffet and the VR food buffet (within-design). The results showed that the energy content of the food that was served in the real-life setting was highly correlated with the energy content of the food served in the VR environment. This holds true for the single meal components (carrots r ¼ .77, pasta r ¼ .75, chicken r ¼ .75; all P < .001) as well as for the total meal (r ¼ .81, P < .001). There was no statistically significant difference in the energy content of the served single food components, but there was a tendency toward a slightly higher total energy content for the complete meals composed in VR. While the meals based on replica food items contained on average 1401 kJ (SD ¼ 469.1 kJ), the meals composed in VR contained on average 1527 kJ (SD ¼ 476.1 kJ). Thus, the participants served themselves meals in the VR condition that contained approximately 8% more energy, that is, 126 kJ or 30 kcal (calories). This difference is very small, and not of practical relevance here. Even though the study participants did not receive any kind of haptic feedback, such as an increase in the weight of the plate or textural information, while they served themselves in the virtual environment, the results showed good agreement between the real-life and the VR conditions. This was the first study to show that the VR food buffet is a useful tool for the assessment and investigation of food choice behavior. More studies with other foods and in other settings are needed in order to further validate the VR buffet. Nevertheless, this new approach enables the manipulation of factors in food and nonfood environments, and examines their impact in a standardized manner. This is especially of interest for a better understanding of factors that influence people’s food choice behavior in buffet situations or self-serving points, such as those found in restaurants, supermarkets, and canteens. This study suggests a lot of potential for further research, such as testing a broader variety of foods, the presence of others on food choices, and changes in the testing environment on people’s behavior.

16.3.2 Supermarkets Comparatively more research on the potential of VR was conducted to assess virtual shopping behavior for academic and commercial research (for a summary see Burke, 2018). More recent studies in the food domain have studied the impact of environmental factors on food purchasing behavior in virtual supermarkets. For example, studies were conducted that assessed the effect of price changes, for example, taxation and discounts, on consumers’ purchasing behavior (Poelman et al., 2017; Waterlander, Steenhuis, de Boer, Schuit, & Seidell, 2013), consumers’ reactions toward innovative food products (Bressoud, 2013) and food labels (Liu, Hooker, Parasidis, & Simons, 2017), as well as consumers’ purchasing behavior toward non-standard fruits and vegetables (Verhulst, Normand, Lombard, & Moreau, 2017). Some of these studies collected data by using an HMD with or without the possibility to interact with the

Virtual reality and immersive approaches to contextual food testing

327

environment (Verhulst et al., 2017), while others relied on a computer desktop system with 3D pictures displayed via one or more PC screens (Bressoud, 2013; van Herpen, van den Broek, van Trijp, & Yu, 2016; Waterlander, Jiang, Steenhuis, & Ni Mhurchu, 2015). A few articles were published in which one research question was whether people’s behavior in VR is comparable to their behavior in real life. Van Herpen et al. (2016) compared shopping behavior in a 3D virtual supermarket displayed on three PC screens with shopping behavior in a 2D pictorial representation of the same supermarket, and shopping behavior in a physical store. They assessed a range of factors in three product categories, such as number, variety, and type of products selected, money spent, and responses to price promotions and shelf displays. They found mixed results. For the number of products bought, money spent and chosen shelf displays, the virtual supermarket was superior to the 2D picture condition, and a better representation of consumers’ behavior in a physical store. However, in both the 2D and 3D conditions, the participants bought more, spent more money, exhibited more variety seeking, and responded more strongly to price promotions compared with the real store setting. The authors further concluded that for some measures, the virtual supermarket improves the representation of a shop under lab conditions toward a 2D picture-based replication. However, for others, a difference between behavior in a physical store and the 3D and 2D display stores persists. Waterlander et al. (2015) found similar purchasing patterns in a 3D supermarket compared with a real supermarket, and no trend of overspending. However, they also observed differences in purchasing patterns on the level of individual food groups. Burke (2018) came to a similar conclusion; shoppers seemed to purchase larger quantities in the simulated store than in the physical store, and seemed to be more responsive to price features. These validation studies are highly relevant; however, they did not take full advantage of VR for conducting consumer research. People were seated in front of computer screens, and they could not physically walk around and interact with the virtual environment. In order to test whether consumers behave similarly in an interactive VR, Siegrist et al. (in press) conducted two experimental studies. The aim of the first study was to compare participants’ food selection behavior while standing in front of real and virtual supermarket shelves. The products used in this study were 33 cereals commonly found in supermarkets. The study participants received two shopping tasks, because the researchers were not interested in habitual shopping behavior, and wanted to avoid the participants just selecting their favorite cereal brand without going through a comprehensive selection process. In the first task, they were instructed to select a cereal for a kid’s camp with children aged 10–12. In the second task, the participants were instructed to buy a cereal for a friend who was on a low-sugar diet. In this second task, it was expected that the participants would need to frequently check the nutrition information. The study participants were randomly assigned to the VR or the real-life condition. In the VR condition, the participants were equipped with an HMD (Oculus Rift DK 2, Oculus VR, Inc. Menlo Park, California, USA) and a hand-tracking device that enabled the researchers to track participants’ grasping behavior. In both conditions, the participants were equipped with an eye-tracking

328

Context

system, and eye movements and gaze direction could be assessed. A detailed description of the system can be found in Siegrist et al. (in press). When people behave similarly in VR and in real life, one would expect them to show the same information-seeking behavior in both conditions; they would focus on similar packages and areas of the shelves, and they would show no differences in product selection. In fact, the results for both tasks showed that information-seeking behavior and the shelf from which the cereal was chosen did not differ in the two conditions. Relative fixation duration was slightly higher in the VR condition, probably because people explored the less familiar virtual environment. The second study involved similar tasks; however, here, the realism of the virtual environment was enhanced because the study participants could walk around in the virtual supermarket as they would in reality (Fig. 16.2). The aim of the study by Siegrist et al. (in press) was to replicate in a VR supermarket the findings of a study conducted in the real word (Visschers, Hess, & Siegrist, 2010). In the real world study, eye-trackers and real products were used. The participants had to select one out of five different cereals. The results of that study suggested that participants with an induced health motivation looked at the nutrition information on the food product longer than those with an induced taste motivation. Therefore, Siegrist et al. (in press) hypothesized for their VR study that, participants with a health

(A)

(B)

(C) Fig. 16.2 Virtual testing environment seen through a head-mounted display and connected with a real walking virtual environment technique used in the study by (Siegrist et al., in press). (A) Top view of the layout of the virtual supermarket, (B) aisle of the virtual supermarket with shopping basket, (C) cereal shelves.

Virtual reality and immersive approaches to contextual food testing

329

motivation will look more often at nutritional information and spend more time looking at the cereal packages compared with study participants with the taste motivation. The participants in the health motivation condition received this directive: “Imagine that you need to buy healthy breakfast cereals for a kindergarten. In the virtual supermarket you need to search the shelf with the breakfast cereals, and then you need to select one package. Once you have selected a product, put it in the shopping cart and leave the shop.” The participants in the taste motivation condition received this directive: “Imagine that you need to bring good-tasting breakfast cereals to a breakfast with friends. In the virtual supermarket you need to search the shelf with the breakfast cereals, and then you need to select one package. Once you have selected a product, put it in the shopping cart and leave the shop.” All participants were randomly assigned to the health or the taste condition. The time spent in the virtual shop and the breakfast cereal chosen were compared between the two conditions. As expected, the participants in the health condition spent more time in the shop, and looked at nutrition tables more often and for a longer duration compared with participants in the taste condition. The results of the following three variables are shown in Table 16.1: (1) Number of times participants looked at the front side of a different package; (2) Number of times participants looked at the back side of a different package; and (3) Number of times participants looked up values in the nutrition table. As shown in Table 16.1, the results were in line with the expectations. The participants in the health condition significantly more often looked at information compared with the participants in the taste condition. First, these results suggest that participants behave similarly in the real world (Visschers et al., 2010) and in VR (Siegrist et al., in press). Second, the results provide additional insights into what information the participants

Table 16.1 How often participants looked at packages and nutrition information in a VR experiment with a health group and a taste group Condition

Number of times looked at the front side of a package Number of times looked at the back side of a package Number of times values looked up in the nutrition table

Health group (n 5 20)

Taste group (n 5 22)

M (SD)

M (SD)

48.95 (23.28)

37.00 (18.55) 2.59 (3.73)

7.85 (8.20) 17.10 (15.40)

10.05 (15.83)

Mann-Whitney U test, z-value 1.83* 2.54** 1.81*

Note: One-tailed tests, * P < 0.05, ** P < 0.01. Participants who looked at all packages or who selected more than one cereal package were excluded from the analyses.

330

Context

took into account when evaluating the healthiness of the cereals. As expected, the nutrition information was important, because the participants in the health condition looked more often at the numbers in the nutrition tables compared with the participants in the taste condition. Furthermore, the same group difference was observed for the front side of the packages. This result suggests that consumers not only use the nutrition information to evaluate the healthiness of products, but also the potentially misleading information on the front side of the package. These studies were some of the first to show the process of product selection, and not only the outcome, that is, which product was chosen. In particular, consumers’ information-seeking and selection strategies and, consequently, their purchase decisions were shown to be similar in the two environments. Consumers’ attention to different products and product information was comparable, which aims to further justify the use of VR in consumer research.

16.4

Food-evoked emotions in VR

Mood and situational emotions can have a profound impact on food evaluation. While the first one is typically accompanied by physiological arousal and lasts for several minutes, the latter is a short-term affective response to certain stimuli (Gibson, 2006). Moods and emotions can be evoked by environmental aspects or the food product itself, which can consequently influence the evaluation of the food and the food choice and purchase behavior (e.g., Edwards, Hartwell, & Giboreau, 2016; King, 2016). In psychological research, it is a standard method to use food pictures in order to evoke positive, for example, happy, or negative, for example, disgust, emotional reactions or to provoke food cravings following visual exposure to high caloric food items (e.g., Harrar, Toepel, Murray, & Spence, 2011). However, compared with real food, affective reactions evoked by food pictures seem to be weaker (Romero, Compton, Yang, & Snow, 2018). Researchers have termed it the “real-exposure effect”―a phenomenon showing that people attribute higher value to real foods compared with their picture-based representations (Bushong, King, Camerer, & Rangel, 2010). Building on the potential of VR for food research, it is crucially important to know whether food presented in VR evokes the same emotional arousal as real food presentations, and whether VR can be helpful in overcoming the limitations of picturebased food item evaluations. In this regard, it is also pertinent to acknowledge that the virtual environment itself can induce certain mood states, which is of relevance for studying the influence of external cues on food behavior. One example of such a research study was conducted by Felnhofer et al. (2015). In their study, participants were exposed to different virtual park scenarios with each eliciting another affective state, that is, joy, sadness, boredom, anger, and anxiety. Mood-inducing elements in this study were certain sounds (e.g., chirping birds), lighting (e.g., darkness), and weather conditions (e.g., sunny weather). The results showed that the various immersive virtual environments succeeded in eliciting the intended mood states. Just by changing a few elements in the virtual environment, the authors were able to elicit one positive and four negative mood states (Felnhofer et al., 2015).

Virtual reality and immersive approaches to contextual food testing

331

As outlined herein, not only can the environment influence affective reactions in real-life settings, but also the food product itself. From studies with clinical populations, such as participants with eating disorders, for example, bulimia nervosa and anorexia, we know that virtual food exposure alone, without the presence of any auditory or olfactory cues, can provoke negative affective responses (Gorini, Griez, Petrova, & Riva, 2010). In addition, these effects were stronger for 3D VR displays of snack foods (via HMD) than for photos. With regard to studies on non-clinical samples and food-evoked emotions in VR, research is scarce. A couple of studies were conducted with regard to food craving, which is defined as an appetitive motivationalemotional state that promotes the search and consumption of foods; mostly highcaloric food (Hill, Weaver, & Blundell, 1991). In a study with 55 non-dieting normal weight women and the utilization of a 3D image of a restaurant buffet (HMD), researchers showed that the food cravings evoked by VR displays did not differ from those evoked by pictures. However, both VR and 3D evoked weaker reactions compared with exposure to real snack food items (Ledoux, Nguyen, Bakos-Block, & Bordnick, 2013). Later, these effects could be reproduced with women and varying VR scenarios that influenced self-reported food cravings (Ferrer-Garcia, GutierrezMaldonado, Treasure, & Vilalta-Abella, 2015). The participants were exposed to high- and low-calorie food scenarios via a 3D PC monitor and polarized glasses. They reported higher levels of food craving after the high-calorie food pictures than after the low-calorie food scenarios. Even though no comparison between the VR environment and real foods or 2D food pictures was performed in this study, the results showed that the VR environments provoked the expected emotional reactions. Nevertheless, there is the risk that the intensity of the arousal is underestimated with VR applications compared with real-life settings. In clinical populations with maladaptive eating symptomatology, such as binge-eating disorder, people pass through phases of high arousal in response to viewing food images (e.g., Schienle, Sch€afer, Hermann, & Vaitl, 2009). Affective reactions following exposure to food and context-related stimuli are presumably lower in persons from the general population. Therefore, it needs to be examined whether presenting foods in VR induces emotional responses similar to those expected in real life in clinical and non-clinical populations. Technical improvements toward more realistic high-resolution graphics of food images in specific contexts within virtual environments will probably increase the chances of people being equally aroused by virtual stimuli and real-food stimuli. Moreover, further research is underway that aims to identify specific cues and contexts that need to be present in VR in order to provoke a certain eating behavioral response, such as cravings in clinical and non-clinical populations (Pla-Sanjuanelo et al., 2015). In addition, the realexposure effect seems to be determined by the accessibility of the food item (Bushong et al., 2010; Romero et al., 2018). Even seeing the real food item behind a transparent barrier reduced its attributed value (Bushong et al., 2010). Thus, at the moment, accessibility of the object might be an important factor for mimicking real-life food evaluations. Haptic feedback applications for VR might be one possibility to overcome such a limitation of 2D displays. Nevertheless, more research on food-evoked emotions is necessary to better understand the effectiveness of VR applications to produce significant emotional reactions in their users.

332

16.5

Context

Appropriateness of contextual cues

Next to visual stimuli, auditory and/or olfactory stimuli are important in contextual food research. They can be used in order to make a VR-based testing situation a realistic food experience that is able to evoke real-life-like reactions. Important to note is that the selection of these additional cues should be reasoned through, because they can influence the evaluation of the product. Bangcuyo et al. (2015) conducted a study in which participants had to indicate their liking for coffee that was sampled either in a sensory booth or in a virtual coffee house. Hathaway and Simons (2017) tested the liking of cookies in a traditional sensory booth and a fully immersive environment based on a video wall that showed men baking. Both studies used somewhat similar approaches to imitate contextual consumption. In particular, participants were exposed to audio-visual stimuli (PC screen or video wall) and olfactory stimuli, that is, coffee or cinnamon roll aromas. In both studies, the test products received higher liking scores in the virtual environments compared with the traditional sensory booths. Audio-visual and olfactory stimuli of a typical consumption context were used in the studies by Brasset, Gachet, Abiven, and Delarue (2017) and Glassl, Lutsch, and Scharf (2017). They tested the liking of beer varieties in a traditional sensory booth, nightclub, and beach scenario, and strawberry yogurt in a traditional sensory booth, mixed-reality, and real-canteen environments. However, no relevant effects of the consumption contexts on the liking of the food items were observed in these two studies. All four studies used somewhat similar approaches to imitate contextual consumption, but found divergent results (Table 16.2). One might speculate that in the first two studies, the smell of the sampled food items, that is, coffee or cinnamon roll, exerted an appetizing effect that led to higher product evaluations. Meanwhile, in the latter two studies, the focus was more on typical consumption environments that do not necessarily smell like the food sampled, that is, beach or nightclub. Thus, when olfactory cues are included in the environmental setting, careful consideration must be given to whether the smell of the food itself or the typical consumption environment will be used as contextual stimuli, because both might influence the evaluation of the product. Therefore, it is of importance whether a contextual evaluation of the food or the effect of contextual variables on food and drink consumption is the main objective of the research.

16.6

A word on presence

Compared with traditional testing environments, such as a sensory booth, study participants are more engaged in immersive and virtual environments, which can improve data quality (Hathaway & Simons, 2017; Schnack et al., 2018). However, it seems to be important for study participants to be given the opportunity to become familiar with the virtual environment with a short training session (Burke, 2018; Siegrist et al., in press). Otherwise, participants might focus on exploring the virtual environment instead of behaving naturally (Siegrist et al., in press). For example, an unrelated scenario could be used to promote familiarization with the virtual system.

Virtual reality and immersive approaches to contextual food testing

333

Table 16.2 Overview of four studies that tested the effect of the virtual environment on product evaluation Author

Test food

Testing environments

Main finding

(Bangcuyo et al., 2015)

Coffee

+

(Hathaway & Simons, 2017)

Cookies

(Brasset et al., 2017)

Alcoholfree Beer

(Glassl et al., 2017)

Strawberry yogurt

Traditional sensory booth vs. coffee house (video wall, coffee house sound, coffee smell) Traditional vs. mixed immersion (PC screen showing men baking video) vs. full immersion (video wall showing men baking, contextual sound, cinnamon roll aroma) Traditional vs. immersive room with a nightclub scenario vs. immersive room with a beach scenario (wide screen, ambient light, sound/music, ambient odor, breeze/ smoke) Traditional vs. traditional & mixed reality (video, sound) vs. CLT + mixed reality (furnished room, sound, music) vs. real canteen

+

Highest liking scores in the virtual coffee house Highest liking scores in the fully immersive environment



Very little effect of testing condition on liking scores



No increase in liking with higher degree of realism

Note: CLT, central location test.

Moreover, the degree to which participants experience a sense of “being” in the virtual environment can influence the degree to which they behave naturally in the virtual environment (Schnack et al., 2018; Waterlander et al., 2015). Some researchers also argue that feeling present in the virtual environment might be a prerequisite for experiencing affective states (Ban˜os et al., 2004; Felnhofer et al., 2015). Questionnaires such as the Presence Questionnaire (Witmer & Singer, 1998), the Multimodal Presence Scale (Makransky, Lilleholt, & Aaby, 2017), and the Engagement Questionnaire (Hathaway & Simons, 2017) enable assessing these aspects in a standardized manner (Table 16.3). All of these questionnaires measure different dimensions of the psychological construct (tele)presence. For example, the Multimodal Presence Scale assesses perceived physical presence based on attributes such as “not paying attention to the real-world environment” and “sense of control in the virtual environment.” Social- and self-presence refer to the extent to which participants experience another person and their own self in the virtual environment (Makransky et al., 2017). The Engagement Questionnaire that was used, for example,

334

Context

Table 16.3 Examples of questionnaires that were used in VR food research to assess perceived presence Questionnaire (source)

Dimensions

Example

Presence Questionnaire (Witmer & Singer, 1998)

32 items, 4 factors: control, sensory, distraction, realism

Multimodal Presence Scale (Makransky et al., 2017)

15 items, 3 subdimensions: physical, social, selfpresence

Engagement Questionnaire (Bangcuyo et al., 2015; Hathaway & Simons, 2017) adapted from O’Brien and Toms (2010) and Witmer and Singer (1998)

19–21 items, 8 dimensions: usability, environmental aesthetics, novelty, involvement, sensory awareness, immersion, realism, distraction

“How natural did your interactions with the environment seem?” “I felt like I was in the presence of another person in the virtual environment.” “How completely did you feel immersed in the testing environment?”

by Bangcuyo et al. (2015) focuses more on an evaluation of aspects of the virtual environment itself, such as the realism of the environment, and the involvement or usability of the environment, for the conducted experimental task. In a few studies, the assessment of (tele)presence was an important part of the study design. For example, in the study by Hathaway and Simons (2017), three test environments were explored―a traditional, a mixed immersive, and a fully immersive environment. They found differences in most of the dimensions of the Engagement Questionnaire in that higher levels of immersion resulted in higher scores on the presence questionnaire. Another example is Waterlander and colleagues (Waterlander et al., 2015), who assessed presence in order to validate the applicability of a virtual simulated supermarket for measuring food purchase behavior. Schnack et al. (2018) found that a fully immersive walk around a virtual supermarket was superior in terms of telepresence, but not usability, compared with a traditional desktop-based simulated virtual supermarket. The availability of technologies for creating immersive and virtual environments is increasing, and the use of such presence questionnaires enables comparisons of different test environments. They can also be used to identify subgroups, for example, older adults whose experiences in VR environments, for example, are not as natural as intended, which could result in reduced data quality.

16.7

Summary and future directions

In sum, the studies highlighted throughout this chapter allow the assumption that VR and immersive environments are promising tools for studying consumer behavior and product evaluation. Of course, these systems have their advantages and disadvantages (Box 16.1), which should be taken into account when planning studies.

Virtual reality and immersive approaches to contextual food testing

335

Box 16.1 Advantages and disadvantages of using VR techniques in food research Advantages  Products can be tested in context  Creating an environmental setting that is difficult to realize in real life  Enables the study of issues that are difficult to tackle in real life  Can be used for targeted and personalized training therapy  Studying the influence of external cues and context on behavior  Controllable conditions  External cues are equal for all study participants  No food waste, cooking effort, or problems with food variability/availability  Testing can be realized early in product development  Higher engagement of test persons compared with sensory booth  Exposure to multiple contexts can be realized Disadvantages  Distorted feelings of presence  Might be experienced as artificial and unnatural  Motion sickness possible1  Expenses for technical applications  Reducing internal validity by increasing external validity  Difficulties concerning handling of food while wearing VR glasses, for

example, eating something 1

Low computer processing power leads to a time lag between the user’s movement in the full immersive digital environment (HMD) and the updated virtual scene that can result in disorientation.

Only a very limited number of studies have been conducted thus far that tested food products in a VR environment based on an HMD, and with the possibility of walking around in the virtual environment. Technological improvements and more study efforts are necessary to further improve the virtual environments and explore their possibilities and limitations. In that vein, we also do not yet know how much contextual information is necessary in order to evoke sensations and emotions that mimic those experienced in real-life situations. In this regard, we also need to explore how much haptic feedback is required for a naturalistic perception of the product. In addition, most studies thus far have been conducted with students and young adults. Because older adults are an increasing population group, it is important to find out how they experience virtual environments. People from higher age groups without gaming experiences might have difficulties using VR technological applications, such as a hand motion-track controller (Schnack et al., 2018). Future studies are needed to assess the perceived presence and usability of virtual environments in higher age groups.

336

Context

Lab experiments concerned with product evaluations and purchasing behavior probably underestimate product acceptance and effect sizes. VR could be a means to increase the prediction accuracy of lab experiments (Schnack et al., 2018). There is, however, still room for improvement of VR research, and not only with regard to technological aspects that limit VR experiences. Nevertheless, further research is not only needed to determine the possibilities of VR for consumer research, but also to overcome limitations of the available equipment and the software (Schnack et al., 2018).

References Bangcuyo, R. G., Smith, K. J., Zumach, J. L., Pierce, A. M., Guttman, G. A., & Simons, C. T. (2015). The use of immersive technologies to improve consumer testing: The role of ecological validity, context and engagement in evaluating coffee. Food Quality and Preference, 41, 84–95. Ban˜os, R. M., Botella, C., Alcan˜iz, M., Lian˜o, V., Guerrero, B., & Rey, B. (2004). Immersion and emotion: Their impact on the sense of presence. Cyberpsychology & Behavior, 7(6), 734–741. Brasset, A.-C., Gachet, J., Abiven, F., & Delarue, J. (2017). Taking control of product testing context thanks to a multi-sensory immersive room. A case study on alcohol-free beer. Abstract Book Pangborn. Bressoud, E. (2013). Testing FMCG innovations: Experimental real store versus virtual. Journal of Product & Brand Management, 22(4), 286–292. Bucher, T., van Der Horst, K., & Siegrist, M. (2012). The fake food buffet–a new method in nutrition behaviour research. British Journal of Nutrition, 107(10), 1553–1560. Burke, R. R. (2018). Virtual reality for marketing research. In Innovative research methodologies in management (pp. 63–82). Springer. Bushong, B., King, L. M., Camerer, C. F., & Rangel, A. (2010). Pavlovian processes in consumer choice: The physical presence of a good increases willingness-to-pay. American Economic Review, 100(4), 1556–1571. Edwards, J. S., Hartwell, H. J., & Giboreau, A. (2016). Emotions studied in context: The role of the eating environment. In Emotion measurement (pp. 377–403): Elsevier. Felnhofer, A., Kothgassner, O. D., Schmidt, M., Heinzle, A.-K., Beutl, L., Hlavacs, H., et al. (2015). Is virtual reality emotionally arousing? Investigating five emotion inducing virtual park scenarios. International Journal of Human-Computer Studies, 82, 48–56. Ferrer-Garcia, M., Gutierrez-Maldonado, J., Treasure, J., & Vilalta-Abella, F. (2015). Craving for food in virtual reality scenarios in non-clinical sample: Analysis of its relationship with body mass index and eating disorder symptoms. European Eating Disorders Review, 23 (5), 371–378. Fox, J., Arena, D., & Bailenson, J. N. (2009). Virtual reality: A survival guide for the social scientist. Journal of Media Psychology: Theories, Methods, and Applications, 21(3), 95. Gibson, E. L. (2006). Emotional influences on food choice: Sensory, physiological and psychological pathways. Physiology and Behavior, 89(1), 53–61. Glassl, S., Lutsch, D., & Scharf, A. (2017). Bringing virtual reality/immersive technologies in the sensory lab. Abstract Book Pangborn. Gorini, A., Griez, E., Petrova, A., & Riva, G. (2010). Assessment of the emotional responses produced by exposure to real food, virtual food and photographs of food in patients affected by eating disorders. Annals of General Psychiatry, 9(30).

Virtual reality and immersive approaches to contextual food testing

337

Harrar, V., Toepel, U., Murray, M. M., & Spence, C. (2011). Food’s visually perceived fat content affects discrimination speed in an orthogonal spatial task. Experimental Brain Research, 214(3), 351–356. Hathaway, D., & Simons, C. T. (2017). The impact of multiple immersion levels on data quality and panelist engagement for the evaluation of cookies under a preparation-based scenario. Food Quality and Preference, 57, 114–125. Hill, A. J., Weaver, C. F., & Blundell, J. E. (1991). Food craving, dietary restraint and mood. Appetite, 17(3), 187–197. Jaeger, S. R., & Porcherot, C. (2017). Consumption context in consumer research: Methodological perspectives. Current Opinion in Food Science, 15, 30–37. Jerald, J. (2015). The VR book: Human-centered design for virtual reality. Morgan & Claypool. King, S. C. (2016). Emotions elicited by foods. In Emotion measurement (pp. 455–472): Elsevier. Ledoux, T., Nguyen, A. S., Bakos-Block, C., & Bordnick, P. (2013). Using virtual reality to study food cravings. Appetite, 71, 396–402. Liu, R., Hooker, N. H., Parasidis, E., & Simons, C. T. (2017). A natural experiment: Using immersive technologies to study the impact of “all-natural” labeling on perceived food quality, nutritional content, and liking. Journal of Food Science, 82 (3), 825–833. Makransky, G., Lilleholt, L., & Aaby, A. (2017). Development and validation of the multimodal presence scale for virtual reality environments: A confirmatory factor analysis and item response theory approach. Computers in Human Behavior, 72, 276–285. O’Brien, H. L., & Toms, E. G. (2010). The development and evaluation of a survey to measure user engagement. Journal of the American Society for Information Science and Technology, 61(1), 50–69. Pla-Sanjuanelo, J., Ferrer-Garcı´a, M., Gutierrez-Maldonado, J., Riva, G., Andreu-Gracia, A., & Dakanalis, A. (2015). Identifying specific cues and contexts related to bingeing behavior for the development of effective virtual environments. Appetite, 87, 81–89. Poelman, M., Poelman, M., Kroeze, W., Kroeze, W., Waterlander, W., & Waterlander, W. (2017). Food taxes and calories purchased in the virtual supermarket: A preliminary study. British Food Journal, 119(12), 2559–2570. Romero, C. A., Compton, M. T., Yang, Y., & Snow, J. C. (2018). The real deal: Willingness-topay and satiety expectations are greater for real foods versus their images. Cortex, 107, 78–91. Schienle, A., Sch€afer, A., Hermann, A., & Vaitl, D. (2009). Binge-eating disorder: Reward sensitivity and brain activation to images of food. Biological Psychiatry, 65(8), 654–661. Schnack, A., Wright, M. J., & Holdershaw, J. L. (2018). Immersive virtual reality technology in a three-dimensional virtual simulated store: Investigating telepresence and usability. Food Research International. https://doi.org/10.1016/j.foodres.2018.01.028. Siegrist M., Ung C.-Y., Zank M., Marinello M., Kunz A., Hartmann C., et al., Consumers’ food selection behaviors in three-dimensional (3D) virtual reality, Food Research International (in press) 10.1016/j.foodres.2018.02.033 Ung, C.-Y., Menozzi, M., Hartmann, C., & Siegrist, M. (2018). Innovations in consumer research: The virtual food buffet. Food Quality and Preference, 63, 12–17. van Herpen, E., van den Broek, E., van Trijp, H. C. M., & Yu, T. (2016). Can a virtual supermarket bring realism into the lab? Comparing shopping behavior using virtual and pictorial store representations to behavior in a physical store. Appetite, 107, 196–207. Verhulst, A., Normand, J.-M., Lombard, C., & Moreau, G. (2017). A study on the use of an immersive virtual reality store to investigate consumer perceptions and purchase behavior

338

Context

toward non-standard fruits and vegetables. In: Paper presented at the virtual reality (VR), 2017, IEEE. Visschers, V. H. M., Hess, R., & Siegrist, M. (2010). Health motivation and product design determine consumers’ visual attention to nutrition information on food products. Public Health Nutrition, 13, 1099–1106. Wansink, B., & Sobal, J. (2007). Mindless eating: The 200 daily food decisions we overlook. Environment and Behavior, 39(1), 106–123. Waterlander, W. E., Jiang, Y., Steenhuis, I. H., & Ni Mhurchu, C. (2015). Using a 3D virtual supermarket to measure food purchase behavior: A validation study. Journal of Medical Research, 17(4), e107. Waterlander, W. E., Steenhuis, I. H., de Boer, M. R., Schuit, A. J., & Seidell, J. C. (2013). Effects of different discount levels on healthy products coupled with a healthy choice label, special offer label or both: Results from a web-based supermarket experiment. International Journal of Behavioral Nutrition and Physical Activity, 10(1), 59. Witmer, B. G., & Singer, M. J. (1998). Measuring presence in virtual environments: A presence questionnaire. Presence, 7(3), 225–240.

Section C Testing products in context

This page intentionally left blank

Healthcare supplements in context

17

Carla Lynn Kuesten Consumer Product Research, Amway, Ada, MI, United States

17.1

Introduction

In-context research is essential to understand, influence, and achieve success in consumer choice and health-related behaviors for health care supplements. Improvements in diet, including nutrient supplementation, given the potential for positive health outcomes, should be adopted, but can be accompanied by conflicting sensations— pleasure and satisfaction of having done something good for oneself, but also feelings of discouragement and despair if one fails to comply with healthy dietary changes, or expectations are unfulfilled. Consumers’ quality of life (QoL) and mindset toward health, wellness, and well-being plays a mitigating role in determining whether or not supplements are used, and can affect supplement choices. As QoL and related assessment tools are relevant for in-context research for the understanding of supplement usage, a brief historical review of these instruments follows. The WHOQOL-BREF is a sound, cross-culturally valid assessment of QoL, as reflected by its four domains: physical, psychological, social, and environment, and has good to excellent psychometric properties of reliability, and performs well in preliminary tests of validity (Skevington, Lotfy, & O’Connell, 2004). Sch€ unemann et al. (2010) address the need for validated tools to assess the relation between nutrition and QoL by development of an instrument for measuring QoL related to nutrition (and potentially compliance), which is of particular importance in the context of the development of new food/supplement products to improve health and well-being. Diener et al. (2009) presented a new measure of well-being to assess Psychological Well-Being (PWB), since renamed as the Flourishing Scale, a short 8-item survey of the person’s self-perceived functioning. Otto, Howerter, Bell, and Jackson (2010) explore whole person wellness with an integrative well-being and psychological flourishing measure, distinguishing flourishers from languishers and the ability to track change with treatment over time. To enhance QoL, Rigby, Ronchi, Graham, and Ronchi (2013) calls out the need for self-managed health and the use of information and communication technologies (ICT’s), providing the informed consumer with: better information for choice, new knowledge sources, healthy lifestyle coaching, and new smart models of care from personalization to ubiquitous care. This chapter examines context and context effects from the perspective of interaction of consumers (attitudes, experiences, memories, knowledge, habits/practices, expectations, and emotions) with supplement products (features, benefits, as well as possible side-effects) and the surroundings/environments in which they are Context. https://doi.org/10.1016/B978-0-12-814495-4.00017-9 Copyright © 2019 Elsevier Inc. All rights reserved.

342

Context

experienced or consumed. Given the challenge and complexity of formulating supplements with functional ingredients that often impart negative sensory characteristics, it is concluded that the necessary product development optimization and sensory testing is best carried out in controlled test settings with trained panelists (using protocols that reflect or emulate consumer consumption). Ultimately, however, context specific, real-life situations are required to accurately assess consumer usage and reactions to these functional products (psychological, physiological, and impact on QoL). The influence and importance of various contexts needs further study.

17.2

Oral nutritional supplements

The use of dietary supplements and functional foods is increasing globally. The vitamin and mineral market is over 13 billion USD in the US, with projected growth through 2020 in most global markets; the strongest markets based on market size worldwide include the US, China, Italy, and India (Mintel, 2018). According to the Dietary Supplement, Health and Education Act (DSHEA), dietary supplements are products intended to supplement the diet that bear or contains one or more of the following dietary ingredients: a vitamin, a mineral, an herb or other botanical, an amino acid, a dietary substance for use by man to supplement the diet by increasing the total daily intake, or a concentrate, metabolite, constituent, extract, or combinations of these ingredients (Zeisel, 1999). It may be taken in a variety of forms including: pill, capsule, tablet, or liquid. It is not represented for use as a conventional food or as the sole item of a meal or diet. It is labeled as a “dietary supplement” (FDA, 2017). From an applied perspective, furthering knowledge about users (and nonusers), why they use dietary supplements (or not), and the determinants of usage is warranted to aid future decisions by government policy makers, industry, and the consumers themselves. In-context research is necessary to understand and drive purchase and use behaviors for supplements. Psychographic, cognitive, emotional, core values, behavioral research and consequent information strategies and motivations play a strong role in choicebased decision-making surrounding nutrition supplements (Bornkessel, Br€oring, Omta, & van Trijp, 2014; Goetzke, Nitzko, & Spiller, 2014; Oakes et al., 2005; Yap, Noor, Marshall, & Liew, 2014). Studies and systematic reviews across the globe for different situations and contexts have surfaced over the years that capture attitudes, beliefs, emotions, and acceptance toward supplements (Egan, Hodgkins, Shepherd, Timotijevic, & Raats, 2011; Harris et al., 2011; Siegrist, Shi, Giusto, & Hartmann, 2015; Pajor, Oenema, Eggers, & de Vries, 2017). Nutrition.gov, a USDA-sponsored website, offers resources and credible information to help individuals make healthful supplement choices (https://www.nutrition.gov/dietary-supplements/questions-to-askbefore-taking-vitamins-and-mineral-supplements).

17.3

Initiation, adherence/compliance, and persistence

Efforts to increase behavioral use of supplements is of keen interest to marketers, as this leads to improved and sustained sales. Initiation (decision to begin supplementation), adherence/compliance (following dosage and usage as recommended or

Healthcare supplements in context

343

instructed), and persistence (ongoing decision to continue with a supplementation regimen) are critical challenges for dietary supplement programs. We draw from the medical compliance literature for learnings that can be applied herein, as the literature for supplements in this area, while growing, is still quite small. In medicine, compliance (also adherence, capacitance) describes the degree to which a patient correctly follows medical advice, in other words, “doctor’s orders”; the medical community then pursuing patient compliance research to assess aspects of compliance (or lack thereof— noncompliance or compliance resistance) and associated strengths and weaknesses of the various methods and measurements of compliance. Adherence is likely a better term to use in relation to supplement regimes, implying more personal choice and responsibility. Glanz (1980) discussed the magnitude of compliance, problems in measuring dietary compliance, and compliance determinants and recommended development of sound measures of dietary compliance, better documentation of diet adherence, and application of research with attention to sociobehavioral correlates of compliance and adoption; advocating testing of intervention strategies as crucial for improvements in health behaviors. Hill and Roberts (2011) discuss the role of adherence in the relationship between conscientiousness and perceived health. Hubbard, Elia, Holdoway, and Stratton (2012) provides a systematic review examining compliance with oral nutrition supplements (ONS) and the influence of patient and ONS-related factors. Influences on initiation and adherence were examined within the Health Belief Model framework by Nechitilo et al. (2016); results showing primary barriers to initiation of supplement use were low perceptions of severity of nutrient deficiencies and personal susceptibility—namely, low knowledge and awareness around deficiencies (varying by nutrient). Numerous other studies can be cited suggesting low compliance for nutrition dietary recommendations and the call for continued efforts to improve compliance for health. Thus, given the known challenges of instilling healthy attitudes and behaviors for health care supplement usage; educational, foundational, and tactical in-context research is mandatory for a better understanding, and for influencing consumer health care supplement behaviors.

17.4

Product factors affecting consumer usage of nutritional supplements

Product factors must be considered and weighed against variable and disparate consumer reactions/responses to them; therefore, the product factors discussed next are taken into consideration with consumer sentiments and behavioral responses in mind. It is considered a given that all the factors highlighted from the literature that are discussed as follows would be impacted by the research context and manner in which the research was conducted. Ares et al. (2016) indicate consumers’ perception of well-being in a food context can affect food choices, and might provide a more holistic evaluation of products than overall liking or healthfulness scores; consumers regarded sensory characteristics, manufacturing processes, nutritional composition, and context of food consumption

344

Context

as the main factors underlying food-related well-being in their study. Lee and Kim (2009) indicate that dietary supplement use might be affected by certain sociodemographic, lifestyle, and health characteristics; and Foote et al. (2003) indicate it may be difficult to separate the effects of supplement use from other lifestyle factors when studying disease etiology. O’Connor and White (2010) examine the role of attitudes, subjective norms, and risk dread and familiarity in the context of vitamin supplements and functional foods among nonusers. Monotony/boredom are ascribed to some compliance issues; time-preference measurement allows a research context in which to optimize a food in order to maintain acceptance but reduce boredom; this is especially relevant under extended use conditions (Moskowitz, 2000). The sensory characteristics of ONS are often blamed for poor compliance, varying by category (Darmon, Karsegard, Nardo, Dupertuis, & Pichard, 2008; Methven et al., 2010). Side effects also contribute to reduced compliance. As Sch€ unemann et al. (2010) point out, the feasibility and success of dietary changes will depend, at least partly, on whether potential negative influences on QoL can be avoided, especially in the context of the development of new food products to improve health and well-being. Consumer confidence in dietary supplement products also impacts compliance; consumers deserve and demand that products meet standards for safety and quality, and want assurance that products are safe from a standpoint of identity, purity, quality, strength, and composition, and that the ingredients used are safe and effective. Fjeldsoe, Marshall, and Miller (2009) examined mobile telephone short-message service (SMS) for delivering health behavior change interventions. Text messages showing positive behavior change outcomes were observed with intervention initiation (researcher or participant), SMS dialogue initiation, tailoring of SMS content, and interactivity were found to be important features of SMS-delivered interventions. Marschollek et al. (2012) demonstrate the benefit of wearable sensor technologies and provide approaches for the implementation of sensor-enhanced health information systems for wearable sensors as an integral part of future pervasive, ubiquitous, and person-centered health care. Products that help consumers form implementation intentions may assist with compliance (Zandstra, den Hoed, van der Meer, & van der Maas, 2010). Connor, Rafter, and Rodgers (2005) provide evidence suggesting that fixed-dose combination pills and unit-of-use packaging are likely to improve adherence in a range of settings, but admit the size of benefits remains unknown. Important aspects of products to keep in mind are labels and communication. The US FDA Nutrition Labeling and Education Act (NLEA) law established mandatory nutrition labeling for most foods, and placed restrictions on the use of food label claims characterizing the levels or health benefits of nutrients in foods; Rowlands and Hoadley (2006) provide an overview of the FDA’s regulations and evidencebased method for evaluating health claims. McDonald and Nicholson (2006) tested how reading a pamphlet by the US Federal Drug Administration titled “Tips for the Savvy Supplement User” affects intention to use and recommend supplements. In addition to advertisement, quality and product attractiveness are noted by consumers as factors influencing dietary supplement purchase intent and consumption. The media (books and magazines) are powerful influences on a person’s decision

Healthcare supplements in context

345

to use supplements; reasons for consuming dietary supplements are often complex, combining social, psychological, knowledge, and economic factors. In addition to labeling, persons involved—physicians, dieticians, coaches, or even friends and family can influence incidence, type, and reasons for supplement use. Lacerda, Carvalho, Hortegal, Cabral, and Veloso (2015) conclude the use of supplements lacks proper professional guidance, which raises the need for monitoring those who work out at a gym. In light of media impact on behaviors, Peters, Shelton, and Nelson (2008) report that advice of a physician underlies most consumer behavior, that consumers take dietary supplements to improve their physical health and gain peace of mind. Beliefs underlying dietary supplement use reveal differences between supplement users and nonusers, highlighting the potential of the theory of planned behavior in exploring supplement-taking behavior and an individual’s motivations to use dietary supplements. Dickinson, Blatman, El-Dash, and Franco (2014) report reasons most often cited for supplement use were for overall health and wellness (58%) and to fill nutrient gaps in the diet (42%). Supplement users were significantly more likely than nonusers to say that they try to eat a balanced diet, visit their doctor regularly, get a good night’s sleep, exercise regularly, and maintain a healthy weight. Detailed results from the 2011 survey confirm that supplement use increases with age, and is higher in women than in men. Supplement ingredients drive usage among supplement users; multivitamins were the most commonly used supplement (71%), followed by omega-3 or fish oil (33%), calcium (32%), vitamin D (32%), and vitamin C (32%). The US FDA Center for Food Safety and Applied Nutrition conducts or supports research for dietary supplements; reports covering consumer surveys, trends, and patterns of nutrient intake can be found at the FDA website: https://www.fda.gov/ Food/FoodScienceResearch/ConsumerBehaviorResearch/ucm188571.htm. Mintel provides up-to-date comparative product usage and frequency reports; the latest being their Vitamins, Minerals and Supplements US Report, (Mintel, 2017). This recent report indicates, despite almost half respondents agreeing VMS are part of a healthy lifestyle, compliance associated behaviors such as forgetfulness and pill fatigue are driving occasional or lapsed use of VMS, citing consumers, rather than just “taking a leap of faith,” want proof that they work, follow-through on claims, highlight of quality ingredients, and information on how they are absorbed to convey efficacy. In order to stay competitive in an ever-changing marketplace, Mintel suggests value incentives to motivate users, or dosage reminders on a wearable health monitor.

17.5

Choice of research design, methodology, and protocol

Selection and design of test methodologies are critical to the successful execution and outcome of any research, and in particular for our topic of discussion—health care supplements (in context)—where, as we have learned from the prior work mentioned herein, consumer health attitudes, values, motivations, and behavior is fundamental to achieving trial, adoption, and consistent (persistent) use; without adherence, treatments can’t be proven effective. Considering new product development, the research

346

Context

questions and objectives must be clearly articulated, target populations/segments defined, and adequately represented (appropriate screening, recruitment and of a sufficient base size), experimental design (i.e., monadic vs. sequential-monadic), variables of interest identified and defined, and number of products to test at a time specified—for accurate and robust measurement of consumer, product, and context effects. Understanding the demographic and psychographic profile of users (vs. nonusers) to target for the product is key. Foundational research to first understand consumer needs (unmet, unarticulated) helps guide technology planning; development efforts can then be focused on meeting those needs. Research during development is generally conducted in stages (or waves), starting in the lab with early prototypes, and progressing through small-scale product cuttings, employee testing, various small research tests with consumers (online, in a central location test facility, virtual or simulated context environments, and/or in-home use tests) for guidance, then finally with larger, perhaps even global-scale testing with consumers within the context of interest for optimization, and usually, last, under normal usage conditions (real-world settings) final product confirmatory testing. Control over the test scenario typically decays through this sequence of testing, as the tests reach past the more controlled conditions to noisy “real-world” settings. Initial controlled experiments conducted with sensory panels for profiling should be conducted on prototypes that vary systematically and are of sufficient difference in the variables of interest to be perceivable. The panel should be monitored for their ability to perform, and replications used if the product is variable. A variety of qualitative, quantitative, and mixed-model research techniques are available to address the research objectives throughout the development path from multiple angles—often with both divergence and convergence of thought along the way. During consumer guidance testing work (CLT’s and HUT’s scenarios), Boutrolle, Arranz, Rogeaux, and Delarue (2005) show hedonic differences can be obtained under these different test scenarios, cautioning with the need for a common comparison criterion (robustness) for comparison across studies. While similar results are comforting and validating, one must be prepared to explain dissimilar results along the way or, ideally, advantage all test scenarios synergistically. It’s true that each specific supplement form will introduce unqiue and specific challenges to consider for testing. While it’s impossible to call out all possible test effects here, two worth mentioning relevant to supplement testing are fatigue and/or carryover effects and long-term range effects; the implications of any of these can be significant. For example, Thomas, van der Stelt, Schlich, and Lawlor (2017) in a study using temporal drivers of liking for oral nutritional supplements for older adults confirmed that drinking an ONS increased thirst during tasting.

17.6

Industry case study applications of health care supplements product research

Three different supplement product category case studies are provided in brief to highlight different stages and contexts of research, starting with controlled laboratory and/ or panel testing through structured and managed Central Location Testing (CLT), and

Healthcare supplements in context

347

finally Home Use Testing (HUT), where the product is consumed in the real-world environment. The fourth, and last, case study is observational in nature. Verification that the type and magnitude of product differences are perceivable, measurable, and as planned/formulated in terms of the desired or intended effects, is an important first step to ensure further testing with consumers is meaningful. CLTs are often used next to optimize and confirm consumer reactions to test prototypes prior to more resourceintensive HUTs, in which the consumer ultimately consumes the product, but this environment is less controlled, ‘noisy,’ and where the consumer can be more prone to distractions, giving less attention to the product. For some product categories, in this particular case, health care supplements, the CLT may not be appropriate for trial/evaluation of multiple prototypes in a single test session, as can be conducted for other consumable products, such as foods and beverages; the controlled laboratory protocol may not be suited for ingestion to achieve physiological responses, or may need to be modified to avoid over-dosage. This is due to the content and nature of the ingredients. Thus, for safety reasons, dosage levels of the ingredients contained in supplements can be a limiting factor, and needs to be managed over multiple test sessions or protocols, adjusted such that safety is taken into account. Each case study presented herein begins with an introduction providing background for the research, and covers objective(s), methodology, results, and conclusions over the various stages of research used for each. Different types of research are applied and represented within the contexts of controlled laboratory and descriptive panel tests, CLTs and HUTs—each staged with intentional protocols, research techniques, and situations to gather product understanding and consumer insights.

17.6.1 Case study 1: Vitamin mineral phytonutrient supplements 17.6.1.1 Introduction This, the most extensive case study, addresses research in various contexts for global harmonization and relaunch of a 3-tablet vitamin mineral phytonutrient supplement product—starting in the sensory laboratory, and followed with consumer immersion, an online community, central location, and home-use testing. Each segment of research provides perceptual or behavioral understanding toward a better consumer experience for a twice-daily regime. Highlights of these studies that advance different research contexts are shared as follows; the preliminary panel work and smaller-scale studies are justified and leveraged to inform later, larger-scale, and more resourceintensive research.

17.6.1.2 Objective(s) Objectives vary across each study segment, as the work evolves toward final in-market formulations. In support of new base materials, the first objective was to characterize the odor profile of the current versus the new base–alone, and with various added odorants to mask some of the offending odors with familiar, pleasant odors congruent with phytonutrients. Next, consumer immersion was applied toward consumption improvements; the objective was to understand the key relationships between tablet variables such as

348

Context

shape, size, thickness, and type of coating, and the impact on Ease of Swallowing to guide development. Separately, an online community with current and lapsed users was used to investigate factors to improve adherence (3 tablets to be taken twice daily). The final studies involved descriptive panel, as well as consumer in-market home-use testing of the current versus new tablets. The objective of the descriptive panel was to profile the flavor aromatics and taste of the tablets to document sensory profile changes, as well as confirm odor type and intensities for relating to consumer hedonic response ratings. The objectives of the consumer 1-week home-use tests were to collect quantitative consumer reactions, emotions, and psychographics based on supplement usage in a real-life consumption context and support product claims.

17.6.1.3 Methodology Supplement dry powder aromas were profiled (Dry Aroma and Off-Notes) by a sensory descriptive panel, and consumers (US, Japan, Korea, and Germany; users and nonusers) in a 1-h consumer CLT (Kuesten, Chopra, Bi, & Meiselman, 2013). Both 10 or 20-item Positive Affect Negative Affect Scale (PANAS) (Thompson, 2007; Watson, Clark, & Tellegen, 1988) scales were applied. See Fig. 17.1.

Fig. 17.1 Global consumer aroma study evaluation protocol.

Healthcare supplements in context

349

Next, consumer immersion work followed a step-wise approach utilizing two distinctive methodologies: (1) Tragon QDA, a descriptive sensory test method, and (2) consumer assessment through a central location test in four key international markets; Japan, China, Germany, and the US. Sensory attributes and DOE variables impacting Ease of Swallowing were first identified, then all tablets were assessed on those attributes. Twenty-five tablets representing oval, round, and diamond shapes were evaluated by the Tragon QDA panel. An efficient, designed subset of variables (shape, size, thickness, and coating), sixteen products in total, were selected and tested. The Market Research Online Community (MROC) was conducted with employees of the company, all users (but only 1 time per day users, i.e., not fully compliant with the package instructions of 2 times per day), or lapsed users of the supplement (those who had stopped taking it). Fig. 17.2 illustrates the 3-month study, outlined by monthly activities. The descriptive panel then profiled the same tablets (current vs. new) as tested in the consumer HUTs. The protocol was designed to simulate the aroma and flavor experienced by consumers during consumption. All HUT studies (Japan, Germany, and the US) involved gathering baseline information on current supplement usage, a 1-week placement for each test sample, tablet consumption diaries, and sensory and consumption experiences—specifically Digestion Comfort-related attributes,

Fig. 17.2 Market Research On-Line Community (MROC) methodology.

350

Context

as well as demographics, psychographics, and emotions—Positive Affect Negative Affect Scale (PANAS) and Profile of Mood States (POMS). The POMS scale was applied (Heuchert & McNair, 2012).

17.6.1.4 Results Phytonutrient tablet aroma studies Aroma of the current tablet samples varied, and was distinguishable in intensity and character. Consumer aroma study results were consistent with descriptive panel results, which showed samples varied in intensity, aroma character, acceptability, and PANAS, globally. This study supports that the PANAS scales are reliable and valid in measurement of consumer emotions evoked by the aromas of phytonutrient supplements. Results suggest that the hedonic, sensory, and emotional attributes represent different dimensions in consumer choice and consumption behaviors. Important psychographic covariates were identified for Overall Liking of Aroma. Further details of this work are available (Kuesten et al., 2013; Kuesten, Dang, Nakagawa, Bi, & Meiselman, 2016).

Consumer immersion—Factors influencing supplement usage Based on the design, shape showed no significant effect; however, all other variables were significant (thickness, coating, diameter level, and coating level). Principal Component Analysis was used to illustrate a two-dimensional view of the product differences/similarities; Factor 1 (Flavor, Coating and Smoothness) and Factor 2 (Easy to Swallow, Stuck in Throat, and Mass). Products that offered unique experiences were prioritized and selected for consumer testing (Fig. 17.3). Results of the consumer acceptance testing showed the best-liked tablets represented all shapes (round, oval, and diamond), and both types of coating. The least-liked tablets had low levels of coating, thickness, and the largest diameters for round and oval shapes. Results were similar and highly correlated across all countries. The swallowability regression model to predict Ease of Swallowing proved successful for later development purposes.

Consumer market research online community—Adherence The MROC study was used to explore supplement attitudes and behaviors with consumers. Using this research context, we were able to easily explore and discuss many topics revolving around supplement usage with consumers. We learned consumers consider taking vitamins/supplements as part of a daily routine—reportedly without much thought or emotion, though they are happy or pleased with themselves for having taken it. One of the exercises was to create collages of how supplements made them feel—physically, psychologically, and emotionally; males and females differed in the emotions expressed. New packaging was also explored through interactive package images toward making the packaging easier to use, and less annoying and frustrating.

Healthcare supplements in context

351

Thick Ap

Round tablets: 1–8

8

15

Stuck in throat Mf Stuck in throat At Mass of tablet Mf Edges Ap

25

Oval tablets: 9–17

5

Diamond tablets: 18–24

Size Ap

2 Handfeel Ap

19 Green tea Fl

13

11 7 1

17 Coating Mf

Factor 2 (23.20%)

Speckled Ap

24 10

Smooth Mf

14 21 4

Tacky Mf Chalky Mf Lingering At Green tea Ar Green tea At Mineral At Seaweed Fl Seaweed Ar Grass Fl Grassy Ar Mineral Fl Grass At

22

12 18

Dissolve rate Mf

Pearly Ap Green Ap

Shiny Ap

16 9

6

3

20 23

Easy to swallow Mf

Factor 1 (36.08%)

Fig. 17.3 Consumer immersion—PCA of tablets of varying shape and size selected for consumer testing, Tragon QDA.

Internal sensory product profiling and global consumer home use testing Descriptive panel profiling of the supplement tablets preceded consumer HUTs, ensuring that the tablets delivered the intended sensory effects. Profiles for the new tablets were generally lighter in aroma and flavor in an effort to improve the consumption sensory experience. Consumers noticed the sensory effects detailed by the descriptive panel and prior immersive research guidance taken from the MROC study, that is, improved aroma profiles and tablet factors that delivered higher key measure ratings (Overall Liking and Overall Digestive Comfort). See Kuesten et al. (2016) and Kuesten, Bi, and Meiselman (2017) for further details.

17.6.1.5 Conclusions Staging research with intentional development based on consumer needs and in consideration of consumer choices using a variety of research contexts (sensory descriptive panel, consumer immersion, MROC, CLT, and HUT), in combination with

352

Context

advanced analytics, as shown in this case study, is a highly effective and efficient means to achieving validated success in the marketplace. It is important to drive reformulation efforts against consumer needs. While controlled internal testing is key to verifying selected design variables based on consumer needs, and likely perceptible to consumers, in-context product testing with the targeted consumer is critical in reaching optimized solutions. And, simply gathering consumer reactions to products lacks depth, without the sound understanding of the consumer who is rating those products. Here, in this case study, we have shown how psychographic and emotion research helps to uncover the complexity of consumer attitudes, beliefs, and how consumer dispositions may influence responses and behaviors toward products.

17.6.2 Case study 2: Protein powder 17.6.2.1 Introduction All-Plant Protein Powder (APP) is a protein product used as a supplement to boost dietary intake of protein. It’s a flagship product for the company, high volume, and a leading product in the marketplace. Manufacturing, quality assurance, procurement, and development concerns are about maintaining delivery of a consistent, high quality product experience; which is a challenge, given these natural ingredients (soy, wheat, and pea) do vary and differ across suppliers. Overly stringent quality control measures can be costly. This case study takes an applied approach using different research contexts—lab instrumental, employee discrimination, and descriptive panel assessments, as well as evaluation with external consumers, to manage and mitigate the risks associated with maintaining a quality protein powder to the globe. In addition to quality maintenance, Product Development took on the initiative of trying to improve the known messy preparation experience of APP for consumers by increasing the rate of powder dissolution, as well as improving the texture and mouthfeel of the finished beverage.

17.6.2.2 Objective(s) The various studies covered in this case study hold different objectives and involve different research contexts. We begin with internal employee discrimination testing to assess similarity across different suppliers and process variables. This is followed by Sureness Difference and Overall Liking ratings by targeted consumers (CLT). Given mouthfeel (Smoothness) was determined important to consumers, we dive deeper with analytical instruments to assess particle size and correlate those with descriptive panel ratings (Viscosity, Powdery Mouthfeel, and Residual Particles). Next, we explore parameters impacting preparation Time to Dissolve by using a controlled, standardized internal panel protocol and attribute ratings versus external consumer perceptions and behaviors; contrasting controlled laboratory, versus both consumer CLT and HUT preparation and consumption results.

Healthcare supplements in context

353

17.6.2.3 Methodology Test samples to assess supplier similarity and substitutability to current APP were prepared. Panels were used in the following sequence: (1) A discrimination panel was used to test differences (2) “Close-to-target” samples were evaluated by a trained descriptive panel, and (3) In-market testing was conducted for consumer validation. The A-Not A Reminded (A-Not AR) and A-Not A scales were used with the employee discrimination panel and consumers, respectively, to measure similarity/difference versus control. See Bi, Lee, and O’Mahony (2013), Bi, O’Mahony, and Lee (2013) and Bi (2005) with application of Benchmark Dose (BMD) methodology (see, e.g., Bi, 2010) and the Benchmark Dose Modeling Software (BMDS) (US EPA, 2013) for determining appropriate levels. Relative importance (RI) analyses were conducted for consumer attributes to Overall Liking and Sureness Difference. See Bi (2012). Further studies aimed at improving and assessing Time to Dissolve and other preparation and consumption aspects were pursued, testing the current APP against alternative treatments with different degrees of agglomeration (preprocessing step to reduce dissolve time). To this end, the analytical group measured particle distribution and size and the descriptive panel measured the time to dissolution, as well as the impact of agglomeration on the flavor and texture of various formulations. Japanese consumers then tested 3 selected samples ranging in dissolution time under both CLT (2 different ways—hand-mixing/spoon stirring versus protein shaker) and HUT conditions (per their usual routine).

17.6.2.4 Results Similarity testing with employees—Protein powder consistency and quality Results from the employee discrimination panel showed differences in similarity to the current control APP across suppliers. See Kuesten and Bi (2015) for further details of this work. Descriptive sensory profiling differences could be traced back and attributed to flavor and texture characteristics inherent in the ingredients. Results showed all samples selected and tested with consumers were “significantly similar” within a similarity limit. Based on RI analysis, the drivers vary by key measure; consumers differentially distinguish Sureness Difference based on texture and Overall Liking based on flavor and sweetness. The main objective of this study was satisfied with determination of appropriate alternative suppliers’ ingredient options via this multi-phased approach.

Controlled versus end-user environments—Characterization, preparation, and consumption Instrumental and descriptive panel methods demonstrated sample differences. The average particle size in the control and 2 test samples varied, ranging from 41 to 103 μm. Descriptive panel results indicated the control had less Powdery Mouthfeel and Residual Particles than both Test 1 and 2 agglomerated samples. Significant correlations of instrumental and sensory measures were evident. The panel Time to Dissolve (secs) with a standardized manual spoon-stir method shows the control with the

354

Context

highest quality flavor is slowest to dissolve (60 s), while Test 2 falls mid-range, and Test 1 is fastest, but also the most bitter sample. So, agglomeration, while it improves rate of dissolution, impacts texture and mouthfeel deleteriously. The next questions became, do consumers notice a difference? And, if so, does it matter to them? Is the decrease in dissolution time worth the trade-off of decreased sensory flavor, texture, or mouthfeel quality? To address these questions, the CLT and HUT tests described herein were executed. In fact, the answer regarding Time to Dissolve is, “it depends” on which scenario in which the samples are tested (CLT forcing the consumers to spoon-stir in a similar manner as the panel standard mixing protocol, yes; HUT usual prep method of using a shaker, no). See Fig. 17.4. And, in fact, these Japanese consumers indicated any advantage in dissolution time, even if noticed, was not worth the sacrifice of taste. Contrary to the descriptive panel findings, no consistent results were observed in Thickness and no salient differences were observed in Smoothness by these consumers. Thus, even though advantages in dissolution time were verifiable in the lab setting, and could be re-created with consumer instructions in the CLT, the significant detriment in mouthfeel and texture observed in the lab were not corroborated by the consumer in a real-life context, perhaps due to lack of attention to those product characteristics under normal use conditions. Their existing behaviors in mixing protein and familiar

Mixing

How fast to dissolve CLT shaker current and fast were found to dissolve significantly faster than medium. While there were no salient differences in the HUT results, current had a directional advantage. For CLT spoon stir, samples performed markedly better in a consi stent order. CLT shaker

(%)

CLT spoon stir

Medium sample

Fast sample

Current APP

Medium sample

Fast sample

Monday

n = 77 A

n = 77 B

n = 77 C

n = 66 D 23 5 3 12

n = 67 E 2 5 8

n = 66 F

n = 120 G

n = 117 H

n = 119 I

21

23

21

18

21

21

26

24

23

24

22

9 Very fast 8

46

7

39

22

52 23

6 5 Neither fast nor slow

17

24

20 17

18

2

9 7 31

8 3 8 1 31

9 5 3

29

7.73

7.38

7.86

BDEFGHIJ

DEFIj

BDEFGHIJ

80.5

76.6

83.1

DEFGIJ

DEFg

DEFGhIJ

1.3

5.2

2.6

Slow (1–3)

15

25

17

3

Fast (7–9)

Current APP

Monday

Friday

Medium sample n = 119 J

21 14

Ave.

Friday

8

21

4

1 Very slow

H UT

Current APP

17

16

9

26

13 9 13

13

10

8 5 53

13

10 5

9 5 3

8 1 1

11 6 2

2.70

4.49

6.62

6.96

7.10

D

DE

DE

DEf

DE

DE

4.5

13.4

57.6

64.2

71.8

66.4

65.5

D

DE

DE

DEfg

DE

DE

75.8

31.3

7.6

1.7

1.7

7.6

5.0

ABCEFGHIJ

ABCFGHIJ

aGH

6.83

7 41 6.88

aGH

*ABCDEFGHIJ/abcdefghij: 95%/90% significance level

How FAST would you say the protein powder dissolved?

Fig. 17.4 CLT and HUT comparison of perceived rate protein samples dissolved. Research conducted and chart provided by IPSOS Japan.

Healthcare supplements in context

355

taste/texture experiences and expectations were still the deciding factors for them. Their usual method of mixing protein at home with a shaker bottle diminished any observable advantages in dissolution time for the agglomerated products in the HUT data. This in-context CLT and HUT research with the consumer saved the company from significant investment in processing to agglomerate protein and at a wouldbe disadvantage in sensory taste quality.

17.6.2.5 Conclusions Supporting a product category is multi-faceted, and requires research from multiple angles for decision-making, or else critical markets, targeted consumers, or dimensions of the product may be sacrificed for the sake of others of lesser importance. This case study has demonstrated the necessary use of in-context research to assist in resolving choices on behalf of the consumer. Here we have shared the relative sensitivity of our internal consumers (lab instrumental, employee discrimination, and descriptive panels) against that of our external consumers (both in a controlled CLT test environment and in an in-home use situation), proving the importance of in-context consumer research for decision-making.

17.6.3 Case study 3: Omega-3 fish oil soft-gel capsules 17.6.3.1 Introduction Fish oil and its products are high in polyunsaturated fatty acids, providing various health benefits such as anti-inflammatory and immune modulation for cognitive, heart, and brain health; but usage may be limited and/or avoided by some, due to a fishy aroma and discomfort during digestion. A method to continuously monitor the immediate and delayed sensory and digestive responses associated with consumption of different fish oil treatments in the context of normal daily use is needed.

17.6.3.2 Objective(s) The main objectives of this research were: (1) to establish a reliable consumer research methodology for measuring digestive comfort and sensory side effects after consumption of fish oil omega-3 softgels, and (2) to understand the effectiveness of an innovative formulation to help improve the sensory consumption experience for consumers.

17.6.3.3 Methodology First, a sensory descriptive panel characterized the flavor profile of experimental treatments of the softgels to ensure aroma and flavor formulations were established and met sensory expectations. A 2
2 factorial design varying enteric coating type and coating percentage were profiled descriptively against the control. Next, a 3-week in-home use test was conducted. Three omega-3 softgel samples were tested: control (fish oil formulation), test (modified fish oil formulation with lemon oil), and placebo

356

Context

(soybean oil formulation). Each participant evaluated all three samples, one sample at a time, for 4 consecutive days followed by a 3-day break. The daily online survey included hedonic and digestive responses including Overall Digestive Comfort, Burp-Back, Burp Bothersomeness, Fishiness, and Amount of Burping—adapted from the BARF pediatric nausea scale (Baxter, Watcha, Baster, Leong, & Wyatt, 2011) by use occasion.

17.6.3.4 Results Sensory and consumption experiences—Minimizing Fishy Burp-back The sensory descriptive panel resolved sample differences in the coating variables, and was used to make the selection of softgel formulation in consumer testing. As expected from the panel results, consumers were able to detect differences between control, treatment, and placebo samples; aroma and flavor intensities were found to be significantly lower in the placebo sample over time, and corresponded to an increase in consumer liking. In regard to digestive side effects, the attributes all reflected similar information. Perceived sensory and digestive side effects were found to be different across and within the demographic groups (Fig. 17.5). Burp amount

5

Rating means

4 1.51a 3

1.29a 1.53a 0.68b

2 1.08b 1

0.60c

Total population

0 Control

Age >50 Treatment Placebo

5

Fishness intensity of burp

Rating means

4 1.95a 3

2.30a

1.41b 1.93a 0.66c

2 1 0.74b

Total population

0 Control

Female Treatment Placebo

Fig. 17.5 Product effects on gender and age subpopulations of sensory side effects. Significant at alpha ¼ 0.1.

Healthcare supplements in context

357

Participants 50 years old and older were able to distinguish between the control and treatment samples. Such a distinction was not found in the total tested population. Females were more sensitive than males to the intensity of fishy burps.

17.6.3.5 Conclusions This 2-phased sensory guided, consumer-based approach has proven beneficial in minimizing the negative sensory side effects and undesirable digestive discomfort for fish oil softgels. The descriptive sensory panel helped to resolve the coating to test with consumers, but is limited (per safety protocol) to a brief (noningested) flavor experience. Given the inability to test Burp-Back effectively with a descriptive panel, testing the final fish oil softgel formulations with consumers in the context in which they are regularly consumed was deemed a requirement to optimize the consumer consumption experience.

17.6.4 Case study 4: MIT PlaceLab study—Validity of dietary assessment surveys—Example of a study to influence health practice 17.6.4.1 Introduction Collaborative research was undertaken between TIAX, LLC, and the MIT PlaceLab in 2004 enabling studies in the context of everyday life (Intille et al., 2006). A major research focus for Placelab was on developing products and technologies that will impact consumer needs for independent living, including proactive health and wellness.

17.6.4.2 Objectives(s) MIT PlaceLab was used as a new tool for calibrating self-reporting instruments used in food-related studies, such as dietary intake. Objectives included characterization between standard self-assessments and data collected with PlaceLab, adjustment for biases, and establishment of a baseline for designing a food-related behavior change intervention. Our goal with this initial pilot study was to identify methodological considerations that would need to be addressed as part of conducting good research in this area.

17.6.4.3 Methodology Food intake measures collected from the survey instruments were compared against true intake, activities, and responses as observed on video. Methods included: Exploratory Sequential Data Analysis (ESDA)—to record and analyze video recordings and descriptive statistics, correlation analysis, and observational analysis with cognitive mapping tools—to elicit, represent, and define content of the video information capture. Two participants were recruited to reside and be monitored for 10 days at Placelab located in Cambridge, MA. The suite was wired with discrete, configurable sensor

358

Context

Fig. 17.6 Placelab residence. Photo credit: Kent Larson, Director of the Changing Places research group and co-Director of the City Science Initiative at the MIT Media Lab.

systems, which allowed comprehensive data to be collected, providing observation of environmental and contextual data. Participants were videotaped in the living areas (kitchen, dining, family rooms; but not bedroom or bathroom). See photo of residence, Fig. 17.6. The observational methodology involved analysis of videotapes, along with interviews of the participants to discover food behaviors and attitudes. Observational deliverables included a summary of findings with transcript and interview excerpts and videotape narration. ProCoder (http://www.procoderdv.com) and MOOSES (http://www.getmooses.com) software tools (The John F. Kennedy Center at Vanderbilt University) were used for data collection and analysis. Detailed examination of habits, practices, and attitudes concerning food/beverage consumption were captured. A participant and moderator reviewed tape fragments retrospectively to gain deeper understanding regarding mood state—normal or modified by a condition or event. Insights were gathered—including discovery of unanticipated ramifications or interactions that transpired throughout the day that impacted consumption behavior.

17.6.4.4 Results The summary of this pilot study noted problem areas, gaps (relative to other instruments), potential opportunities for intervention or strategies (where appropriate), and areas for new ideas. Some inconsistencies were observed between self-report and actual food behaviors.

Healthcare supplements in context

359

17.6.4.5 Conclusion PlaceLab offers the unique ability to assess human behavior and the impact of various technologies on behavior modification in an applied setting. This pilot study provided a multi-faceted body of knowledge about behaviors associated with food (and supplement) consumption that will advance the understanding of effective nutrition, health, and well-being intervention strategies. The benefits of the Placelab are to gain insights that are not possible or efficient with traditional methods, and observe actual behaviors–not self-reported and not in a lab.

17.7

Product story telling—Making sense of it all

Product storytelling for health care supplements takes the form of claims, as claims provide the basis for the product story and communicate the nutrition/health benefits to the consumer. It is incumbent upon the seller to ensure claims are legitimate, follow strict regulatory guidelines, and are interpreted and understood by the consumer. The research context in which claims are generated (sensory, consumer, or clinical research studies) may profoundly impact assessments and the manner in which the results may be used to inform consumers. As mentioned in previous sections, educated consumers expect and demand understanding. Agarwal, Hordvik, and Morar (2006) describe the different types of claims that can be made: (1) nutrient content claims, (2) structure/function claims, (3) health claims, (4) dietary guidance statements, (5) other claims, and (6) nutritional claims displays on packages. A systematic literature review of studies investigating the extent and the ways in which health claims influence consumers is presented by Pothoulaki and Chryssochoidis (2009); six thematic categories emerged, namely knowledge/awareness of dietary issues, effects of health claims on purchase decisions, effects of health claims on perception/attitudes/beliefs, sources of information and trust, framing of health claims, and disease-risk reduction and healthenhancing claims, as well as consumer purchase decision. Wu, Linn, Fu, and Sukoco (2012) examined the effects of endorsers, message framing, and rewards on consumers’ responses toward dietary supplement advertisements, indicating behavioral intentions could be influenced, and suggested results could be useful for marketers with regard to developing and implementing their marketing activities to specific customer segments. Global efforts to streamline regulations beg the question of what will be the exact framework, research context, and criteria to use when assessing consumer understanding; answers are still outstanding.

17.8

Practical considerations and future needs

The reality is that consumer behavior is hard to manage or change, especially, it seems, for the health care supplement product category. Cumulative evidence suggests clinical results have not been motivational enough to impact health behaviors broadly or significantly. For example, the USDA and other governmental agencies have recommended for some time that persons markedly increase their consumption

360

Context

of fruits and vegetables to reduce the risk of cancer and other diseases (USDA, 2010); and yet few US adults meet the MyPyramid recommendations for consumption of fruit and vegetable servings based on estimated 2-day average intakes (de Abreu et al., 2013; Murphy et al., 2012). The recent nutrition evolution has shifted to a more holistic focus aimed at coping with conditions of excess—calories, sedentary lifestyles, and stress, and recognize the complexities of nutritional, dietary, social, behavioral, and environmental factors (Shao et al., 2017). Ongoing studies investigating the contribution of supplements to Health Related Quality of Life (HRQoL) are warranted. Glanz (1980) warned, years ago, that it seems unreasonable to expect changes in long-lasting and complex patterns of behavior without intense and sustained time and effort. Koster (2009) advises us to rethink the methods used for sensory and consumer research—given the realization that much decision making occurs at the nonconscious level. Piqueras-Fiszman and Jaeger (2015) take a lead with exploration into in-context emotions research. In order to effect change, we need to put emphasis on the influence real-life contexts exert on motivations and consequent behaviors. Areas to further consider and explore are endless, but should include focus on: (1) social and family influencers, (2) contemporary lifestyles and nutrition, (3) modifiable regimens to effect favorable change, including productive nutrition interventions for maladaptive behaviors, (4) targeted decision strategies to improve informed decision making, and (5) emotional research (in context). Georgiou, Garssen, and Witkamp (2011) write to tell of the rising interaction between pharmacology and nutrition science, the synergy hopefully reflecting favorably on future changes on our lives. We anticipate personalized nutrition as the lines between the “Pharma-Nutrition Interface” blur; Eussen et al. (2011) even suggest health technology assessments should be used more to compare the cost-effectiveness and benefit-risk ratios of drugs, functional foods, and dietary supplements, and to evaluate the added value of functional foods or dietary supplements to drug therapy—leaving us with more to study.

17.9

Summary

This chapter (with research context in mind) has covered various aspects of health care supplement research: health-related wellness, well-being and QoL measures; oral nutritional supplements—an historical research perspective; initiation, adherence/ compliance, and persistence—definitions and considerations; product (and consumer) factors affecting usage; case studies highlighting applied product development, and research efforts that enlisted the services of consumers in various contexts, and communication challenges of health-related claims. The complexity of the topic, consumer comprehensions, and behavior being among the dominant forces to reckon with, demonstrates there is continued work to be done for instilling successful health-related practices with nutritional supplements. As focus has shifted to be even stronger on the consumer, in-context research (with all the approaches, methods, and techniques available, as well as future developments) will undoubtedly strengthen our capabilities, understanding, and consumer influence as this field of endeavor continues to evolve.

Healthcare supplements in context

361

Acknowledgments The author would like to thank the following individuals for their contributions to the case studies cited in this chapter: Kristi Pelc, Amway R&D Director; Miki Nakagawa and Jenny Wu, Amway consumer research scientists; Jennifer Dang and Jessica Gavin, Amway supplement developers; Gene Maly and Troy Nietling, Amway protein product development formulators; Jian Bi, Sensometric Research and Service; Herb Meiselman, Herb Meiselman Training and Consulting Services; Rebecca Bleibaum, Tragon Corporation, and, all of the consumers and others who dedicated their time, but are not mentioned by name, who contributed to these works.

References Agarwal, S., Hordvik, S., & Morar, S. (2006). Nutritional claims for functional foods and supplements. Toxicology, 221, 44–49. Ares, G., Gimenez, A., Vidal, L., Zhou, Y., Krystallis, A., Tsalis, G., et al. (2016). Do we all perceive food-related wellbeing in the same way? Results from an exploratory crosscultural study. Food Quality and Preference, 52, 62–73. Baxter, A. L., Watcha, M. F., Baster, W. V., Leong, T., & Wyatt, M. M. (2011). Development and validation of a pictorial nausea rating scale for children. Pediatrics, 127, 1542–1549. Bi, J. (2005). Similarity testing in sensory and consumer research. Food Quality and Preference, 16, 139–149. Bi, J. (2010). Using the benchmark dose (BMD) methodology to determine an appropriate reduction of certain ingredients in food products. Journal of Food Science, 75(1), R9–16. Bi, J. (2012). A review of statistical methods for determination of relative importance of correlated predictors and identification of drivers of consumer liking. Journal of Sensory Studies, 27, 87–101. Bi, J., Lee, H.-S., & O’Mahony, M. (2013). Statistical analysis of ROC curves for the ratings of the A-not A and the same-different methods. Journal of Sensory Studies, 28, 34–46. Bi, J., O’Mahony, M., & Lee, H.-S. (2013). Non-parametric estimation of d and its variance for the A-not A with reminder. Journal of Sensory Studies, 28, 381–386. Bornkessel, S., Br€oring, S., Omta, O., & van Trijp, H. (2014). What determines ingredient awareness of consumers? A study on ten functional food ingredients. Food Quality and Preference, 32, 330–339. Boutrolle, I., Arranz, D., Rogeaux, M., & Delarue, J. (2005). Comparing central location test and home use test results: Application of a new criterion. Food Quality and Preference, 16, 704–713. Connor, J., Rafter, N., & Rodgers, A. (2005). Do fixed-dose combination pills or unit-of-use packaging improve adherence? A systematic review. Bulletin of the World Health Organization, 82(12), 935–939. Darmon, P., Karsegard, V. L., Nardo, P., Dupertuis, Y. M., & Pichard, C. (2008). Oral nutritional supplements and taste preferences: 545 days of clinical testing in malnourished in-patients. Clinical Nutrition, 27, 660–665. de Abreu, D., Guessous, I., Vaucher, J., Preisig, M., Waeber, G., Vollenweider, P., et al. (2013). Low compliance with dietary recommendations for food intake among adults. Clinical Nutrition, 32, 783–788. Dickinson, A., Blatman, J., El-Dash, N., & Franco, J. C. (2014). Consumer usage and reasons for using dietary supplements: Report of a series of surveys. Journal of the American College of Nutrition, 33(2), 176–182. Diener, E., Wirtz, D., Tov, W., Kim-Prieto, C., Choi, D., Oishi, S., et al. (2009). New measures of well-being: Flourishing and positive and negative feelings. In: E. Diener (Ed.), Social

362

Context

Indicators Research Series: vol. 39. Assessing Well-Being: The Collected Works of Ed Diener. Egan, B., Hodgkins, C., Shepherd, R., Timotijevic, L., & Raats, M. (2011). An overview of consumer attitudes and beliefs about plant food supplements. Food & Function, 2, 747–752. Eussen, S. R. B. M., Verhagen, H., Klungel, O. H., Garssen, J., van Loveren, H., van Kranen, H. J., et al. (2011). Functional foods and dietary supplements: Products at the interface between pharma and nutrition. European Journal of Pharmacology, 668, S2–S9. FDA (Food and Drug Administration). (2017). Dietary Supplements Guidance Documents & Regulatory Information. Available from: https://www.fda.gov/food/guidanceregulation/ guidancedocumentsregulatoryinformation/dietarysupplements/default.htm. Accessed 27 December 2017. Fjeldsoe, B. S., Marshall, A. L., & Miller, Y. D. (2009). Behavior change interventions delivered by mobile telephone short-message service. American Journal of Preventive Medicine, 36(2), 165–173. Foote, J. A., Murphy, S. P., Wilkens, L. R., Hankin, J. H., Henderson, B. E., & Kolonel, L. N. (2003). Factors associated with dietary supplement use among healthy adults of five ethnicities. The Multiethnic Cohort Study. American Journal of Epidemiology, 157, 888–897. Georgiou, N. A., Garssen, J., & Witkamp, R. F. (2011). Pharma–nutrition interface: The gap is narrowing. European Journal of Pharmacology, 651, 1–8. Glanz, K. (1980). Compliance with dietary regimens: Its magnitude, measurements and determinants. Preventative Medicine, 9, 787–804. Goetzke, B., Nitzko, S., & Spiller, A. (2014). Consumption of organic and functional food. A matter of well-being and health? Appetite, 77C, 94–103. Harris, E., Kirk, J., Rowsell, R., Vitetta, L., Sali, A., Scholey, A. B., et al. (2011). The effect of multivitamin supplementation on mood and stress in healthy older men. Human Psychopharmacology: Clinical and Experimental, 26, 560–567. Heuchert, J. P., & McNair, D. M. (2012). The profile of mood states 2nd edition™ (POMS 2TM). North Tonawanda, NY: Multi-Health Systems. Hill, P. L., & Roberts, B. W. (2011). The role of adherence in the relationship between conscientiousness and perceived health. Health Psychology, 30(6), 797–804. Hubbard, G. P., Elia, M., Holdoway, A., & Stratton, R. J. (2012). A systematic review of compliance to oral nutritional supplements. Clinical Nutrition, 31, 293–312. Intille, S. S., Larson, K., Munguia Tapia, E., Beaudin, J. S., Kaushik, P., Nawyn, J., et al. (2006). Using a live-in laboratory for ubiquitous computing research. In: K. P. Fishkin, B. Schiele, P. Nixon, & A. Quigley (Eds.), vol. 3968. Proceedings of PERVASIVE 2006 (pp. 349–365). Berlin, Heidelberg: Springer-Verlag. Koster, E. P. (2009). Diversity in the determinants of food choice: A psychological perspective. Food Quality and Preference, 20, 70–82. Kuesten, C., & Bi, J. (2015). Risk assessment of negative sensory effects in product development using the benchmark dose (BMD) methodology for continuous data. Journal of Sensory Studies, 30, 128–135. Kuesten, C. L., Bi, J., & Meiselman, H. L. (2017). Analyzing consumers’ profile of mood states (POMS) data using the proportional odds model (POM) for clustered or repeated observations and R package ‘repolr’. Food Quality and Preference, 61, 38–49. Kuesten, C. L., Chopra, P., Bi, J., & Meiselman, H. L. (2013). A global study using PANAS (PA and NA) scales to measure consumer emotions associated with aromas of phytonutrient supplements. Food Quality and Preference, 33, 86–97. Kuesten, C. L., Dang, J., Nakagawa, M., Bi, J., & Meiselman, H. L. (2016). Propensity score analysis (PSA) for sensory causal inference – Global consumer psychographics and applications for phytonutrient supplements. Food Quality and Preference, 51, 77–88.

Healthcare supplements in context

363

Lacerda, F. M. M., Carvalho, W. R. G., Hortegal, E. V., Cabral, N. A. L., & Veloso, H. J. F. (2015). Factors associated with dietary supplement use by people who exercise at gyms. Rev Sau´de Pu´blica, 49, 63. Lee, J., & Kim, J. (2009). Factors affecting the use of dietary supplements by Korean adults: Data from the Korean National Health and nutrition examination survey III. Journal of the American Dietetic Association, 109, 1599–1605. Marschollek, M., Gietzelt, M., Schulze, M., Kohlmann, M., Song, B., & Wolf, K. (2012). Wearable sensors in healthcare and sensor-enhanced health information systems: All our tomorrows? Healthcare Informatics Research, 18(2), 97–104. McDonald, D. D., & Nicholson, N. R. (2006). Dietary supplement information and intention to continue and recommend supplements. International Journal of Nursing Studies, 43, 51–57. Methven, L., Rahelu, K., Economou, N., Kinneavy, L., Ladbrooke-Davis, L., Kennedy, O. B., et al. (2010). The effect of consumption volume on profile and liking of oral nutritional supplements of varied sweetness: Sequential profiling and boredom tests. Food Quality and Preference, 21, 948–955. Mintel (2017). Vitamins, minerals and supplements – US report. Mintel (2018). Vitamins and minerals retail market size global forecast. Information Resources, Inc., Retail World, Economist Intelligence Unit, Mintel (search Jan 11, 2018). Moskowitz, H. R. (2000). Engineering out food boredom: a product development approach that combines home use tests and time-preference analysis. Food Quality and Preference, 11, 445–456. Murphy, M. M., Barraj, L. M., Herman, D., Bi, X., Cheatham, R., & Randolph, R. K. (2012). Phytonutrient intake by adults in the united states in relation to fruit and vegetable consumption. Journal of the Academy of Nutrition and Dietetics, 112(2), 222–229. Nechitilo, M., Nguyen, P., Webb-Girard, A., Gonzalez-Casanova, I., Martorell, R., DiGirolamo, A., et al. (2016). A qualitative study of factors influencing initiation and adherence to micronutrient supplementation among women of reproductive age in Vietnam. Food and Nutrition Bulletin, 37(4), 461–474. O’Connor, E. L., & White, K. M. (2010). Willingness to trial functional foods and vitamin supplements: The role of attitudes, subjective norms, and dread of risks. Food Quality and Preference, 21, 75–81. Oakes, M. E., Slotterback, C. S., Bator, T., Corrigan, E. R., Alaimo, S., & Vagner, E. (2005). The vitamin mystique: An analysis of beliefs and reputations of vitamin supplements among a regional sample of grocery shoppers. Food Quality and Preference, 16, 151–155. Otto, L. M., Howerter, A., Bell, I. R., & Jackson, N. (2010). Exploring measures of whole person wellness: Integrative well-being and psychological flourishing. Explore, 6(6), 364–370. Pajor, E. M., Oenema, A., Eggers, S. M., & de Vries, H. (2017). Exploring beliefs about dietary supplement use: Focus group discussions with Dutch adults. Public Health Nutrition, 20 (15), 2694–2705. Peters, C. L. O., Shelton, J., & Nelson, P. S. T. (2008). An investigation of factors that influence the consumption of dietary supplements. Health Marketing Quarterly, 21(1–2), 113–135. Piqueras-Fiszman, B., & Jaeger, S. R. (2015). The effect of product–context appropriateness on emotion associations in evoked eating occasions. Food Quality and Preference, 40, 49–60. Pothoulaki, M., & Chryssochoidis, G. (2009). Health claims: Consumers’ matters. Journal of Functional Foods, 1, 222–228. Rigby, M., Ronchi, E., Graham, S., & Ronchi, E. (2013). Evidence for building a smarter health and wellness future—Key messages and collected visions from a joint OECD and NSF workshop. International Journal of Medical Informatics, 82, 209–219.

364

Context

Rowlands, J. C., & Hoadley, J. E. (2006). FDA perspectives on health claims for food labels. Toxicology, 221, 35–43. Sch€ unemann, H. J., Sperati, F., Barba, M., Santesso, N., Melegari, C., Akl, E. A., et al. (2010). An instrument to assess quality of life in relation to nutrition: Item generation, item reduction and initial validation. Health and Quality of Life Outcomes, 8, 26. Shao, A., Drewnowski, A., Willcox, D. C., Kr€amer, L., Lausted, C., Eggersdorfer, M., et al. (2017). Optimal nutrition and the ever-changing dietary landscape: A conference report. European Journal of Nutrition, 56(Suppl 1), S1–S21. Siegrist, M., Shi, J., Giusto, A., & Hartmann, C. (2015). Worlds apart. Consumer acceptance of functional foods and beverages in Germany and China. Appetite, 92, 87–93. Skevington, S. M., Lotfy, M., & O’Connell, K. A. (2004). The World Health Organization’s WHOQOL-BREF quality of life assessment: Psychometric properties and results of the international field trial. A Report from the WHOQOL Group. Quality of Life Research, 13, 299–310. Thomas, A., van der Stelt, A. J., Schlich, P., & Lawlor, J. B. (2017). Temporal drivers of liking for oral nutritional supplements for older adults throughout the day with monitoring of hunger and thirst status. Food Quality and Preference, 70, 40–48. https://doi.org/10.1016/ j.foodqual.2017.05.001. Thompson, E. R. (2007). Development and validation of an internationally reliable short-form of the positive and negative affect schedule (PANAS). Journal of Cross-Cultural Psychology, 38(2), 227–242. US Department of Health and Human Services, US Department of Agriculture. (2010). Dietary guidelines for americans (7th ed.). Washington, DC: US Government Printing Office. http://www.cnpp.usda.gov/Publications/DietaryGuidelines/2010/PolicyDoc/PolicyDoc. pdf. Accessed 26 December 2017. US Environmental Protection Agency (EPA). (2013). Benchmark Dose Modeling Software (BMDS). http://www.epa.gov/ncea/bmds/. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. Wu, W., Linn, C. T., Fu, C., & Sukoco, B. M. (2012). The role of endorsers, framing, and rewards on the effectiveness of dietary supplement advertisements. Journal of Health Communication, 17(1), 54–75. Yap, S., Noor, N. A. M., Marshall, R., & Liew, K. (2014). Promoting preventive health behaviour among young Malaysian consumers: Toward an integrated conceptual framework. Australasian Marketing Journal, 22, 268–278. Zandstra, E. H., den Hoed, W., van der Meer, N., & van der Maas, A. (2010). Improving compliance to meal-replacement food regimens. Forming implementation intentions (conscious IF-THEN plans) increases compliance. Appetite, 55, 666–670. Zeisel, S. H. (1999). Regulation of “nutraceuticals”. Science, 285, 1853–1855.

Further reading Tragon, Q. D. A., Stone, H., Bleibaum, R. N., & Thomas, H. A. (2012). Sensory evaluation practices. In Descriptive analysis (pp. 233–289). (4th ed.). New York: Elsevier/Academic Press (Chapter 6).

Personal and home care products in context

18

Porcherot C. *, Vignon-Mares M.C. *, Goisbault I. † *Firmenich SA, Geneva, Switzerland, †Strategir, Bordeaux, France

18.1

Main considerations while testing personal and home care products in context

Fragrances can contribute to making personal and home care experiences delightful, highly sensorial, and memorable by enhancing routines. On a daily basis, consumers enjoy fragrances in fabric care with detergents and softeners, in home care with surface and air care products, and in personal care with shampoos, body washes, soaps, deodorants, and creams. Despite the fact that most of these fragranced products are mainly experienced at home, this experience, like that of food products, is affected by the context in which it occurs. Contextual factors interact with the fragranced product and person-related factors in determining use, enjoyment, and emotional responses ( Jaeger & Porcherot, 2017; K€ oster, 2003; Piqueras-Fiszman & Spence, 2015). These factors can be defined by the physical context (time, environment, appropriateness, etc.), the social context (product usage, being alone or with others, etc.), or the personal and cultural context. The role of these different contextual factors is complex, as they all contribute in a holistic way to the ecology, or natural condition, of a situation and affect how consumers perceive and judge a product (Meiselman, 2006).

18.1.1 Consideration of physical context Personal and home care products are experienced repeatedly at different moments of use. Their fragrance impact could be effective starting from their application to several hours, weeks, or even months of usage. Repeated evaluation of the fragrance, as would happen in real life, is therefore critical. Several authors have shown the influence of repeated exposure on the hedonic response. Two phenomena could be observed: either a decrease in acceptability, also called boredom over repeated exposures (Meiselman, de Graaf, et al., 2000), or an increase in acceptability, also called the “mere exposure effect” (Levy, 1998; Porcherot & Issanchou, 1998). Fragrance impact should therefore be measured after several uses on different days to keep track of these potential boredom or mere exposure phenomena. In addition, the fragrances are smelled at different moments of the experience. For example, the shampoo experience starts from smelling the liquid out of the bottle, to its dilution with water when it is perceived on wet hair after rinsing, to the final perception on dry hair after brushing. The shampoo fragrance performs differently in Context. https://doi.org/10.1016/B978-0-12-814495-4.00018-0 Copyright © 2019 Elsevier Inc. All rights reserved.

366

Context

these different stages of hair washing, brushing, and later on. Although a fragrance may deliver well during application and washing, it may not fully satisfy consumers while drying their hair with a blow dryer, and it may not linger long enough on the hair. Overall satisfaction and experience may be affected by these differences in performance, depending on the moment that contributes most to the memorability of the fragrance in the whole experience. For laundry detergents, the experience starts with smelling the undiluted product out of the bottle or pack and continues with removal of the washed laundry from the machine, hanging the wet clothes on the line, drying the laundry, and using or wearing the laundry. Again, the fragrances perform differently in these different stages of laundry washing and drying. One stage might also be more important than another and contribute the most to the memorability of the fragrance in the whole experience, which we could define as the moment of truth. Experiencing fragrances in natural and different conditions also induces the notion of appropriateness. Situational appropriateness was used in a variety of studies to assess the degree to which any product “fit” different usage or consumption situations (e.g., Cardello et al., 2016; Cardello & Schutz, 1996; also refer to Chapter 6 from Giacalone for more details on appropriateness). Different smells could be appropriate in certain situations and not in others. A laundry fragrance will be experienced differently depending on the washing machine location: whether it is in the bathroom, where other fragranced products are stored; in the kitchen, where food odors may interfere; or in a hallway, basement, or even outside the house. Lack of appropriateness will most probably occur when food smells are mixed with the smell of laundry detergent.

18.1.2 Consideration of social context The social context may play an important role because many personal and home care products convey an image of self. Reactions and comments from peers may interfere with user perception and experience. In Mexico, for example, the cleanliness of a house is judged by the intensity and freshness of the fragrance of surface care products and comments by peers are important.

18.1.3 Consideration of personal and cultural context Consumers strongly differ in personal and cultural rituals while experiencing personal and home care products (Dreyfus, 2018). For example, consumers may choose to wash their hair more regularly in the bathtub than in the shower, with a higher or lower temperature, and with more or less shampoo, which will affect fragrance perception. Hair drying could be natural or done with a blow dryer, which may again change the fragrance experience. The fragrance of drying laundry will be more or less dependent on if it is outside or inside the home, or if the laundry is put in a dryer. Moreover, the perception of one personal care product is generally incorporated in the whole beauty routine, and the product of interest may interfere with other products. Interactions of smells with other fragranced products or home odors should not be neglected. As part of their routine, consumers may apply their usual hair conditioner.

Personal and home care products in context

367

The overall judgment of the shampoo may therefore be influenced by the use of the conditioner. Consumers may also add softener to their laundry rituals, which may also interfere with the perception of the laundry detergent.

18.2

What are the implications while testing personal and home care products?

The best way to consider all contextual factors (physical, social, personal, and cultural) is to test personal and home care products at home, in a home use test (HUT). However, HUTs are generally long because they are conducted for several weeks for each product in order to consider normal usage duration and frequency of use (also refer to Chapter 4 from Zandstra on “In Home Testing of Foods”). Tests conducted at home also present many practical challenges because of the lack of control, i.e., it is difficult to control what consumers do, the quantity of product they use, and whether or not they add another product to the product of interest. The experimenters do not know who smelled and rated the product and the circumstances under which they did so (Cardello & Meiselman, 2018). In addition, the capture of fragrance evaluation “in the moment” of use could become a challenge when consumers are in their bathroom and wash their hair, for example. In general, there will be no immediate recording of the evaluation, and they will have to remember their experience to fill in the questionnaire after their bath or shower. The laundry experience is long and we do not know at which particular moment consumers filled in the questionnaire, even when we ask them to focus on the specific moment. More important than these practical reasons, however, is that the results may be disappointing because of not enough discrimination among the tested products. When testing only one product at home, consumers have to evaluate and compare the tested product with their memory of their usual experience. In addition, consumers are less likely to give responses on the qualitative attributes, which are crucial for the reworking of perfume. Another alternative is to conduct consumer product tests in central location tests (CLTs), which allow the control of extraneous and confounding variables found in natural situations. These tests are generally designed to reduce individual differences in rituals and product usage so that the experimenter can ensure delivery of products in the exact same way to all consumers. The quantity of the product, without the use of any other product, the time of the evaluation, and the defined stages are all cautiously chosen and standardized to foster product comparison. Fragranced product tests are generally conducted in a standard room with a neutral environment to control the evaluation conditions. Therefore, conducting consumer tests in CLTs means defining fixed parameters for physical factors on the one hand, and minimizing social and individual factors on the other (Cardello & Meiselman, 2018). Compared with testing in natural conditions, CLTs offer the advantage of presenting several products to participants following a monadic presentation. CLTs are also quicker and allow product comparison. In consideration of incomplete block designs, our standard requirement for the number of respondents is lower for CLTs than for HUTs (n ¼ 60 for CLT, n ¼ 200 for HUT). CLTs allow us to compare up

368

Context

to 10 fragranced products in order to select the best-performing candidates to submit to our clients or to be reworked by our perfumers. Last but not least, CLTs are generally preferred because they are less expensive to perform. A true comparison of CLTs and HUTs is difficult because of the different test conditions related to the number of products (in general several in CLTs and only one, or a few in HUTs) and the frequency of use (in general several in HUTs and only one in CLTs). However, when we compared results from these two test conditions for fragranced shampoos, we generally observed good agreement, with the liking scores obtained from HUTs being higher than those from CLTs. This difference in liking levels was also observed by other authors (Boutrolle, Arranz, Rogeaux, & Delarue, 2005; Boutrolle, Delarue, Arranz, Rogeaux, & Koster, 2007; Hersleth, Mevik, Næs, & Guinard, 2003; King, Weber, Meiselman, & Lv, 2004; Meiselman, de Graaf, et al., 2000; Meiselman, Hirsch, & Popper, 1988; Meiselman, Johnson, et al., 2000). Higher well-being in natural environments, as well as higher appropriateness in HUTs, may be responsible for these higher scores (Boutrolle et al., 2007). Another explanation could be that consumers pay less attention to the products and fragrances when they test them at home in a natural situation than when during CLTs. In addition, more fragrance discrimination was obtained in CLTs because consumers could pay more attention to differences by comparing the products. The drawback is that such differences, while being significant in the laboratory, may have little or no relevance in real-life situations. The potential danger of differences in product discrimination in the two test conditions is that the preferred fragrance in CLTs may be not be noticed in the natural condition of use, thus biasing our conclusions on fragrance performance. Concluding that CLTs could be used for screening purposes is therefore too risky. In addition, the main problem related to CLTs is that they lack external validity. This finding has been supported by a number of studies that compared CLTs to real-life tests (King et al., 2004; King, Meiselman, Hottenstein, Work, & Cronk, 2007; Meiselman, 1992, 2000, 2013). Indeed, many studies showed that CLTs do not predict consumer usage and long-term preferences (Thomson, 2016). With testing in a real-life context (i.e., HUTs) being so important, although dependent on uncontrolled factors, it may be interesting to consider testing by recreating natural environments to better control testing conditions. This raises the question of whether intermediate alternatives between HUTs and CLTs, such as evoking or mimicking/simulating contexts in CLT, could be envisaged and whether they could help maintain a good level of discrimination while yielding relevant fragrance descriptions to guide the reworking of fragrance.

18.3

Intermediate alternatives such as evoking or mimicking/simulating contexts in CLT

Evoking or mimicking/simulating contexts in CLTs may be a good alternative to place the consumer in a relevant context, as would be the case in HUTs, while maintaining controlled test conditions. This approach retains the idea of a controlled setup in laboratory settings but builds on the manipulation of contextual variables and the

Personal and home care products in context

369

observation of their possible effects on behavior. Recently, different strategies have been proposed to introduce relevant contexts into consumer tests ( Jaeger & Porcherot, 2017) to align them more closely with natural conditions of consumption and use and to immerse consumers in a realistic moment of consumption. These strategies encompass evoking contexts with the use of descriptions, scenarios, or pictures, or mimicking/simulating contexts with use of furniture and/or immersive techniques, such as large screens, virtual reality (VR), and 360° pictures or videos.

18.3.1 Methods for evoking context in research and testing Various investigators have tried novel approaches to create realistic contextual situations in the laboratory. In these studies, situations of consumption were evoked and participants were instructed to imagine the situations from brief text descriptions ( Jaeger & Meiselman, 2004), pictures (Hersleth, Monteleone, Segtnan, & Naes, 2015), audio recordings (K€ oster, 2003), or written scenarios (Hein, Hamid, Jaeger, & Delahunty, 2010, 2012; Jaeger & Rose, 2008). Written scenarios are brief texts meant to evoke a sense of presence in a real situation. The written scenario used in Hein et al. (2010) instructed consumers to imagine an occasion when a food or a beverage could be consumed and required them to think of that occasion during the test. Each consumer imagined his or her own personal occasion. Piqueras-Fiszman and Jaeger (2014a, 2014b, 2014c) later explored how evoked consumption contexts can affect associative emotional responses to products. Results showed that positive emotions were more frequently selected in appropriate consumption contexts, whereas negative emotions were more frequently selected in less appropriate contexts. We implemented evoked contexts in our laboratory with brief text descriptions and pictures of the situation. We compared three test conditions in a specific study on softeners: a real test situation in which participants were invited to smell out of laundry machines availables in our laboratories as if they were doing their laundry at their home, a traditional CLT without any context, and a CLT in which four laundry moments were evoked with brief text descriptions and pictures. Respondents were instructed to imagine that they were in one of the four particular laundry situations while smelling different fragrances that we provided: (1) when opening the bottle and pouring the softener into the washing machine, (2) when opening the machine and removing the wet laundry, (3) when drying the laundry in a tumble dryer or on the line, and (4) when ironing or folding the laundry after it was dry. Different products were also provided to best illustrate these different laundry moments: (1) neat or undiluted liquid softener, (2) wet washed laundry, (3) dry washed laundry, and (4) dry washed laundry that needed to be scrubbed by the respondents with their hands. Different emotional responses were observed from these four laundry moments (Fig. 18.1). Energy and well-being were highest after the laundry was rubbed by the respondents (point 4 above), and lowest when the laundry was presented dry (point 3 above), soothing being the strongest emotional response for dry laundry. The same observation was reported by Schifferstein, Fenko, Desmet, Labbe, and Martin (2013), who explored consumers’ feelings at different stages of a food product consumption

370

Context

Fig. 18.1 Emotional responses corresponding to four laundry moments: (1) when opening the bottle of softener (2) when removing the wet laundry from the machine (3) when laundry is drying (4) when ironing or folding the laundry.

Fig. 18.2 Emotional responses corresponding to three test conditions: CLT without context, CLT with evoked context, real test situation.

sequence, from shopping to preparation to consumption, and found that different feelings were elicited during the various stages of user-product interactions. In addition, the results of the two conditions “real test situation” and “CLT with evoked context” were closer than were the results of the CLT conducted without any context (Fig. 18.2). From our perspective, the added value of evoked contexts in neutral CLTs needs no further demonstration and seems easy to implement. However, while the results of CLTs with evoked context match the emotional profile of the real situation with more sensuality and well-being feelings, they did not exactly match the results of the real test situation in terms of intensity of the emotional response, and therefore this warrants further research.

18.3.2 Mimicking context with furniture and/or video screens Other authors have implemented immersive approaches by recreating a full consumption experience in laboratory settings. Sensory laboratory environments were setup to mimic a natural consumption situation or point of purchase and were recreated in terms of visual, auditory, and olfactory cues.

Personal and home care products in context

371

Petit and Sieffermann (2007) recreated coffee shop environments by using physical elements that recreated a coffee shop atmosphere. This approach was advocated in consideration of its optimized cost, organization, and administration in comparison to tests conducted in natural environments. Bangcuyo et al. (2015) simulated a coffee shop environment in a laboratory using a 20-min video recording of a natural setting on high-definition LCD screens with accompanying olfactory stimulation and sounds of a coffee shop. The immersive environment had appropriate furniture for a coffee shop. Affective responses were more discriminative and predictive of future liking in this recreated environment than were those obtained in standard CLT settings. Panelists also reported being more engaged in the test on the engagement questionnaire (O’Brien & Toms, 2010; Witmer & Singer, 1998). Kim, Lee, and Kim (2016) compared an evocative condition with a simulated cafe condition in laboratory settings. In the evocative condition, subjects were instructed to read evocative phrases and to imagine the situation. In the simulated cafe condition, the room resembled a common cafe with tablecloths and furniture as might be found in a cafe. Results indicated that simulating the physical environment was more effective than imagining a situation and coffees were rated higher in the simulated cafe setting. Sester et al. (2013) implemented the immersion approach in bars and compared the ambience of two bars by changing only the color of the furniture (wooden or blue tables), the music, and the videos projected on a wall. The authors showed that differences in overall warmth led to different drink choices. For instance, participants more frequently selected warm beverages when the video clip presented icebergs. These immersive approaches seem to be valuable for studying the effects of manipulating contextual variables. They also appear to be interesting and effective in studying the effects of context when the recreated environment comes close to natural situations. Moreover, these studies emphasized the importance of environmental effects and advocated more consumer engagement in recreated environments.

18.3.3 Simulating context with immersive virtual reality More immersive contexts are possible. A variety of approaches were presented as part of the European Sensory Network workshop on immersive tools at the 2016 EuroSense Conference in Dijon, France (Porcherot, Kremer, Dreyfuss, & Almi, 2016). One of these approaches is immersive virtual reality (iVR). Although it was mainly developed for gaming purposes and therapy (to reduce pain and anxiety in burn victims, Morris, Louw, & Grimmer-Somers, 2009; to help restore memory deficits in people with acquired brain injury, Yip & Man, 2013; and to reduce compulsive eating-related disorders, Cesa et al., 2013; Gutierrez-Maldonado, Wiederhold, & Riva, 2016; Perpina˜ & Roncero, 2016), this new technology could offer many possibilities for research on consumer behavior. iVR has recently been introduced in consumer research and is seen as a future direction to follow in sensory and consumer science ( Jaeger et al., 2017). It is also mentioned in several chapters of this book. VR (virtual reality) is computer simulation that immerses people into almost real environments with the perception of being

372

Context

physically present in a nonphysical world. It has the potential to replace physically created environments with immersive settings for product research ( Jaeger & Porcherot, 2017). With this technology, consumers wear VR headsets or goggles, allowing them to experience a complete visual and auditory environment. The advantage of such virtual experiences is that they activate multiple sensory pathways in which all senses work together, and this may elicit more emotions and increase consumer engagement by increasing the realism of the environments (Dinh, Walker, Hodges, Chang, & Kobayashi, 1999; Scholz & Smith, 2016; Witmer & Singer, 1998). This method produces better immersion in the contextual situation and is more realistic and more engaging, and may therefore improve the extrinsic validity of laboratorybased research. In addition, iVR retains the idea of the controlled setup of laboratory settings. This new immersive technology also has the advantage of being able to recreate the complexity of the real world, at the same time allowing experimenters to constrain experimental parameters (Ischer et al., 2014). Porcherot, Delplanque, Ischer, De Marles, and Cayeux (2015) and Porcherot, Kremer, et al. (2016) investigated the extent to which virtual experimental settings could be applied to the study of emotions elicited by scents and could highlight contextual influences. Olfaction was introduced in three different VR environments (bathroom, kitchen, and public transport). A within-subject experiment was conducted with three different experimental setups and different levels of immersion: (1) in sensory booths without immersion, (2) in sensory booths with interactive virtual worlds shown on a large computer screen, and (3) in a three-dimensional (3D) immersive virtual laboratory, consisting of four sides that presented 3D images to participants wearing iVR glasses. Results showed that this iVR system successfully provided fun and interactive experiences, generating different emotional responses to the odors and environments and with more physiological arousal. The two-dimensional (2D) experimental settings in the study (i.e., point (2) above) appeared to be a good alternative to the 3D settings, with generally higher levels of feelings than with the two other conditions (i.e., point (1) and (3) above) despite the fact that participants declared being more immersed in the 3D condition than in the two other conditions. Participants’ experiences of immersion differed in the three virtual locations, pointing out the relevance of 3D immersion. In the following studies, the authors used video screens to investigate consumer behavior at the point of purchase and to study the influence of packaging, shelf placement, pricing, and advertising (Daugherty, Li, & Biocca, 2005). In the same vein, Domina, Lee, and MacGillivray (2012) studied factors affecting consumer intention to shop in a virtual world and showed that the perceived enjoyment of consumers influenced their shopping intentions. Other authors (Pantano & Servidio, 2012) studied the benefits of VR in retail stores and observed that consumers were more attracted to these stores, and more attracted to use them, than they were to traditional retail stores. Finally, Van Herpen, Van den Broek, Van Trijp, and Yu (2016) found that, compared with a pictorial (2D) store, a virtual store elicits behavior that is more similar to behavior in a physical store in terms of product selection and amount of money spent. However, the VR condition did not completely reproduce behavior in natural

Personal and home care products in context

373

conditions in that participants in VR stores were tempted to buy more products, spend more money, buy more well-known brands, or respond more strongly to price promotions than they were in physical stores. In this particular shopping case, VR may be more game-like with less constraints (e.g., time) than in real situations. The evolution of VR methods in sensory and consumer research is still in its infancy, and more studies needed to define the best immersive conditions and the best way to immerse consumers.

18.3.4 Simulating context with 360° immersion Although expensive and complex equipment in 3D laboratories was necessary in the past, progress in this field has made equipment more affordable. Approaches other than iVR were also presented at the immersive tools workshop at the 2016 EuroSense Conference (Porcherot, Kremer, et al., 2016): 360° immersion with VR headsets, Oculus Rift, PlayStation VR, Gear VR, and HTC Vive. We conducted three studies to investigate whether 360° immersion could be used to create immersive environments for product testing in CLTs and to help in the selection of the best products according to emotional state and hedonic scores. This immersive approach consisted of recreating a full experience with a 360° picture displayed by the Samsung Gear VR technology. Respondents were immersed in an appropriate 360° environment while smelling the products. This technique was designed by Strategir (Bordeaux, France) and participants used a Samsung Gear VR 360 headset. Three tests were conducted in 360° immersion with VR headsets to compare a traditional CLT protocol with a protocol integrating VR immersion. Two tests were conducted in Russia and China for the hair care category (shampoos), and one test was conducted in Germany for the home care category (laundry detergents). The most critical part is the choice of the virtual environment, which should be relevant to the consumers of the chosen countries. The 360° immersive context should correspond to the moment of truth, or the key moment when the fragrance is the most important for the consumers during the product experience. Despite being virtual, the context should be as real as possible to truly immerse consumers. With the 360° device, consumers should be immersed in a relevant virtual context and be more connected to the product category while they smell the perfumes. For shampoos, we defined beforehand that the main fragrance experience of interest corresponded to the buying moment in a supermarket, with consumers being situated in front of a shampoo shelf on which several shampoo products were presented. The stores and the shelfs were virtually developed in order to immerse Chinese and Russian shoppers in a plausible store and in front of a representative shampoo shelf from their market. Consumers smelled the undiluted shampoos while in this situation (Fig. 18.3A). For laundry detergents, we defined that the main fragrance experience of interest corresponded to the moment when the consumers took their laundry from the machine and put it on the drying line. In Germany, this situation is generally performed in a

374

Context

Fig. 18.3 Environments used for 360° immersion with VR headsets: (A) shampoo shelf storage (B) laundry room.

dedicated laundry room. The virtual laundry room was developed with 3D objects based on pictures of real laundries. The chosen laundry room included a washer, line dryer, and washed laundry. It was clean without too many objects so that consumers would feel confident and pleased enough to be immersed in this environment. Consumers smelled wet washed laundry with the detergents while in this situation (Fig. 18.3B). For each product category, two tests were run in parallel with two groups of consumers: one group who performed the test with 360° immersion and another who performed it without (Fig. 18.4A and B). For laundry detergents, 220 participants tested 6 out of 11 perfumes, each fragrance being tested 120 times. For shampoos, 312 respondents tested 5 out of 13 shampoo fragrances.

(A)

(B)

Fig. 18.4 The two test conditions for the laundry category: (A) sniffing liquid detergent scents in a neutral environment (35 min), (B) sniffing liquid detergent scents while immersed in a 360° environment (40 min).

Personal and home care products in context

375

Apart from 360° immersion, an identical test procedure was applied for the two test conditions without immersion so that only the context differed. The only difference was that in 360° immersion, respondents were familiarized with the device and environment for a few minutes before starting the test. The two tests were conducted in the same room. Respondents were prompted to sniff each fragrance as often as needed with a small break after every three products. For laundry detergents, the fragrances were evaluated on humid laundry tissues that had been washed with the fragranced detergent. These tissues were presented in white plastic containers. Shampoos were presented undiluted in white plastic bottles. Sessions lasted about 30 min. In the two conditions, an interviewer read out the questions and the scales to respondents and recorded their answers. The same protocol was implemented in the two conditions to avoid any bias of scaling and to avoid any interviewer effects (Fig. 18.5). Questions were about fragrance liking (5-point scale), intensity (5-point scale), fragrance descriptors (in a check-all-that-apply [CATA] list), product benefits (5-point scale), emotions (ScentMove, CATA list, Porcherot et al., 2010), and justabout-right intensity (5-point scale). We analyzed the results of these studies by looking at both consumer behavior and comfort and, the quality of the data.

Fig. 18.5 Test conditions for the laundry category: (A) washing machines, (B) plastic containers where laundry tissues are stored, (C) briefing the interviewers, (D) interview with a subject undergoing 360° immersion.

376

Context

Consumer interviews indicated that the headsets were not too heavy and were easy to wear. None of the respondents felt uncomfortable while wearing the headset. Despite the fact that they were allowed to remove it every three samples, many respondents decided to keep it on during the entire test session. Respondents were fully immersed in the situation; they felt they were in a real environment in the test condition with 360° immersion and as though they were at home for the laundry test or in a supermarket for the shampoo test. For the laundry test, the consumers felt as though they were using the machine as in a natural situation. For the shampoo test, Chinese consumers felt as though they were standing in front of a shelf at a real store, they were excited about wearing the headset during the test, and they could answer the questions while thinking of shampoos. Both Russian and Chinese consumers felt comfortable and were interested in this unusual but realistic experience. They found the interview to be easy and fast. In general, more engagement in the task could be observed with 360° immersion, as more participants focused on the smelling experience and smelled the same fragrance more often and for a longer time than the nonimmersion group did. Respondents were also more concentrated on their task because they had no external distractions. This positive “wow” effect with 360° immersion did not influence the participants’ ratings, because results showed that the liking scores were similar for the two test conditions i.e., with and without 360° immersion. Differences in liking were obtained for some of the fragrances between the two conditions. However, a comparison of liking and product benefit scores in the two conditions showed that the higher level of engagement obtained with 360° immersion resulted in more discrimination and the 360° immersion showed less of a halo effect (i.e., tendancy to give the same scores to all questions for a same product) regarding product benefits. More important, we observed better data quality and more actionable results for reworking of fragrance because the consumers were selective in the CATA task, choosing more descriptive and relevant terms (fruity, light, etc.) and fewer generic terms (clean, fresh) than they did in the test condition without immersion. From an open question at the end of the experiment, consumers had to spontaneously give three words that best characterized their experience. Thirty-eight percent of the consumers found the experience with 360° immersion exciting, great, and fun compared with 23% for the nonimmersion group. Among the 360° immersion group, 27% of respondents felt the experience was new; 20% felt that it was pleasant and comfortable, and this percentage is similar for the nonimmersion group. In addition, and more importantly, more consumers spontaneously mentioned that they were involved, concentrated, and focused compared to the nonimmersion group (Fig. 18.6). These three experiments provided good results and information about 360° immersion. However, more studies are needed to investigate the added valued of this immersive technique. Predictability in such an environment as compared with real life is the key remaining question. Moreover, even if virtual environments are constructed cautiously and with good knowledge of the moment of truth, or key moment of the experience, we still do not know how these environments influence product perception.

Personal and home care products in context

377

Fig. 18.6 Frequency of words used to describe the 360° immersive experience (in red) as compared with the standard test experience (in gray).

18.4

Recommendations and perspectives when evoking or mimicking/simulating context in CLTs

Several of the studies mentioned earlier have shown the importance of context in product perception, and product acceptability results may suffer with the removal of the product from the context in which it is normally used or consumed. The level of involvement has also been raised as being important to perception, and the accuracy of results may be in question if consumers feel less involved with the product and the situation (Lyman, 1989). Because of the complexity of implementing tests in natural settings, several authors have tried novel approaches to implementing context in laboratory settings by evoking or simulating a real context for consumption. These experiments have shown that evoking or simulating context in laboratory settings could help by maintaining controlled test conditions, and many studies have succeeded in implementing this approach. However, many questions remain. Is it better to evoke context and let consumers think and recreate their own appropriate situation, or is it better to immerse consumers in a given environment, with the risk that this environment may not be relevant for some consumers? Which immersive technique is best for simulating context? Different studies have demonstrated that diverse tools such as iVR or 360° immersion have been successful in reproducing an almost real environment, where consumers were able to move and possibly interact with products and/or avatars. If and how immersive techniques transform research practices remains to be seen. The following paragraphs summarize what has been learned about these techniques and what remains to be validated.

378

Context

18.4.1 Mimicking/simulating environments The first point of discussion is dedicated to the creation of virtual environments, which should be done with caution and in consideration of the key moment of product use or consumption. The context needs to be meaningful to consumers and appropriate to the product category and the situation ( Jaeger & Porcherot, 2017). The problem of relevancy has been mentioned by Petit and Sieffermann (2007). Immersion techniques simulate a physical context that may not be similar enough to real life or relevant enough for consumers and their personal experiences. Although 360° videos may help increase the impression of being similar to real life, one could still question the validity of these measures and the reality of these environments, which are not those that participants are used to in real life. One could also question whether these aspects need to be even closer to those in real life, i.e., whether these contexts are realistic and relevant to consumers. However, our latest experiments conducted with 360° immersion led us to think that our design of a virtual environment was good enough for consumers, who felt as though they were in a real world and a real situation of doing laundry or buying shampoo. When compared with traditional CLTs without any immersion, reliability has been partly demonstrated in our experiments because of the consistency between liking scores and product benefits. The better quality of the data collected and the more actionable results to improve fragrances were also mentioned.

18.4.2 Context relevance Recent studies presented at the 12th Pangborn Sensory Science Symposium (Providence, Rhode Island, 2017) have shown no effect of different contexts: i.e., party, classroom, and home living room for cola beverages (Arato, Liu, Hooker, Parasidis, & Simons, 2017) and night club or beach for beer tasting (Brasset, Gachet, Abiven, & Delarue, 2017). This absence of influence by context may be explained by the lack of product differences, or the lack of relevance of these contexts, or because these beverages could be consumed in different contexts. Context might also have less effect on well-known products and brands. It therefore appears important to find valid measures of context relevance and realism. What it takes to physically create an environment that immerses consumers and successfully creates what Sester et al. (2013) refer to as a relevant “moment of consumption” remains to be established. It may even be more challenging for fragrances, where the moment of use or moment of experience is not that clear, or may correspond to different times of the day, different environments, and different situations. Further studies are necessary to check context validity, and comparisons of test results conducted on different key moments by country are warranted (i.e., self vs supermarket vs bathroom; undiluted from bottle vs wet hair vs bloom for shampoos).

18.4.3 Potential bias It is also important to consider that evoking or simulating context might underestimate or overestimate the contextual effect in comparison with what might happen in a real situation. In the case of immersive techniques, because the exercise is new to the

Personal and home care products in context

379

participants and they enjoy the immersive experience, they might pay more attention to the environments than to the tested products. Participants may also be more positive towards products because of the enjoyable immersive experience. The influence of potential biases from the immersive environments therefore needs to be better identified. Some objects added to the virtual environment may have strong and unpredicted effects; i.e., the use of colors or other elements in a consumer test may give consumers certain associations and emotional feelings. Careful pretesting of such environmental factors should be considered. Our laundry environment was chosen to be clean so that consumers would have good feelings. Would our results have changed had the laundry room been tidier and contained more objects, or had the laundry room been more strongly colored? We thus recommend comparisons of the results of tests conducted with subtle differences in the environment, such as the addition of colors or other elements.

18.4.4 Evoking versus mimicking/simulating context The effect of potential bias raises the question of whether it is better to immerse consumers in a predefined environment or to allow them to imagine their own situation and context. Jaeger et al. (2017) found that a personalized evoked context resulted in smaller sample discrimination than a predefined context did. This result could be explained by the variability in individual situations while allowing consumers to imagine on their own. Therefore, one could question the balance between the consumer’s own situation versus the control and homogeneity of the environment. More studies need to be conducted on such comparisons.

18.4.5 Level of immersion regarding presence and engagement The level of immersion with the manipulation of the context variables should also be investigated (more complexity by adding sound, touch, and other sensory modalities). Various authors have tried different immersive techniques such as projection, VR headset, 360° pictures or videos, 3D environments, and mixed reality, but it is still not clear whether one approach is better than another. Considering the constraints of developing an environment in iVR, perhaps fewer immersion levels would be enough, as might be observed by projection on a large screen. Some authors have shown that it is possible to implement a plausible moment of consumption in a controlled setting with only a few environmental elements (Bangcuyo et al., 2015). Some studies have also shown more presence with VR than with images and photographic cues. Which evocating or simulating technique is best for adding context to laboratory settings or immersing consumers is still not well-defined; thus, more studies need to be conducted to compare the different approaches and techniques. In addition, the challenge of delivering odor or taste products with immersive approaches should not be neglected. For any approach, measuring the sense of presence and engagement by using the questionnaire developed by Witmer and Singer (1998) is recommended because the validity of the measure may be higher when consumers feel more involved (Lyman, 1989).

380

Context

18.4.6 Predictive value of immersive approaches Of similar importance, the predictive value needs to be further investigated in comparison with home tests or market data ( Jaeger et al., 2016). Indeed, contradictory results have been obtained by different authors. Bangcuyo et al. (2015) reported more discrimination with immersion compared with no immersion, and a study by Jordan, Hemmerling, Lutsch, and M€ oslein (2017) showed that consumer segmentation with immersive techniques was closer to real in-home tests than were traditional laboratory tests without context. However, a study conducted by Gerigk, Hehn, Koch, M€oslein, and Pessel (2017) showed the same level of difference between two products independent of the test condition, meaning that the test conducted with 360° immersion was no better than the test without context in comparison to real-life settings. These results could also be explained by the level of product differences. The product type is another reason for variation in effects. Andersen, Kraus, and Bredie (2017) showed that the immersive approach induced a specific beverage choice, whereas no similar effect was observed for body lotions. More studies need to investigate the extent to which inducing context in CLTs helps to bridge the gap between traditional CLTs and HUTs and correlates well with the results of HUTs. Comparison studies in several countries and for several product categories are recommended.

18.5

Conclusion

Some authors are in favor of considering multiple contextual factors to recreate reallife situations and replicate the complexity of real-world experiences when testing consumer products (Go´mez-Corona, Chollet, Escalona-Buendı´a, & Valentin, 2017; Meiselman, Johnson, et al., 2000; Porcherot, Petit, Giboreau, Gaudreau, & Cayeux, 2015; Pouyet, Cuvelier, Benattar, & Giboreau, 2015). But HUT conditions are typically uncontrolled, bringing noise to the variable of interest and affecting the outcome. New methods and techniques that consist of evoking or simulating context in laboratory settings have been developed. They allow product testing in controlled conditions in CLTs by adding context that is relevant to product consumption and use but avoiding a neutral and sterile environment. These methods open the field for innovation as interactions between product and context offer multiple combinations and allow for a better understanding of the role of contextual elements in product acceptance. The future of contextual research by evocation or mimicking/simulating context in the laboratory is promising. The studies that have been conducted so far offer encouraging results and good information. More studies are needed, however, to investigate the validity of these methods and techniques and to determine whether they introduce any bias, the main challenge being to design the most relevant evocating or simulating context as compared with real-life settings. To date, there is no proof that immersing consumers is close enough to real-life conditions and that it improves the ecological validity of consumer research performed in a laboratory, or

Personal and home care products in context

381

that it even predicts the real-life experience. According to K€oster (2009), it is important to: develop additional and ecologically valid research methods that take the dynamic aspects of the complex interactions between the product, the individual consumer and the environment into account and thus provide a better basis for the prediction of consumer food choice (p. 72).

We conclude by underscoring the challenge: the way in which the context is evoked, recreated, or simulated is a critical factor that needs attention.

References Andersen, I. N., Kraus, A., & Bredie, W. L. (2017). Assessing product evaluations in neutral or pictorial and immersive beach-like contexts. In: Paper presented at the 12th Pangborn Sensory Science Symposium, Providence, RI. Arato, M., Liu, R., Hooker, N. H., Parasidis, E., & Simons, C. T. (2017). Consumer testing using immersive technologies. In: Paper presented at the 12th Pangborn Sensory Science Symposium, Providence, RI. Bangcuyo, R. G., Smith, K. J., Zumach, J. L., Pierce, A. M., Guttman, G. A., & Simons, C. T. (2015). The use of immersive technologies to improve consumer testing: the role of ecological validity, context and engagement in evaluating coffee. Food Quality and Preference, 41, 84–95. Boutrolle, I., Arranz, D., Rogeaux, M., & Delarue, J. (2005). Comparing central location test and home use test results: application of a new criterion. Food Quality and Preference, 16(8), 704–713. Boutrolle, I., Delarue, J., Arranz, D., Rogeaux, M., & Koster, E. P. (2007). Central location test vs. home use test: contrasting results depending on product type. Food Quality and Preference, 18, 490–499. Brasset, A. C., Gachet, J., Abiven, F., & Delarue, J. (2017). Taking control of product testing context thanks to a multi-sensory immersive room: a case study on alcohol-free beer. In: Paper presented at the 12th Pangborn Sensory Science Symposium, Providence, RI. Cardello, A. V., & Meiselman, H. L. (2018). Contextual influences on consumer responses to food products. In: G. Ares & P. Varela (Eds.), Vol. 2. Methods in consumer research: Alternative approaches and special applications (pp. 4–32). Duxford: Elsevier. Cardello, A. V., Pineau, B., Paisley, A. G., Roigard, C. M., Chheang, S. L., Guo, L. F., et al. (2016). Cognitive and emotional differentiators for beer: an exploratory study focusing on “uniqueness” Food Quality and Preference, 54, 23–38. Cardello, A. V., & Schutz, H. G. (1996). Food appropriateness measures as an adjunct to consumer preference/acceptability evaluation. Food Quality and Preference, 7(3), 239–249. Cesa, L. G., Manzoni, M. G., Bacchetta, M., Castelnuovo, G., Conti, S., Gaggioli, A., et al. (2013). Virtual reality for enhancing the cognitive behavioral treatment of obesity with binge eating disorder: randomized controlled study with one-year follow-up. Journal of Medical Internet Research, 15, e113. Available from: http://www.jmir.org/2013/6/e113. Accessed 2 June 2013. Daugherty, T., Li, H., & Biocca, F. (2005). Experiential ecommerce: a summary of research investigating the impact of virtual experience on consumer learning. In C.

382

Context

Haugtvedt, K. Machleit, & R. Yalch (Eds.), Online consumer psychology: Understanding and influencing consumer behavior in the virtual world. Mahwah, NJ: Lawrence Erlbaum. forthcoming. Dinh, H. Q., Walker, N., Hodges, L. F., Chang, S., & Kobayashi, A. (1999). Evaluating the importance of multi-sensory input on memory and the sense of presence in virtual environments. In Proceedings of the Virtual Reality Conference, 1999, VR ‘99 (pp. 222– 228). Washington, DC: IEEE Computer Society. Domina, T., Lee, S., & MacGillivray, M. (2012). Understanding factors affecting consumer intention to shop in a virtual world. Journal of Retailing and Consumer Services, 19(6), 613–620. Dreyfus, L. (2018). Cosmetic products. G. Ares & P. Varela (Eds.), Methods in consumer research: Alternative approaches and special applications (pp. 399–410). Vol. 2 (pp. 399–410). Duxford: Elsevier. Gerigk, A., Hehn, P., Koch, J., M€oslein, R., & Pessel, F. (2017). Sniff test with toilet rim blocks in different settings comparing virtual reality, sensory lab, and home use tests. In: Paper presented at the 12th Pangborn Sensory Science Symposium, Providence, RI. Go´mez-Corona, C., Chollet, S., Escalona-Buendı´a, H. B., & Valentin, D. (2017). Measuring the drinking experience of beer in real context situations: the impact of affects, senses, and cognition. Food Quality and Preference, 60, 113–122. Gutierrez-Maldonado, J., Wiederhold, B. K., & Riva, G. (2016). Future directions: how virtual reality can further improve the assessment and treatment of eating disorders and obesity. Cyberpsychology, Behavior, and Social Networking, 19(2), 148–153. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2010). Application of a written scenario to evoke a consumption context in a laboratory setting: effects on hedonic ratings. Food Quality and Preference, 21(4), 410–416. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2012). Effects of evoked consumption contexts on hedonic ratings: a case study with two fruit beverages. Food Quality and Preference, 26(1), 35–44. Hersleth, M., Mevik, B.-H., Næs, T., & Guinard, J.-X. (2003). Effect of contextual factors on liking for wine—use of robust design methodology. Food Quality and Preference, 14, 615–622. Hersleth, M., Monteleone, E., Segtnan, A., & Naes, T. (2015). Effects of evoked meal contexts on consumers’ responses to intrinsic and extrinsic product attributes in dry-cured ham. Food Quality and Preference, 40(Part A), 191–198. Ischer, M., Baron, N., Mermoud, C., Cayeux, I., Porcherot, C., Sander, D., et al. (2014). How incorporation of scents could enhance immersive virtual experiences. Frontiers in Psychology, 5, 736. Available from: https://www.frontiersin.org/articles/10.3389/fpsyg. 2014.00736/full. Accessed 17 July 2014. Jaeger, S. R., Fiszman, S., Reis, F., Chheang, S. L., Kam, K., Pineau, B., et al. (2017). Influence of evoked contexts on hedonic product discrimination and sensory characterizations using CATA questions. Food Quality and Preference, 56, 138–148. Jaeger, S. R., Hort, J., Porcherot, C., Ares, G., Pecore, S., & MacFie, H. J. H. (2016). Future directions in sensory and consumer science: four perspectives and audience voting. Food Quality and Preferences, 56, 301–309. Jaeger, S. R., & Meiselman, H. L. (2004). Perceptions of meal convenience: the case of at-home evening meals. Appetite, 42(3), 317–325. Jaeger, S. R., & Porcherot, C. (2017). Consumption context in consumer research: methodological perspectives. Current Opinion in Food Science, 15, 30–37.

Personal and home care products in context

383

Jaeger, S. R., & Rose, J. M. (2008). Stated choice experimentation, contextual influences and food choice: a case study. Food Quality and Preference, 19(6), 539–564. Jordan, G., Hemmerling, S., Lutsch, D., & M€oslein, R. (2017). Virtual reality as a bridge between realism and standardization: combining strengths of traditional lab and in-home tests. In: Paper presented at the 12th Pangborn Sensory Science Symposium, Providence, RI. Kim, S., Lee, S. M., & Kim, K. (2016). Consumer acceptability of coffee as affected by situational conditions and involvement. Food Quality and Preference, 52, 124–132. King, S. C., Meiselman, H. L., Hottenstein, A. W., Work, T. M., & Cronk, V. (2007). The effects of contextual variables on food acceptability: a confirmatory study. Food Quality and Preference, 18, 58–65. King, S. C., Weber, A. J., Meiselman, H. L., & Lv, N. (2004). The effect of meal situation, social interaction, physical environment and choice on food acceptability. Food Quality and Preference, 15(7–8), 645–653. K€ oster, E. P. (2003). The psychology of food choice: some often encountered fallacies. Food Quality and Preference, 14(5), 359–373. K€ oster, E. P. (2009). Diversity in the determinants of food choice: a psychological perspective. Food Quality and Preference, 20(2), 70–82.  Levy, C. M. (1998). Elaboration de m ethodologies adapt ees a‘ l’ etude de la dynamique des pr ef erences (Methodological contribution to the study of the dynamics of preference). Doctoral thesis. Dijon, France: Universite de Bourgogne. Lyman, B. (1989). A psychology of food, more than a matter of taste. New York: Van Nostrand Reinhold. Meiselman, H. L. (1992). Methodology and theory in human eating research. Appetite, 19, 49–55. Meiselman, H. L. (Ed.), (2000). Dimensions of the meal. Gaithersburg, MD: Aspen. Meiselman, H. L. (2006). The role of context in food choice, food acceptance and food consumption. In R. Shepherd & M. Raats (Eds.), The psychology of food choice (pp. 179– 199). Wallingford: CABI Publishing. Meiselman, H. L. (2013). The future in sensory/consumer research: evolving to a better science. Food Quality and Preference, 27, 208–214. Meiselman, H. L., de Graaf, C., & Lesher, L. L. (2000). The effects of variety and monotony on food acceptance and intake at a midday meal. Physiology & Behavior, 70 (1), 119–125. Meiselman, H. L., Hirsch, E. S., & Popper, R. D. (1988). Sensory, hedonic, and situational factors in food acceptance and consumption. In D. M. H. Thomson (Ed.), Food acceptability. Barking, UK: Elsevier Science. Meiselman, H. L., Johnson, J. L., Reeve, W., & Crouch, J. E. (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35(3), 231–237. Morris, L. D., Louw, Q. A., & Grimmer-Somers, K. (2009). The effectiveness of virtual reality on reducing pain and anxiety in burn injury patients: a systematic review. The Clinical Journal of Pain, 25, 815–826. O’Brien, H. L., & Toms, E. (2010). The development and evaluation of a survey to measure user engagement. Journal of the American Society of Information Science and Technology, 61, 50–69. Pantano, E., & Servidio, R. (2012). Modeling innovative points of sales through virtual and immersive technologies. Journal of Retailing and Consumer Services, 19(3), 279–286.

384

Context

Perpina˜, C., & Roncero, M. (2016). Similarities and differences between eating disorders and obese patients in a virtual environment for normalizing eating patterns. Comprehensive Psychiatry, 67, 39–45. Petit, C., & Sieffermann, J. M. (2007). Testing consumer preferences for iced-coffee: does the drinking environment have any influence? Food Quality and Preference, 18(1), 161–172. Piqueras-Fiszman, B., & Jaeger, S. R. (2014a). The impact of evoked consumption contexts and appropriateness on emotion responses. Food Quality and Preference, 32, 277–288. Piqueras-Fiszman, B., & Jaeger, S. R. (2014b). Emotion responses under evoked consumption contexts: a focus on the consumers’ frequency of product consumption and the stability of responses. Food Quality and Preference, 35, 24–31. Piqueras-Fiszman, B., & Jaeger, S. R. (2014c). The impact of the means of context evocation on consumers’ emotion associations towards eating occasions. Food Quality and Preference, 37, 61–70. Piqueras-Fiszman, B., & Spence, C. (2015). Sensory expectations based on product extrinsic food cues: an interdisciplinary review of the empirical evidence and theoretical accounts. Food Quality and Preference, 40, 165–179. Porcherot, C., Delplanque, S., Ischer, M., De Marles, A., & Cayeux, I. (2015). Applying immersive virtual reality to the study of emotions elicited by scents: what is the value of a virtual experimental setting? In: Paper presented at the 11th Pangborn Sensory Science Symposium, Gothenburg, Sweden. Porcherot, C., Delplanque, S., Raviot-Derrien, S., Le Calve, B., Chrea, C., Gaudreau, N., et al. (2010). How do you feel when you smell this? Optimization of a verbal measurement of odor-elicited emotions. Food Quality and Preference, 21, 938–947. Porcherot, C., & Issanchou, S. (1998). Dynamics of liking for flavoured crackers: test of predictive value of a boredom test. Food Quality and Preference, 9(1–2), 21–29. Porcherot, C., Kremer, S., Dreyfuss, L., & Almi, V. (2016). Presented at the European Sensory Network workshop on immersive tools, EuroSense Symposium, Dijon, France. Porcherot, C., Petit, E., Giboreau, A., Gaudreau, N., & Cayeux, I. (2015). Measurement of selfreported affective feelings when an aperitif is consumed in an ecological setting. Food Quality and Preference, 39, 277–284. Pouyet, V., Cuvelier, G., Benattar, L., & Giboreau, A. (2015). Influence of flavour enhancement on food liking and consumption in older adults with poor, moderate or high cognitive status. Food Quality and Preference, 44, 119–129. Schifferstein, H. N. J., Fenko, A., Desmet, P. M. A., Labbe, D., & Martin, N. (2013). Influence of package design on the dynamics of multisensory and emotional food experience. Food Quality and Preference, 27, 18–25. Scholz, J., & Smith, A. N. (2016). Augmented reality: designing immersive experiences that maximize consumer engagement. Business Horizons, 59, 149–161. Sester, C., Deroy, O., Sutan, A., Galia, F., Desmarchelier, J.-F., Valentin, D., et al. (2013). Having a drink in a bar: an immersive approach to explore the effects of context on drink choice. Food Quality and Preference, 28, 23–31. Thomson, D. M. H. (2016). Conceptual profiling. In H. Meiselman (Ed.), Emotion measurement (pp. 239–272). Duxford: Woodhead Publishing. Van Herpen, E., Van den Broek, E., Van Trijp, H. C. M., & Yu, T. (2016). Can a virtual supermarket bring realism into the lab? Comparing shopping behavior using virtual and pictorial store representations to behavior in a physical store. Appetite, 107, 196–207.

Personal and home care products in context

385

Witmer, B. G., & Singer, M. J. (1998). Measuring presence in virtual environments: a presence questionnaire. Presence: Teleoperators and Virtual Environments, 7, 225–240. Yip, B. C. B., & Man, D. W. K. (2013). Virtual reality-based prospective memory training program for people with acquired brain injury. NeuroRehabilitation, 32, 103–115.

Further Reading Porcherot, C., Delplanque, S., Ferdenzi, C., Gaudreau, N., & Cayeux, I. (2016). Studying emotions in practice: emotions of odors/personal and home care products. In H. L. Meiselman (Ed.), Emotion measurement (pp. 427–450). Duxford: Woodhead Publishing.

This page intentionally left blank

Beverages in context Sara Spinelli DAGRI, University of Florence, Florence, Italy

19.1

19

Context has many meanings

The scientific literature on context uses words such as “appropriateness,” “situation,” etc., that refer to different aspects of the issue (see Chapter 1 in this book). “Context” has in itself many meanings, from a “narrower” to a “wider” sense (Meiselman, 2007; see Table 19.1). If we consider the context of a study in which beverages are evaluated (in an experimental setup), at a narrower level we have several contexts as external frames created by the other stimuli (e.g., hedonic and contextual contrast), by the end anchors in scales, and in general by the experimental design (for details see Cardello, 2017; Lawless & Heymann, 2010). Contextual variables are also created by the language that is used in the experiment, within the definition of the task instructions and in the items the individuals are asked to evaluate. There may be an impact from the language used, e.g., colloquial, technical, but also the selection of the items incorporated within a questionnaire define a “context.” Thus, the participant’s response can be influenced by either the presence of items (e.g., synonyms) or the lack of items in the list that are considered more appropriate (see the dumping effect, Lawless & Heymann, 2010). Context is very important in reducing the ambiguity that is usual in a language and which indeed corresponds to its richness. A word has multiple meanings, which are activated depending on the context. In textual linguistics, two kinds of contexts that help to select the meaning activated against the virtual/potential meaning of a word were identified: – –

“co-text” is an intra-textual linguistic context: it is the sentence (the other words in the sentence) or the group of sentences within which a word is included. “context,” instead, refers to an extra-textual context, the setting/scenario in which the words should be interpreted (i.e., the situational context); for example, “sweet” has the double meaning of “sugary” and “lovable,” but in the context of the drinking experience of a beer, the contextual meaning activated is probably the former and not the latter.

In a questionnaire presented as a list of adjectives, the co-text is only constituted by the list format and the context is limited to the product under test. For this reason, it can be difficult for the respondent to “correctly” interpret the word (and “correctly” here means “in the sense intended by the researcher” who aims to “measure” something). In a wider sense, context may refer to the setting of the study. A study may be conducted in a laboratory, in a natural context for eating, e.g., a restaurant, home, or in a realistic context such as a room prepared to recall a situation such as a party or a bar. If we focus on natural contexts, in the sense of “situations” in which we have beverages, Context. https://doi.org/10.1016/B978-0-12-814495-4.00019-2 Copyright © 2019 Elsevier Inc. All rights reserved.

388

Context

Table 19.1 Beverages in context Contextual effects

Examples

External frames created by the other stimuli in the test Internal frames created by the end anchors in scales/instructional frames Language

Contrast/range mapping/frequency bias/ centering bias/transfera Definition of the range of the scale/ contraction bias/logarithmic biasa Type of language used: jargon, colloquial, technical, etc. Co-text (e.g., items in the questionnaires) Natural/naturalistic-immersive/laboratory/ central location test, evoked context

Setting of the study

a

Lawless and Heymann (2010) and Cardello (2017).

we may describe them in terms of social environment (are you drinking alone, with friends, with colleagues, with family, or with clients?); setting/environment (cafeteria, bar, restaurant, home, party, street, on the train, on a flight); action (are you drinking while eating, during a meal, during a break, while walking, while watching a movie, or during a conversation?); or, goal (are you drinking because you are thirsty, to have an opportunity to chat with friends, or to do something in a break during work?). The context of beverage consumption may be very heterogeneous and activate quite different meanings. It is not difficult to imagine that the experience of consuming a beverage in a different context (¼ situations) may be completely different, because the meaning associated with the beverage is different. The importance of integrating these “contextual” aspects in the study of product experience has been highlighted by many scholars in the sensory and consumer science fields (K€ oster, 2003; Koster & Mojet, 2018; Meiselman, 2006, 2007). In the last three decades, in almost every discipline within the human and social sciences, a debate about what constitutes a context (e.g., extraneous and confounding variables), and whether and how it can be considered as an object of analysis, has clearly highlighted the need for an integration of this issue in the research paradigm. In psychology, two views of the relationships between the individual and the situation have been identified. In one perspective, the situation causes the process of evaluation within the individual. This view of the situation as something external that triggers processes that are internal to the person places a boundary between the person and the situation, as if they are truly independent causes that can interact to produce the behavior in question. This view corresponds to what has been identified as a situational fallacy (K€ oster, 2003). In the other perspective, the person and situation are not separable sources of variation that interact and influence each other; rather, they help to constitute each other (Mesquita, Barrett, & Smith, 2010). This perspective is based on the concept of “ecological niche” developed in ecology. An organism’s ecological niche is not defined by the physical surroundings, but by what aspects of its environment are relevant to its actions and behaviors. Some aspects are physically present, but go unnoticed, depending on the individuals, because they are not

Beverages in context

389

“relevant” (not coherent with their interest or goals) for them. As noted by Barrett, Mesquita, and Smith (2010, p. 9): To borrow an example of the ecological niche from the evolutionary biologist Richard Lewontin (2000), two species of bird (phoebes and thrushes) live in exactly the same territory within the northeastern United States that includes both grasses and rocks, but whereas a phoebe’s niche includes grass to build nests, a thrush requires rocks to crack open seeds. Rocks are physically present for a phoebe but go unnoticed; the same is true for grass and thrushes. For humans, the psychological situation includes only those aspects of the physical surroundings […] that are relevant to the goals of the perceiver, so that within the exact same physical surrounding there exist different “situations” for different people (or for a single individual at different points in time).

Hence, while physical surroundings exist separately from observers, “situations”— in the sense intended here—do not. It is the person that determines the meaning of the physical surroundings (i.e., the psychological situation). Thus, in this perspective, even the situation is itself contextual, in part being determined by the person (who acts as a form of context). The implication of this view is that the situation is not a description of the physical properties of the environment, but it can be characterized as containing just those aspects that are relevant to the thoughts, feelings, and behaviors of that particular person at that particular point in time. What are the implications of such a view for the investigation of product experience, and in particular of the drinking experience? If we take an example such as different people having a drink in the physical environment of a bar, we may imagine that one individual is focusing their attention on the sensory characteristics of the cocktails they have just ordered (has something changed from the last time?), another is completely absorbed in a conversation with a friend and thus what they are drinking goes unnoticed, while a third individual is not paying very much attention to the conversation, but is instead observing the setting and the furniture, which reminds them of an experience during a trip to Spain some years earlier. All three persons are seated at the same table, but if asked about the situation in which they are having a drink, they would probably each describe it differently, or rather they would probably give value to different aspects of the situation that are salient for each of them. In this chapter, we first present the findings that have emerged from studies conducted in the last three decades with the objective of integrating context in the investigation of the drinking experience and the perception of beverages. Two main approaches have been identified, namely studies aimed at investigating how context contributes to shape expectations and responses for beverages (Section 19.2) and studies on the situational appropriateness (Section 19.3). Then we present a new original approach that fully adopts the perspective of the psychological niche described earlier, in which the mind of the individual determines the psychologically “active variables” of the situation so that a “situation” does not exist separately from the person, together with a case study on the coffee drinking experience.

390

19.2

Context

Context contributes to shape expectations and responses to beverages

A number of studies have focused on the influence of “situational factors,” along with sensory factors, on the experience of consuming a beverage. With “situational factors,” numerous variables in the drinking environment that influence the ease with which drinking is commenced, continued, or reiterated have been identified (Meiselman, Hirsch, & Popper, 1988). These include temperature, lighting, comfort, and accessibility. These effects may be mediated by the expectations generated by the situation in which the product is experienced. In fact, since product expectations are based on a person’s individual memory of the different situations in which they have encountered that product or a similar one (Koster & Mojet, 2018), the context contributes to shape expectations and responses to products. Several studies have investigated the expectations triggered by different situational contexts, from the environment (Stroebele & De Castro, 2004), to the information provided to the consumers (Caporale & Monteleone, 2004; Meillon, Urbano, Guillot, & Schlich, 2010), and the collocation of the beverage in a meal (see the studies on food-beverages pairings: Eschevins, Giboreau, Allard, & Dacremont, 2018; Paulsen, Rognsa˚, & Hersleth, 2015). Many situational factors have been shown to influence hedonic scores. For example, it has been observed that beverages are most preferred when presented at their usual (i.e., expected) temperature of consumption, a function primarily of culture (Zellner, Stewart, Rozin, & Brown, 1988). Several studies have compared the affective responses to different beverages in a laboratory setting and in a natural drinking context (see Table 19.2). Some studies found that hedonic ratings increased in a naturalistic and natural drinking context (Hersleth et al., 2003; Kim et al., 2016), supporting the findings already reported elsewhere for foods (Meiselman, Johnson, Reeve, & Crouch, 2000). Other studies did not find a difference between conditions in the liking ratings, but instead in emotional responses whether testing in a natural context (Danner et al., 2016) or using a written scenario (Dorado et al., 2016). In addition, studies that used a written scenario found that product discrimination was higher in the condition of evoked context in some cases (Dorado et al., 2016; Hein et al., 2010, 2012), but not in others (Hein et al., 2012; Lusk et al., 2015). These controversial findings suggest that it is very likely that it is not the introduction of situational factors in itself that has an impact on product evaluation, but rather the introduction of an “appropriate” drinking context. When the evoked context is not perceived as appropriate (e.g., in the case of the evoked scenario) or realistic (e.g., in the case of naturalistic context), there is no reason to expect that the evaluation of the product will be positively influenced. In a study that compared different products, including two beverages (sparkling water and a milk beverage), Boutrolle, Delarue, Arranz, Rogeaux, and K€ oster (2007) found that although the liking scores for these two beverages were higher in the home use test (HUT) compared to the central location test (CLT), the ordering of results was not different: the same sample was preferred in both conditions. Commenting on these results, the authors noted that in the case of the two

Type of beverage

Geographical location

Between/within subjects

Apple juices

New Zealand

Between subjects

Apple juices

Apple and blackcurrant juices

New Zealand

New Zealand

Between subjects

Between subjects

Situational factors

Effects

References

1. Sensory laboratory 2. Evoked context using a written scenario (written response):

Product discrimination was higher in the evoked condition

Hein, Hamid, Jaeger, and Delahunty (2010)

Best-worst scaling discriminated between the samples, while 9-point hedonic scale did not discriminate

Lusk, Hamid, Delahunty, and Jaeger (2015)

Hedonic ratings did not change in the different condition for apple juices, while they changed for blackcurrant juices (lower in context condition) Rank order for liking was similar in the two conditions

Hein, Hamid, Jaeger, and Delahunty (2012)

“Think about an occasion in which you want something refreshing to drink” Evoked context using a written scenario: “When having something refreshing to drink”: 1. 2. 1. 2.

Product discrimination was higher in the evoked condition for blackcurrant juices

391

Best-worst scaling 9-point hedonic scale Sensory laboratory Evoked context using a written scenario (written response): breakfast on a weekend morning 3. Evoked context using a written scenario (written response): watching a movie at the theater 4. Evoked context using a written scenario (written response): when having something refreshing to drink

Beverages in context

Table 19.2 Studies on beverages that take into account situational factors

Continued

392

Table 19.2 Continued Beers

France

Between subjects

United Kingdom

Between subjects

1. Naturalistic context: immersive bar—cold condition (using video clips, furniture, and music) 2. Naturalistic context: immersive bar—warm condition (using video clips, furniture, and music) Laboratory: 1. Without written scenario 2. With written scenario (“imagine you are having a beer,” with focus on location, social setting, and time)

Sester et al. (2013)

Liking: No difference in magnitude between the two conditions, slight increase in product discrimination in the “with scenario” condition Emotions: Emotional responses were higher for positive emotions and lower for negative emotions in the “with scenario” condition No condition could be identified as more discriminating in terms of emotional responses Increase in product discrimination in familiarity in the “with scenario” condition

Dorado, Chaya, Tarrega, and Hort (2016)

Context

Difference in choice of one of five beers

United States

South Korea

Within subjects

Between subjects

1. Laboratory (under red light) 2. Virtual coffeehouse: Evocation of a natural drinking situation using video footage depicting sights and sounds regularly experienced in a coffeehouse, typical furniture, and a cinnamon roll aroma Laboratory (individual booths): 1. With evocation 2. Without evocation Simulated cafe (using furniture and oral description)

Iced-coffee

France

Within subjects

3. With evocation 4. Without evocation Laboratory (common condition: hot temperature: 22°C): 1. Classical laboratory 2. Evocation of a natural drinking situation of hot temperature in a laboratory using visual, auditory, and olfactory stimuli

Differences in preference order were identified in the two conditions Hedonic data collected in the virtual coffeehouses were more discriminating

Bangcuyo et al. (2015)

An effect of the environment, but not of evocation, on liking was found The impact of situational factors was higher for consumers less involved in the product

Kim, Lee, and Kim (2016)

Hedonic ratings were lower in the natural contexts compared to the laboratory In the laboratory, the two conditions gave very similar results

Petit and Sieffermann (2007)

393

Continued

Beverages in context

Coffee

394

Table 19.2 Continued Natural consumption situation:

Wines

United States

Within subjects

3. Office meeting room 4. Cafeteria Laboratory: 1. Individual sensory booths without food 2. Individual sensory booths with food Naturalistic consumption situation with creation of a social atmosphere with music, flowers Participants (eight per time) were encouraged to talk:

Wines

Australia

Within subjects

3. Reception without food 4. Reception with food 1. Sensory laboratory 2. Natural drinking environment: Restaurant 3. Natural drinking environment: Home

Hedonic ratings were higher in the reception room than in the individual booths The positive effect of serving food together with the wines was larger in the reception room than in the booths in the sensory laboratory

Hersleth, Mevik, Næs, and Guinard (2003)

No effect of context on liking but effect on emotional responses, with some positive emotions higher in the restaurant condition

Danner et al. (2016)

Context

Beverages in context

395

beverages, the usual consumption situation was between meals, and not during a meal, thus making the evaluation in a CLT less defective in terms of situational factors. For the other product included in the study, the salted crackers that are usually consumed in a social situation, the HUT was more discriminating. The authors suggested that the usual context of consumption has a leading role in the hedonic evaluation of food products, and recommended using a HUT whenever the product consumption is expected to be associated with situational factors not replicable in the CLT (e.g., social aspects).

19.2.1 Methodological considerations when evaluating the choice of a natural/naturalistic/virtual context for beverage evaluation The importance of moving product evaluations increasingly away from laboratories to natural contexts, for example with the HUT (Meiselman, 1992, 2013), and of opting for situational research that integrates many different influences and their developing interaction over time (K€ oster, 2003; Koster & Mojet, 2018) has been widely highlighted. The choice of the specific setting for the test is crucial (for an up-to-date overview of the pros and cons of each choice, see Jaeger & Porcherot, 2017). While natural contexts (such as restaurant, bars, home) have higher external validity, experimental control is reduced. The use of so-called naturalistic contexts (immersive context) in which a bar or a restaurant is recreated as a way of manipulating environmental variables is an alternative, but the difficulties in making it credible and in activating what in theater is called “suspension of disbelief” should not be underestimated. The fact of asking individuals to go into a place that it is not a bar but to behave such as if it was, does not ensure that the attention of the participants will be attracted by features that the experimenter is interested in. It is very important to avoid perception being focused on the fact that the situation is “fictional.” This is even more important in virtual environments, where new digital devices usually allow one to recall specific situations visually, but also with auditory, tactile, and olfactory cues. Because such environments are novel and interactive, it may be that attention is not attracted by the object of the test but by the setting in itself (see Jaeger & Porcherot, 2017). An interesting possibility to ensure control, while at the same time keeping some features of the natural context, may be the experimental restaurant (see for example the Institut Paul Bocuse in Lyon or the Restaurant of the Future in Wageningen; Giboreau, 2018). Even the use of an evoked context triggered by descriptions (written scenario), pictures or using story-telling, may be beneficial. The response to a written scenario, so that respondents must write what they are imagining, proved to be useful to evaluate fruit juices and beers (Dorado et al., 2016; Hein et al., 2010, 2012; Lusk et al., 2015). However, further studies are needed to better understand how the approach may be improved and what the specificities of different types of beverages are. A path to further explore may be the analysis of the responses to the evoked scenario; see Spinelli et al., 2017 and Section 19.4 in this chapter.

396

19.3

Context

Beverages and situational appropriateness

Another route for integrating the context in product evaluation is the investigation of the situational appropriateness (see also Giacalone, Chapter 6 in this book). Traditionally, situational appropriateness has been evaluated by means of the item-by-use appropriateness approach, a method that was developed by Schutz (1988) and Cardello and Schutz (1996). In this method, the uses, situations, or attributes to be evaluated are obtained by eliciting information on when it would be appropriate to use/consume the food product. Individuals are asked to rate appropriateness of each food item for each of the various uses, attributes, and situations in a matrix format in a self-administered questionnaire using a scale ranging from “never appropriate” (1) to “always appropriate” (7). In the original version of the method, these uses include various aspects, depending on the product, such as time of the day, occasion, location where served, physiological states, how used, psychological characteristics, and person served. Beer experience, including situational factors, has been extensively investigated in the last decade. Giacalone et al. (2015) applied an item-by-use appropriateness approach to beers and found an association between the situational appropriateness and the familiarity of the beers. Familiar beers were considered appropriate mostly for refreshments and while attending sport events, while novel ones were perceived as more self-indulgent and appropriate for dining events and special occasions. Interestingly, the set of four studies consistently showed that familiar beers were perceived as more versatile (appropriate to a larger number of situations), while unfamiliar beers were considered appropriate only for a specific situation. As the authors suggest, this link may be due to previous experiences with the product, while figuring out situations of use of an unfamiliar product is much more difficult. Further studies conducted on beers integrating the measure of the situational appropriateness with a wider group of measurements, including affective, emotional, and attitudinal measures confirmed this link with familiarity (Cardello et al., 2016) and, in addition, with complexity; Jaeger et al. (2017) found that well-liked beers with high drinkability were also more versatile, with high perceived situational appropriateness for social occasions and when together with friends and family, while more complex beers were found to be more appropriate when people want “something different” and to taste with friends interested in beer. A further extension of this concept of situational appropriateness has also been involved in developing concepts for new products. The integration of the contexts of use at this stage (concept development/concept refinement) has been stressed by different scholars (Christensen & Raynor, 2003; Moskowitz, 2012). The information about how the product is used, which situation is considered more appropriate and which emotional links it activates in different situations may be usefully collected at the concept stage of the innovation process. Lieberman, Moskowitz, and Moskowitz (2012) developed an approach (Drink it!) based on conjoint analysis, in which ideas related to different aspects of the experience (benefits, emotions, situations, sensory features, etc.) are mixed and matched by a computer program in order to create brief, easy to understand concepts about beverages. These concepts are then

Beverages in context

397

evaluated by consumers and the analysis conducted at an individual level reveals what particular ideas drive interest for the respondents. The approach is based on a segmentation that identifies different “mind sets” towards the product category under investigation (e.g., white wines), namely different patterns of response to the white wine concept.

19.3.1 Methodological considerations when evaluating situational appropriateness of beverages An important aspect for this approach aimed at measuring the situational appropriateness to be effective is the selection of the contexts. This is a crucial stage because the effectiveness of a study is very much linked to whether participants understand the contexts and recognize them as familiar. Generally, the selection is based on previous studies as a starting point, but, as recommended by Jaeger and Porcherot (2017), product-specific consumer knowledge is paramount and preliminary research is a must. A pilot test is always recommended (Giacalone et al., 2015). Traditionally sentences were used to describe contexts, but photographs have also been recently employed in addition (Giacalone et al., 2015). Pilot studies are recommended to validate the correspondence between the description and the image. The use of images instead of words may be more impactful; however, it should be considered that an image has a lower level of generality compared to words. For example, when breakfast is recalled using the phrase “during breakfast,” the participant is free to imagine their own breakfast, from toast and coffee only to a rich meal with eggs, bacon, bread, etc. Selecting a particular image may mean fixing and restricting the meaning for a participant. This may be important in some cases in which the aim is to focus on specific aspects of the situation, but it may also create a feeling of unfamiliarity. Thus, a respondent might perceive the product as appropriate for breakfast but may have difficulty in identifying themselves in the photograph that it is shown because the environment/the objects shown are unfamiliar. This aspect is an important consideration when planning a study aimed at measuring situational appropriateness.

19.4

Individual differences in preferred context to consume a product: A case study on coffee

Food preferences are related to multiple factors, including genetics, phenotype, physiological, and psychological and socio-cultural factors (Prescott, 2012; Rozin, 2006). These same factors also influence the ways in which individuals prefer to consume the foods they like. The most appropriate context for having a beverage typically corresponds to the context in which the beverage is most liked. In contrast, if a food or beverage is served in an inappropriate situation, its liking and consumption may suffer (Schutz, 1988; Schutz, 1994). It is currently unclear how individual differences might relate to such preferred contexts. It is inevitable that the most appropriate (preferred) context will vary between individuals, as might the number and variety of appropriate consumption contexts, at least for some products. If we adopt the perspective

398

Context

described earlier (Section 19.1), that within a common physical surrounding there exists different “situations” for different people (or for a single individual at different points in time) and that the “active ingredient” or psychological feature of the situation is individually determined, we should consider that a “situation” does not exist separately from a person (Barrett, 2006). It is therefore plausible that differences in preferred context might be related to differences in physiological and psychological variables of the individual. Recently, Jaeger et al. (2013) demonstrated that genotype can influence whether individuals perceive an odor as appropriate for use in specific situations. The study showed that genotype for sensitivity to an odor explains aspects of perception such as perceived odor experience and emotional association, as well as consumption related responses, product appropriateness, and choice of stimuli containing the focal odor. Such findings encourage exploration of the linkages between genotype, physiological variables, perception, and behavioral measures to better understand food choice and preferences. One way of examining the impact of consumption contexts on preferences or choices is through experimentally evoking the details of the consumption situation in a way that the respondent can identify with. Adopting this approach allows investigation not only of the effects of the physical environment, but also of a mix of context features that are emotionally, individually, socially, and culturally determined. Several studies have shown that a written scenario can be useful to evoke a context effectively (see Section 19.2; Belk, 1975; Hansen, 2005; Jaeger & Meiselman, 2004; Hein et al., 2010; Hersleth, Monteleone, Segtnan, & Næs, 2015), by allowing individuals to link the evoked context with their own memories. Some studies have taken an additional step towards ensuring context effectiveness by asking consumers to provide input to the written scenario used to evoke a context (Hein et al., 2010, 2012; Qiu & Yeung, 2008). This has been said to increase the availability of such events that exist in the memory (Schwarz & Clore, 1983). In addition, this written input can also provide qualitative data regarding aspects of an individual consumer’s contexts that would allow further understanding of how the context was created (Hein et al., 2010). Descriptions provided by consumers in their own words may thus be an important source of information about the aspects they value in the situation. However, this aspect has received little attention up to now. With this in mind, we developed an approach to investigate the relationship between individual differences in taste responsiveness and physiological indices and the preferred product context, using the example of coffee (Spinelli et al., 2017). Coffee was chosen for this study because it is a universally popular product, often consumed more than once a day, and in a variety of different contexts. In many countries, coffee is consumed often at breakfast, but also in the middle of the morning or afternoon (the “coffee break”), or after a meal. Coffee contains caffeine, which has strong pharmacological and physiological effects, including cardiovascular, respiratory, renal, and smooth muscle effects, as well as effects on mood, memory, alertness, and physical and cognitive performance (McLellan, Caldwell, & Lieberman, 2016). Thus, the preference for a specific coffee drinking context may be related not only to the social features characterizing the situation, but also for sensory enjoyment or for its alerting effects (Labbe, Ferrage, Rytz, Pace, & Martin, 2015). Coffee is traditionally

Beverages in context

399

consumed to wake up, to gain energy, to help digestion, and to have a break to recover from work, and it has been demonstrated that coffee consumption can condition flavor preferences in contexts in which it is used to alleviate caffeine abstinence (Tinley, Yeomans, & Durlach, 2003). Recent studies have also shown the importance of individual variations in caffeine metabolism rate on responsiveness to caffeine bitterness, and the effect of this link on coffee perception and consumption (Masi, Dinnella, Monteleone, & Prescott, 2015; Masi, Dinnella, Pirastu, Prescott, & Monteleone, 2016). Moreover, individual differences in the density of lingual fungiform papillae (FP), in the sensitivity to the bitterness of the compound 6-n-propylthiouracil (PROP) (Masi et al., 2015), and in responsiveness to astringent stimuli (Dinnella, Recchia, Tuorila, & Monteleone, 2011; Fleming, Ziegler, & Hayes, 2016) have all been found to influence the perception of coffee bitterness and the use of sweeteners in coffee to moderate this bitter taste. The main question behind this study was whether the preference for a specific context of coffee consumption was related to individual differences, represented by variations in taste responsiveness (PROP status) and physiological variables (fungiform papillae density and caffeine metabolism rate). Italian regular coffee consumers were asked to describe their own preferred situation for having coffee. In fact, while many people consume more than one coffee per day, it is common to have a preferred (one or more than one) context to have coffee. The rationale behind this study was that the choice of one setting instead of another can provide valuable information about the most favored aspects of the coffee experience, that may reflect individual differences in taste sensitivity and caffeine metabolism. If each situation (context) is defined by the meanings that individuals attribute to their surroundings (K€oster & Mojet, 2007) studying the situations that are preferred by the individuals to consume coffee should reveal meaningful aspects of their experience. Moreover, because these meanings vary from individual to individual, we assume that taste responsiveness and physiological variables may play a role in this variation. If, for example, the preferred context described by an individual highlights very precisely how the coffee should be in terms of temperature, taste, etc., this might reflect a preference that is driven by the sensory characteristic of coffee which, in turn, may be determined by the sensitivity to bitterness. Conversely, if the preferred context in which to have coffee is described as in the middle of the afternoon, chatting with friends during a break from work, this might reflect a situation where the preference for coffee was developed in association with its social function, perhaps independently of its taste.

19.4.1 The design of the study The participants, all consumers of at least one cup of coffee per day, were asked to answer the open-ended question: “Describe the moment of the day in which you prefer to have coffee.” Subjects were encouraged to provide details in terms of time of the day, place, and company. This was done to create an immersive scenario (evoked context) prior to the sensory evaluation of six samples of coffee, similar to the procedure suggested by Hein et al. (2010). We will report here only the data related to the participant characteristics and the text analysis of the evoked scenario. In addition

400

Context

to the usual procedure (Hein et al., 2010) in which the written responses to the scenario are used to promote the “immersion” of the respondents in the situation evoked, the responses to the evoked context, namely the descriptions provided by the participants, were analyzed using text analysis. Subjects (n ¼ 119) were characterized in terms of bitterness sensitivity, caffeine metabolism rate, attitudes, and food behavior (see Masi et al., 2015, 2016; Spinelli et al., 2017). The analysis of consumer responses was conducted by combining the statistical tools provided by the software for text analysis T-LAB 9.1 (Italy) with semiotic analysis, a methodological approach aimed at investigating the meaning-making of texts (Greimas & Courtes, 1983; Rastier, 2015; Violi, 2001). We proposed an original approach to conduct a guided preprocessing based on a semiotic analysis on a subset of texts (10%). This way, it was possible to build a “customized dictionary” based on a preliminary semiotic analysis on a subset (i.e., grouping words based on semantic criteria; see Spinelli et al., 2017 for details). A thematic analysis based on an unsupervised clustering (bisecting K means) was conducted on the texts produced by consumers to describe their preferred context for coffee consumption using the software T-LAB 9.1 (Lancia, 2012, 2015) (see Spinelli et al. (2017) for details). The analysis allowed the identification of the most frequent themes and an examination of their relationship with the subjects’ individual variables.

19.4.2 Thematic analysis of the open-ended questions: Preferred contexts and habits, socio-demographics, and physiological variables Our findings show that individuals preferred different contexts to have coffee during the day. Four thematic clusters were identified, focused on different aspects of the coffee experience, representing respectively 24.6%, 25.4%, 30.6%, and 19.4% of the subjects (we will refer to these as “EXPERIENCE” themes). Each theme was labeled according to the focus on the preferred context, as follows: –

– – –

Theme 1: SAVORING: focus on an experience of relaxation, comfort (with characteristic terms relating to /relax/, /savoring/, /sensation/, /reading/, /moka/—the Italian coffee machine used at home) and attention to the sensory properties of coffee (/bitter/, /with added sugar/, /temperature of serving/). Theme 2: BREAK: focus on break function of the coffee (e.g., break from /work/) and on the social dimension (/chatting/); they describe their preferred context as coffee had in the afternoon with friends to recover from work. Theme 3: AFTER A MEAL: focus on the situation; coffee taken after meal, associated with a social ritual (coffee taken in company with friends or colleagues). Favorite coffee is also described as being spiked with liquor by some people in this cluster. Theme 4: AWAKENING; focus on a “waking up” stimulating function and on the tasting experience (coffee drunk alone, without company); and food eaten with coffee during breakfast (e.g., biscuits, croissants).

Fig. 19.1 shows the results of correspondence analysis performed on the terms x theme contingency table. According to the thematic analysis, the descriptions were clearly

Beverages in context –0.5

Y = Fact 2 (27.98%)

–1.0

401 0.0

0.5

1.0

1.5

1.5

bed before 1.0

4 AWAKENING

eating

tiredness

awakening biscuit & co

Y - axis

together

milk comfort

morning

waiting alone sensation

0.5

kitchen important

smoking

espresso after a meal company work • PROP ST ritual 3 AFTER A bar MEAL 2 BREAK • PROP NT break

0.0

spiked • HFP excuse

coffee temperature bitter 1 SAVOURING • HCMI relax • PROP MT reading • LEP

• LCMI chatting half morning/afternoon

brief

recovering afternoon

savouring sugar

add

–0.5

homemade

X - axis

–1.0

X = Fact 1 (44.59%)

Fig. 19.1 Thematic analysis. Characteristic terms of each theme are represented with the corresponding symbol: 1 SAVORING; 2 BREAK; ⧫3 AFTER A MEAL; ★ 4 AWAKENING. LCMI, slow caffeine metabolizers; HCMI, fast caffeine metabolizers; LFP, low fungiform papillae density; HFP, high fungiform papillae density; PROP NT, nontaster; MT, medium tasters; ST, supertasters.

▪

separated along the first axis based on the focus on the sensory experience (SAVORING) or on the social situation (AFTER A MEAL; BREAK). On the second dimension, the main differences are registered between theme 4 (AWAKENING) and all the other themes. We can hypothesize that these different themes reflect different ways to interpret the rewarding value of coffee: (1) for its sensory characteristics and for representing an experience to savor alone (SAVORING); (2) for its social value (creating a moment to share together with other people) and its alerting effect of recovering during work—improving cognitive performances (BREAK); (3) for its social value (creating a moment to share together with other people) and its alerting effect after a meal and during digestion—improving physical performance (AFTER A MEAL); and (4) for its psychoactive and alerting effect after sleeping

402

Context

(AWAKENING). Physiological variables such as fungiform papillae density and caffeine metabolism index play a role in interpreting the rewarding value of coffee. For individuals more sensitive to taste (high density of fungiform papillae on the tongue; HFP), the coffee seems more rewarding for its psychoactive and alertness effect after sleeping, than for its sensory properties, compared to LFP (low density of fungiform papillae). Conversely, individuals less sensitive to taste (LFP) find coffee more rewarding for its sensory and emotional enjoyment and less rewarding for its stimulating function after sleeping. These results could be related to the different degree of sensitivity between the two groups, since there is evidence that fungiform papillae density correlates with taste sensitivity (Delwiche, Buletic, & Breslin, 2001). Masi et al. (2015) reported that HFP individuals rated the bitterness of caffeine and quinine solutions and sourness in coffee as stronger than did LFP individuals. For this more sensitive group, coffee may be less appreciated as an experience to savor (i.e., for its sensory properties). Fast caffeine metabolizers, who consume more coffee daily, seem to find coffee more rewarding both for its sensory and emotional enjoyment, while for them the social and the psychoactive functions to recover after a break are less rewarding than for slow metabolizers. Slow metabolizers, who consume less coffee daily, seem to find more rewarding the social aspects related to coffee drinking experience, e.g. chatting with friends and colleagues, compared to the fast metabolizers.

19.5

Conclusions

In this chapter, attention has been drawn to the approaches used to investigate contextual factors in the beverage consumption experience. The adoption of a perspective on situation as co-created by the individual and the environment may further assist in the understanding of the role of the context in product experience. The approaches that rely on the evocation of context have been presented and a situational approach based on the analysis of written responses to the evoked context of preferred situation to have coffee in a day has been proposed. These findings suggest an interesting pathway to explore, that of the link between the contexts preferred by subjects to have a product and their individual differences. Further studies in this direction could increase understanding of food preferences, through investigating not only the dimensions of the products (i.e., the foods we like) but also the dimensions of the context (i.e., how we like the foods we like). Both expressions seem to be related to our biological and physiological background, beyond an undiscussed role of the cultural dimension and of psychological factors. The proposed approach could be easily applied to gain a deeper understanding of product experience, investigating the preferred contexts in relationship with individual differences. Different variables, not only related to physiological variables and taste responsiveness, but also psychological factors, sociodemographics, and affective and sensory responses to products may be of interest. The proposed approach showed promising results to link individual differences to textual data. The analysis of responses to open-ended questions to explore the contexts was found to be effective, allowing the collection of information about the most

Beverages in context

403

important features valued by individuals. Unlike closed-ended questions, open-ended questions help to avoid a focusing of the attention of the respondents on specific aspects predetermined by the researcher. This is particularly important in the study of context in order to understand the meaning assigned by individuals to the situation. Research employing statements made by consumers can be limited if the dataset is not sufficiently broad. However, progress in text analysis can provide tools that allow investigation of this issue in more depth, considering that an access to how people experience and prefer to experience foods expressed in their own words is needed. The proposed approach, which includes semiotic analysis on a subset of text, allows the development of a hypothesis that can guide pre-processing, solving some practical issues highlighted by several scholars. Particularly, this helps in reducing the arbitrariness of the pre-processing step because it allows the researcher to base the preprocessing on an interpretative hypothesis rested on the semantic specificities of the texts that are analyzed (e.g., meaning of the words in this context, especially when the meaning is figurative and metaphorical). Context research can benefit from an improvement in text analysis, which will allow an easier and more effective treatment of textual data directly provided by the respondents.

References Bangcuyo, R. G., Smith, K. J., Zumach, J. L., Pierce, A. M., Guttman, G. A., & Simons, C. T. (2015). The use of immersive technologies to improve consumer testing: the role of ecological validity, context and engagement in evaluating coffee. Food Quality and Preference, 41, 84–95. Elsevier Ltd https://doi.org/10.1016/j.foodqual.2014.11.017. Barrett, L. F. (2006). Valence is a basic building block of emotional life. Journal of Research in Personality, 40(1), 35–55. https://doi.org/10.1016/j.jrp.2005.08.006. Barrett, L. F., Mesquita, B., & Smith, E. R. (2010). The context principle. In B. Mesquita, L. F. Barrett, & E. R. Smith (Eds.), The mind in context (pp. 1–22). New York, London: The Guilford Press. Belk, R. W. (1975). Situational variables and consumer-behavior. Journal of Consumer Research, 2, 157–164. Boutrolle, I., Delarue, J., Arranz, D., Rogeaux, M., & K€oster, E. P. (2007). Central location test vs. home use test: contrasting results depending on product type. Food Quality and Preference, 18(3), 490–499. Elsevier https://doi.org/10.1016/J.FOODQUAL.2006.06.003. Caporale, G., & Monteleone, E. (2004). Influence of information about manufacturing process on beer acceptability. Food Quality and Preference, 15(3), 271–278. https://doi.org/ 10.1016/S0950-3293(03)00067-3. Cardello, A. V. (2017). Hedonic scaling: assumptions, contexts and frames of reference. Current Opinion in Food Science, 15, 14–21. Elsevier https://doi.org/10.1016/J.COFS.2017. 05.002. Cardello, A. V., Pineau, B., Paisley, A. G., Roigard, C. M., Chheang, S. L., Guo, L. F., et al. (2016). Cognitive and emotional differentiators for beer: an exploratory study focusing on “uniqueness” Food Quality and Preference, 54, 23–38. https://doi.org/10.1016/ j.foodqual.2016.07.001. Cardello, A. V., & Schutz, H. G. (1996). Food appropriateness measures as an adjunct to consumer preference/acceptability evaluation. Food Quality and Preference, 7(3–4), 239–249. https://doi.org/10.1016/S0950-3293(96)00012-2.

404

Context

Christensen, C. M., & Raynor, M. E. (2003). The innovator’s solution : Creating and sustaining successful growth. Harvard Business School Press. Danner, L., Ristic, R., Johnson, T. E., Meiselman, H. L., Hoek, A. C., Jeffery, D. W., et al. (2016). Context and wine quality effects on consumers’ mood, emotions, liking and willingness to pay for Australian Shiraz wines. Food Research International. 89, https://doi. org/10.1016/j.foodres.2016.08.006. Delwiche, J. F., Buletic, Z., & Breslin, P. A. S. (2001). Relationship of papillae number to bitter intensity of quinine and PROP within and between individuals. Physiology and Behavior, 74(3), 329–337. https://doi.org/10.1016/S0031-9384(01)00568-6. Dinnella, C., Recchia, A., Tuorila, H., & Monteleone, E. (2011). Individual astringency responsiveness affects the acceptance of phenol-rich foods. Appetite, 56(3), 633–642. https://doi. org/10.1016/j.appet.2011.02.017. Dorado, R., Chaya, C., Tarrega, A., & Hort, J. (2016). The impact of using a written scenario when measuring emotional response to beer. Food Quality and Preference, 50, 38–47. Elsevier Ltd https://doi.org/10.1016/j.foodqual.2016.01.004. Eschevins, A., Giboreau, A., Allard, T., & Dacremont, C. (2018). The role of aromatic similarity in food and beverage pairing. Food Quality and Preference, 65, 18–27. Elsevier https:// doi.org/10.1016/J.FOODQUAL.2017.12.005. Fleming, E. E., Ziegler, G. R., & Hayes, J. E. (2016). Investigating mixture interactions of astringent stimuli using the isobole approach. Chemical Senses, 41(7), 601–610. https:// doi.org/10.1093/chemse/bjw064. Giacalone, D., Frøst, M. B., Bredie, W. L. P., Pineau, B., Hunter, D. C., Paisley, A. G., et al. (2015). Situational appropriateness of beer is influenced by product familiarity. Food Quality and Preference, 39, 16–27. Elsevier Ltd https://doi.org/10.1016/j.foodqual. 2014.06.012. Giboreau, A. (2018). Situational factors and the design of in situ evaluations. Methods in consumer research (pp. 109–116). Vol. 2(pp. 109–116). Elsevier. https://doi.org/10.1016/ B978-0-08-101743-2.00005-4. Greimas, A. J., & Courtes, J. (1983). Semiotics and language: An analytical dictionary. Bloomington and Indianapolis: Indiana University Press. Hansen, T. (2005). Understanding consumer perception of food quality: the cases of shrimps and cheese. British Food Journal, 107(7), 500–525. https://doi.org/10.1108/ 00070700510606909. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2010). Application of a written scenario to evoke a consumption context in a laboratory setting: effects on hedonic ratings. Food Quality and Preference, 21(4), 410–416. Elsevier Ltd https://doi.org/10.1016/ j.foodqual.2009.10.003. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2012). Effects of evoked consumption contexts on hedonic ratings: a case study with two fruit beverages. Food Quality and Preference, 26(1), 35–44. Elsevier Ltd https://doi.org/10.1016/j.foodqual.2012.02.014. Hersleth, M., Mevik, B. H., Næs, T., & Guinard, J. X. (2003). Effect of contextual factors on liking for wine—use of robust design methodology. Food Quality and Preference, 14(7), 615–622. https://doi.org/10.1016/S0950-3293(02)00190-8. Hersleth, M., Monteleone, E., Segtnan, A., & Næs, T. (2015). Effects of evoked meal contexts on consumers’ responses to intrinsic and extrinsic product attributes in drycured ham. Food Quality and Preference, 40, 191–198. https://doi.org/10.1016/ j.foodqual.2014.10.002. Jaeger, S. R., Cardello, A. V., Chheang, S. L., Beresford, M. K., Hedderley, D. I., & Pineau, B. (2017). Holistic and consumer-centric assessment of beer: a multi-measurement approach. Food Research International, 99, 287–297. https://doi.org/10.1016/j.foodres.2017.05.004.

Beverages in context

405

Jaeger, S. R., McRae, J. F., Bava, C. M., Beresford, M. K., Hunter, D., Jia, Y., et al. (2013). A mendelian trait for olfactory sensitivity affects odor experience and food selection. Current Biology, 23(16), 1601–1605. https://doi.org/10.1016/j.cub.2013.07.030. Jaeger, S. R., & Meiselman, H. L. (2004). Perceptions of meal convenience: the case of at-home evening meals. Appetite, 42(3), 317–325. https://doi.org/10.1016/j.appet. 2004.01.005. Jaeger, S. R., & Porcherot, C. (2017). Consumption context in consumer research: methodological perspectives. Current Opinion in Food Science, 15, 30–37. https://doi.org/10.1016/ j.cofs.2017.05.001. Kim, S. E., Lee, S. M., & Kim, K. O. (2016). Consumer acceptability of coffee as affected by situational conditions and involvement. Food Quality and Preference, 52, 124–132. Elsevier Ltd https://doi.org/10.1016/j.foodqual.2016.04.008. K€ oster, E. P. (2003). The psychology of food choice: some often encountered fallacies. Food Quality and Preference, 14(5–6), 359–373. https://doi.org/10.1016/S0950-3293(03) 00017-X. K€ oster, E. P., & Mojet, J. (2007). Theories of food choice development. In L. Frewer & H. C. M. van Trijp (Eds.), Understanding consumers of food products (pp. 93–124). Woodhead Publishing. https://doi.org/10.1533/9781845692506.1.93. Koster, E., & Mojet, J. (2018). Complexity of consumer perception: thoughts on pre-product launch research. G. Ares & P. Varela (Eds.), Methods in consumer research (pp. 24–45). New approaches to classic methods: Vol. 1(pp. 24–45). Woodhead Publishing. Labbe, D., Ferrage, A., Rytz, A., Pace, J., & Martin, N. (2015). Pleasantness, emotions and perceptions induced by coffee beverage experience depend on the consumption motivation (hedonic or utilitarian). Food Quality and Preference, 44, 56–61. https://doi.org/ 10.1016/j.foodqual.2015.03.017. Lancia, F. (2012). T-LAB pathways to thematic analysis, www.tlab.it. Available from: http:// www.mytlab.com/tpathways.pdf. Lancia, F. (2015). User manual T-Lab 9.1. Tools for text analysis. Available from: http://tlab.it/ en/download.php. Lawless, H. T., & Heymann, H. (2010). Context effects and biases in sensory judgment. In H. T. Lawless & H. Heymann (Eds.), Sensory evaluation of food. Principles and practices, sensory evaluation of food–principles andpractices (pp. 203–225). New York, NY: Springer. https://doi.org/10.1007/978-1-4419-6488-5_9. Lieberman, L. E., Moskowitz, D. I., & Moskowitz, H. R. (2012). Consumer research in the wine industry: new applications of conjoint measurement. In Alcoholic beverages (pp. 395– 435). Elsevier. https://doi.org/10.1533/9780857095176.4.395. Lusk, K. A., Hamid, N., Delahunty, C. M., & Jaeger, S. R. (2015). Effects of an evoked refreshing consumption context on hedonic responses to apple juice measured using best–worst scaling and the 9-pt hedonic category scale. Food Quality and Preference. https://doi.org/ 10.1016/j.foodqual.2015.01.007. Masi, C., Dinnella, C., Monteleone, E., & Prescott, J. (2015). The impact of individual variations in taste sensitivity on coffee perceptions and preferences. Physiology and Behavior, 138, 219–226. Elsevier Inc. https://doi.org/10.1016/j.physbeh.2014.10.031 Masi, C., Dinnella, C., Pirastu, N., Prescott, J., & Monteleone, E. (2016). Caffeine metabolism rate influences coffee perception, preferences and intake. Food Quality and Preference, 53, 97–104. Elsevier Ltd https://doi.org/10.1016/j.foodqual.2016.06.002. McLellan, T. M., Caldwell, J. A., & Lieberman, H. R. (2016). A review of caffeine’s effects on cognitive, physical and occupational performance. Neuroscience and Biobehavioral Reviews, 71, 294–312. Elsevier Ltd https://doi.org/10.1016/j.neubiorev.2016.09.001.

406

Context

Meillon, S., Urbano, C., Guillot, G., & Schlich, P. (2010). Acceptability of partially dealcoholized wines – measuring the impact of sensory and information cues on overall liking in real-life settings. Food Quality and Preference, 21(7), 763–773. Elsevier https://doi.org/10.1016/J.FOODQUAL.2010.07.013. Meiselman, H. L. (1992). Methodology and theory in human eating research. Appetite, 19(1), 49–55. https://doi.org/10.1016/0195-6663(92)90235-X. Meiselman, H. L. (2006). The role of context in food choice, food acceptance and food consumption. In R. Shepherd & M. Raats (Eds.), The psychology of food choice (pp. 179– 199). Wallingford: CABI. https://doi.org/10.1079/9780851990323.0179. Meiselman, H. (2007). 3 – The impact of context and environment on consumer food choice. In L. Frewer & H. C. M. van Trijp (Eds.), Understanding consumers of food products (pp. 67–92). Cambridge,England: Woodhead Publishing Limited. https://doi.org/ 10.1533/9781845692506.1.67. Meiselman, H. L. (2013). The future in sensory/consumer research: evolving to a better science. Food Quality and Preference, 27(2), 208–214. Elsevier Ltd https://doi.org/10.1016/j. foodqual.2012.03.002. Meiselman, H. L., Hirsch, S., & Popper, R. (1988). Sensory, hedonic and situational factors in food acceptance and consumption. In D. M. H. Thomson (Ed.), Food acceptability (pp. 77– 87). London: Elsevier Applied Science. Meiselman, H. L., Johnson, J. L., Reeve, W., & Crouch, J. E. (2000). Demonstrations of the influence of the eating environment on food acceptance. Appetite, 35(3), 231–237. Academic Press https://doi.org/10.1006/APPE.2000.0360. Mesquita, B., Barrett, L. F., & Smith, E. R. (2010). The mind in context. Guilford Press. Moskowitz, H. R. (2012). “Mind genomics”: the experimental, inductive science of the ordinary, and its application to aspects of food and feeding. Physiology & Behavior, 107 (4), 606–613. Elsevier https://doi.org/10.1016/J.PHYSBEH.2012.04.009. Paulsen, M. T., Rognsa˚, G. H., & Hersleth, M. (2015). Consumer perception of food–beverage pairings: the influence of unity in variety and balance. International Journal of Gastronomy and Food Science, 2(2), 83–92. Elsevier https://doi.org/10.1016/J.IJGFS.2014.12. 003. Petit, C., & Sieffermann, J. M. (2007). Testing consumer preferences for iced-coffee: does the drinking environment have any influence? Food Quality and Preference, 18(1), 161–172. https://doi.org/10.1016/j.foodqual.2006.05.008. Prescott, J. (2012). Taste matters. Why we like the foods we do. London: Reaktion Books. Qiu, C., & Yeung, C. W. M. (2008). Mood and comparative judgment: does mood influence everything and finally nothing? Journal of Consumer Research, 34(5), 657–669. https:// doi.org/10.1086/522096. Rastier, F. (2015). Interpretative semantics. In R. Nick (Ed.), Routledge handbook of semantics (pp. 1–14). Routledge. Rozin, P. (2006). The Integration of biological, social, cultural and psychological influences on food choice. In R. Shepherd & M. M. Raats (Eds.), The psychology of food choice (pp. 19– 40). CABI. https://doi.org/10.1079/9780851990323.0019. Schutz, H. G. (1988). Beyond preference: appropriateness as a measure of contextual acceptance of food. In D. M. H. Thomson (Ed.), Food acceptability (pp. 115–134). New York: Elsevier. Schutz, H. G. (1994). Appropriateness as a measure of the cognitive-contextual aspects of food acceptance. In Measurement of food preferences (pp. 25–50). Boston, MA: Springer US. https://doi.org/10.1007/978-1-4615-2171-6_2. Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: informative and directive functions of affective states. Journal of Personality and Social Psychology, 45(3), 513–523. https://doi.org/10.1037/0022-3514.45.3.513.

Beverages in context

407

Sester, C., Deroy, O., Sutan, A., Galia, F., Desmarchelier, J.-F., Valentin, D., et al. (2013). “Having a drink in a bar”: an immersive approach to explore the effects of context on drink choice. Food Quality and Preference, 28(1), 23–31. Elsevier https://doi.org/10.1016/ J.FOODQUAL.2012.07.006. Spinelli, S., Dinnella, C., Masi, C., Zoboli, G. P., Prescott, J., & Monteleone, E. (2017). Investigating preferred coffee consumption contexts using open-ended questions. Food Quality and Preference, 61, 63–73. https://doi.org/10.1016/j.foodqual.2017.05.003. Stroebele, N., & De Castro, J. M. (2004). Effect of ambience on food intake and food choice. Nutrition, 20(9), 821–838. https://doi.org/10.1016/j.nut.2004.05.012. Tinley, E. M., Yeomans, M. R., & Durlach, P. J. (2003). Caffeine reinforces flavour preference in caffeine-dependent, but not long-term withdrawn, caffeine consumers. Psychopharmacology, 166(4), 416–423. Springer-Verlag https://doi.org/10.1007/ S00213-002-1354-1. Violi, P. (2001). Meaning and experience. Bloomington and Indianapolis: Indiana University Press. Zellner, D. A., Stewart, W. F., Rozin, P., & Brown, J. M. (1988). Effect of temperature and expectations on liking for beverages. Physiology & Behavior, 44(1), 61–68.

This page intentionally left blank

Automobiles in context Nathalie Herbeth*, David Blumenthal† *Groupe Renault, Guyancourt, France, †UMR Ing
enierie Proc
ed
es Aliments, AgroParisTech, Inra, Universit
e Paris-Saclay, Massy, France

20.1

20

Introduction

For the last 10 years, research in the car industry has been tending towards a user experience perspective. Although no consensus on a general definition of user experience (UX) has been reached (K€ orber, Eichinger, Bengler, & Olaverri-Monreal, 2013), the definition of Hassenzahl and Tractinsky (2006) in the field of human-computer interaction (HCI) and interaction design for instance includes the notion of context: UX is a consequence of: – A user’s internal state (predispositions, expectations, needs, motivation, mood, etc.), – The characteristics of the designed system (e.g. complexity, purpose, usability, functionality, etc.) and – The context (or the environment) within which the interaction occurs (e.g. organisational/social setting, meaningfulness of the activity, voluntariness of use, etc.)

Focusing on the system appraisal, Mahlke and Th€uring (2007) developed the model “Components of User Experience” (Fig. 20.1). UX researchers describe two types of benefits: the functional benefits referring to the instrumental and practical characteristics of the product and the hedonic benefits referring to the aesthetic, sensory, and symbolic characteristics (Chitturi, Raghunathan, & Mahajan, 2008; Hassenzahl, 2004; Mahlke & Th€ uring, 2007). Finally, UX theory underlines that products exceeding customers’ expectations on relevant benefits will elicit positive emotional responses and enhance satisfaction (Hassenzahl, 2003). Although there are some discrepancies in consumers’ strategies to evaluate products, Chitturi, Raghunathan, and Mahajan (2007) demonstrated that consumers focus more on the functional benefits than on the hedonic benefits of a product until their minimum expectations of fulfilling utilitarian goals are met. In addition to this, Chitturi et al. (2008) underlined that products exceeding customers’ functional expectations enhance customer satisfaction, and products exceeding hedonic wants enhance customer delight. Dealing with the context in that UX perspective means that researchers will have to consider the interactions between the context and both ergonomic and hedonic appraisals. Customers’ hedonic evaluations of cars are strongly impacted by the specific tasks performed by the customers, and therefore by the context of these tasks. In the Human Factors domain, Chater and Novick (2001) address the characterization of context by defining three main categories of contexts in evaluation (Fig. 20.2): –

The work context, considering the product in real use.

Context. https://doi.org/10.1016/B978-0-12-814495-4.00020-9 Copyright © 2019 Elsevier Inc. All rights reserved.

410

Context

Components of user experience Perception of noninstrumental qualities Visual esthetics, haptic quality, identification, etc.

Consequences

System User

Interaction characteristics

Emotional reactions Feelings, motor expressions, physiological reactions

Context

Overall judgment, usage behavior, choice of alternatives, etc.

Perception of instrumental qualities Controllability, effectiveness, learnability, etc.

Fig. 20.1 Model “components of user experience” (Mahlke & Th€ uring, 2007). From Michael Minge, Manfred Th€uring, Hedonic and pragmatic halo effects at early stages of User Experience, Int. J. Human-Computer Studies 109 (2018) 13–25, Original from M. Th€ uring, S. Mahlke, Usability, aesthetics, and emotions in human-technology-interaction, Int. J. Psychol., 42 (2007), pp. 253–264.

1 Work context

Team Operational context Time-critical & dynamic

User 1

Tasks

Other artifacts

2 OSCT context

User i

Artifact in use

3 Evaluation context

Fig. 20.2 A multidimensional structure of context (Chater & Novick, 2001).

Automobiles in context

411

Characterizing the real or natural context of car use is essential because users’ perceptions of actions that are situated within specific spatial and temporal contexts may vary across these contexts. For instance, touchscreen tablets that were used in car cockpits 10 years ago and touchscreen tablets in the entertainment industry have different specifications. In fact, car developers check the compatibility of applications for safe use while driving. When Apple launched the iPhone, the customers were astonished by the touch feeling of the screen, which allowed them to brush the screen to interact instead of pressing the screen. This is the mean difference for a customer between a capacitive screen (the iPhone) and a resistive screen (all the others at the time). In a car, it could be counterproductive to have such a sensitive screen to brush because of the movements of the body during driving. But it introduces a discrepancy between what customers can have in their pocket and in their car. Users’ tasks are central in the work context. –

–

The Organizational, Social, Cultural and Technical (OSCT) context, considering the user culture, knowledge, product standards, and safety rules. For instance, when evaluating car human-machine interfaces, the researcher should consider how familiar the customer is with technology: What is his current car and level of equipment? Does he often use connected devices? How does he interact with connected devices? The evaluation context, considering test objectives and the design of experiment. The evaluation context must answer the traditional questions of an evaluation in absolute or with competitors, of a monadic or a sequential presentation of cars, of a test in static or while driving. Car features that should be tested while driving (steering wheel for feedback for instance) does not allow a real comparison between products (cars can only be tested one by one), unless you have a mock-up that allows you to switch very fast from one proposal to another.

The evaluation realism is defined by the gap between the work context and the evaluation context. The first section of this chapter will address the context characteristics in static and dynamic evaluations of cars (dynamic evaluations meaning while driving). The second section of this chapter will then address context specifications to design relevant evaluations of cars’ subparts. In fact, during the new car development process, the car is not always the product of interest. In many cases, subparts of the car (i.e., the seat, the dashboard, or the navigation system) are the products that are considered by engineers and designers. The whole car is then a part of the context.

20.2

Characterizing the context to design relevant cars’ evaluations: “Between-products” context

One characteristic of cars compared to food or cosmetic products traditionally studied in sensory science is motion. The changing context of a car in motion will impact the sensory characteristics of this car. For instance, changing lights impact the readability of instruments because of reflection; the vehicle speed and road surface characteristics

412

Context

impact cockpit noise perception. Consequently, there are two main evaluation contexts to consider in the car industry: – –

In the static context, the car will be evaluated with the engine off. That research context corresponds to the car discovery, as people would do in the street or in a car dealer shop. In the dynamic context, the car will be evaluated while driving. That research context corresponds to the car use on the short term (such as a test drive before purchase) and on the long term (from 1 month to several years).

20.2.1 Static evaluations During their interactions with a car, drivers and passengers are exposed to a wide variety of stimuli, depending on the vehicle itself (styling, engine power, road handling) but also on the driving situation (e.g., speed, road coating, bends, temperature, and hygrometry). We will first focus on the static properties of a car, referring to the perception a consumer would have when observing the car in a show-room context. Product design is a key point for companies as it is often the first point of contact for consumers and can determine whether the product is considered for purchase (Bloch, 1995). Creusen and Schoormans (2005) define six roles of product appearance in product choice: communication of aesthetics, symbolic, functional and ergonomic product information, attention drawing, and categorization. The esthetics and symbolic roles being the most relevant for consumers. In competitive business environments in which products are often similar regarding their technical definitions, quality and price, product design has become a decisive buy-argument and plays an important role in the product choice (Demirbilek & Sener, 2003) and product commercial success (Smith, 1994; Yamamoto & Lambert, 1994). Extensive research has been dedicated to the analysis of the relationships between product design and product attractiveness (Hoegg, Alba, & Dahl, 2010; Noble & Kumar, 2010) or product elicited emotions (Desmet, Hekkert, & Jacobs, 2000; Norman, 2004). Although product appearance and visual evaluations of products are of interest, it is important that product appearance is congruent with other sensory aspects, as visual properties create expectations about what the other senses will perceive (Smets & Overbeeke, 1995). How does the context interact with visual evaluations of products? In their comprehensive framework for consumer response to the visual domain in product design (Fig. 20.3), Crilly, Moultrie, and Clarkson (2004) characterize the environment by the physical condition of the context of interaction between the user and the product. The authors distinguish between five types of context parameters. First, environmental distractions deal with material conditions (such as issues of illumination, colors interfering with the product colors or the media with which the product is to be represented—commercialized product, picture, or movie) and the time available to view the product (the full details of a product take time to be explored). To assess styling properties of cars, car manufacturers have rooms where light origins and parameters are controlled. Empirical studies dealing with cars’ exterior styling manage environmental conditions by reproducing car dealers’ arrangement: the product of interest is presented among its competitors, at equal distance from the customers and with a three-quarters front viewpoint (Fig. 20.4). Authors would like to underline that the space required for these kinds of experiments is considerable.

Context of consumption

Metaphors

Characters

Conventions

Cliches

Environment Consumer

Producer Design team

Product

Senses

Response Cognition

Individual(s) activities management

Geometry dimensions textures materials Colors graphics details

Vision (Touch) (Taste) (Smell) (Hearing)

Esthetic impression

Semantic interpretation

Objective information Subjective information Objective concinnity Subjective concinnity

Description Expression Exhortation Identification

Affect

Behavior

Instrumental Esthetic social surprise interest

Approach avoid

Automobiles in context

Visual references Similar products

Stereotypes

Symbolic association Self-expressive categorical (inward/outward)

Organization issues

Production quality

Sensory capabilities

Personal characteristics

Communication Resources Brand style

Tolerances Finish Aging

Visual acuity Range-of-vision Color vision

Age, gender Experience Personality

Environmental distractions Background Viewing time

Tastes Trends Fashions Styles

Situational factors Motivation Opportunity Marketing Social setting

413

Fig. 20.3 Framework for consumer response to the visual domain in product design.

Cultural influences

414

Context

Fig. 20.4 Show-room conditions for static car evaluations (Blumenthal, Priez, Sieffermann, & Danzart, 2003).

Secondly, cultural influences, defined as established conventions of taste, general trends, and transient fashions, will influence responses. Product branding for instance will strongly influence perception of quality and social value. Besides, responses to design are often described as involving innate, personal, and cultural factors. A great deal of historical art and architecture is based on the notion of inherently pleasing proportions (such as the golden number and the Gestalt rules of symmetry, similarity, or repetition), based on an innate desire of consumers for order. These innate preferences are relatively universal and constant. Some researchers have attempted to verify the reality of the golden ratio in car proportions (Akhtaruzzaman & Shafie, 2011; Page, Thorsteinsson, & Ha, 2010). In a globalized world, cross-cultural studies usually aim at comparing western and eastern cultures: US and Korean consumers (Oliver & Lee, 2010), Australian and Chinese drivers (Young, Rudin-Brown, Lenn
e, & Williamson, 2012), and United Kingdom and Indian car users (Khan, Pitts, & Williams, 2016). Fetscherin and Toncar (2010) underlined that US consumers’ brand images varied according to the country of origin of the car brand and the country of manufacture of the car. A Chinese car made in the US was perceived to have a stronger brand personality than the US car made in China, suggesting that the country of manufacture of cars exerts a greater influence on the perceived personality of a brand than the country of origin of the brand. Thirdly, situational factors refer to the consumer’s motivation to view an object. For example, intrinsically motivated consumers may focus on a product’s hedonic quality and thus be more focused on aesthetic impressions than product’s meaning or product’s use. The immediate social setting within which products are consumed may also moderate consumer response. Mainstream economists have long debated the role that social influence should play in the theory of consumption (Grinblatt, Keloharju, & Ikaheimo, 2008). The automobile purchase has been a popular arena in which to study marital roles (Davis, 1976). In contrast to the housing purchase, all the studies have found the husbands’ influence to be greater than the wives’. Grinblatt et al. (2008) analyzed the automobile purchase behavior of all residents of two Finnish provinces over several years. Results indicated that the purchases of neighbors, particularly by those who are geographically most proximate, influence a consumer’s purchases of automobiles. There is little evidence that emotional biases, such as envy, account for the observed social influence on consumption. Information transmission of some sort is better at explaining the observed phenomenon. Nowadays, peer influence is not only studied but also managed by companies through social networks (Aral & Walker, 2011). Furthermore, the marketing program that surrounds a product may also moderate consumer response. In particular, the impact of brand image on customers’

Automobiles in context

415

evaluations must be considered as it is so powerful with cars. Literature has shown that brands can influence product demand by providing information about quality. Brand quality has a significant, positive effect on prices of cars, suggesting that car buyers use information about brand quality to infer the quality of models (Sullivan, 1998). Brucks, Zeithaml, and Naylor (2000) demonstrated that consumers vary on the use of price and brand name when evaluating six dimensions of quality of cars (versatility, durability, performance, ease of use, serviceability, and prestige). The brand name and price availability interactions were not significant, indicating that the brand name results are not driven by price information. Brand name was used as a cue to judge the prestige of the cars. In addition, brand name was also related to judgments of ease of use in some cases. Higher prices signaled prestige to consumers. In their evaluation of early design sketches, Herbeth, Dessalles, and Desmet (2017) underlined that the brand does not impact the overall evaluation by a contribution to the hedonic evaluation but by the functional evaluation. Without detailed information about product characteristics, brand name is used by consumers as a substitute and the brand equity transfer is done much more through the quality dimension than the image dimension of cars. Getting rid of this brand effect in sensory studies seems difficult, not to say impossible. In fact, making cars anonymous is absurd for two main reasons. First, car logos often contribute to the styling of the front faces. Secondly, subjects can easily recognize cars, even without brand identification. Blumenthal, Priez, Sieffermann, and Danzart (2004) proposed a methodology to consider the brand image effect on hedonic scores of cars. In his study, consumers had access to the car brands and based their hedonic scores from the perspective of the brand (the car is in accordance, beneath or above customers’ expectations according to its brand). Fourthly, visual references will help the consumers to categorize and to give sense to the product. These visual references, based on the consumers’ personal experience, can be: – –

–

–

Stereotypes of the product category. Stereotypes are mental images of conventional, typical, and generic examples of a product category. For instance, a sport car should be low with flashy colors and a family car (multipurpose vehicle) should be high with a big trunk. Similar products of the same category. Products are usually compared to competing products or products from previous generations. Beyond references to recent designs, products may evoke iconic designs. The automotive industry capitalizes on that “neoretro” trend; examples are the new Volkswagen Coccinelle, Fiat 500, and BMW Mini. References from other product categories such as metaphors. They are defined as analogies to concepts that are already familiar. Metaphors relate to how products operate and are particularly common in electronic products to facilitate intuitive use (for instance stretching the picture with two fingers for zooming). The semiotic perspective on product design focuses on viewing products as signs capable of representation (Krippendorff & Butter, 1984). Semiotic studies are appropriate tools to analyze metaphors and are used more and more in the automotive industry. Characters refers to objects personification, which consists of giving properties of living things to products to favor interactions. For instance, the parallel between car front views and human facial expressions has been studied (Aggarwal & McGill, 2007; Windhager et al., 2012), as well as our tendency to anthropomorphize cars (Burgess & King, 2004).

416

–

–

Context

Conventions are conventional cues, culturally accepted. For instance, a passenger car must have four wheels and at least two passengers in a row. The Renault Twizzy concept broke the conventions by proposing a car for two passengers in line (one in the front and one in the back). Finally, clich
es arise when too many products are seen with the same visual references.

Fifthly, the interpersonal differences between consumers result in not only variations in the preferences they express for products, but also on the emphasis they place on the appearance of products. Psychographics (such as gender, age, income, place of living, and family situation), previous experience with cars and attitudes towards cars help car manufacturers to build customers’ profiles (e.g., the “Adventurers,” the “Functionals,” or the “Tech seekers”). These profiles can be used as screeners according to the objectives of the study. If questionnaires and market segmentations are confidential, some studies have been published in marketing science (Hashim, Whitfield, Jackson, & Effendi, 2012; Lee, 2009; Shende, 2014). Whilst there has been much research into the visual aesthetics of cars, hearing, touch, and smell also play an important role in customers’ evaluations (Blumenthal, Priez, Sieffermann, & Danzart, 2001). One has already heard consumers express how they “love that new car smell” (Hoyer & Stokburger-Sauer, 2012). The new car smell comes from the vapors of plastics, stabilizers, glues, and other foams. The large number of agents contributing to the new car smell makes it difficult to attribute a characteristic of smell to a component of the car cabin. Extensive research has been conducted by car manufacturers to analyze not only the impact of cabin materials on the new car smell but also on health. The mastering of cabin smell acquired by car manufacturers over the past 20 years has forced them to begin a new era: the quest for a brand smell. Instead of containing bad smell emissions, how about making the car smell good? A sensorial smell strategy is now applied to allow a scent to become an element of a car brand’s identity and image (Hult
en, 2011). Tactile properties of the compartment also strongly contribute to the first overall quality judgment a consumer has in a car show-room (Giboreau, Navarro, Faye, & Dumortier, 2001). Sensory science methods have been used to characterize the tactile properties of cabin materials. Wellings, Williams, and Tennant (2009) studied the impact of the context on the hedonic scores of cars switch-feel. To do so, participants separately assessed the feel of push-switches either seated inside the cars (in context), or with the samples on a bench (out of context). Results showed a significant impact of the context for only one switch on five. Authors conclude that the expense of hiring vehicles can be saved as conducting similar assessment of switches on a bench yield accurate enough results for correct decisions to be made during new product development.

20.2.2 Dynamic evaluations 20.2.2.1 When the focus is elicited sensations We have described how to deal with the context of the evaluation when cars are tested “engine-off.” However, many key features of cars will be evaluated while driving, most of these features having a link with kinesthesia: engine performances, brake and steering sensations, road handling perception, vibration and noise pollutions, and so on.

Automobiles in context

417

So, researchers must consider all the parameters that will affect the driver’s sensations. Usually, they define the experimental route to maximize the impact of the studied parameters. For instance, imposed driving contexts have been used to study the perception of the brake feel (Dairou, Priez, Sieffermann, & Danzart, 2003), the steering feel (Herbeth & Dairou, 2005), and the gearshift sensations (Herbeth, 2007). For the roll and the lateral support study, Astruc and Blumenthal (2005) chose a driving procedure close to reality for consumers: freeway with large curves (100 km/h) and traffic circle (35 km/h) to induce road-holding sensations (Fig. 20.5). Bergeron, Astruc, Berry, and Masson (2010) studied the perception of internal automotive road noise by establishing sensory profiles. Seven cars were driven under four well-defined conditions varying in speed and road surface. In fact, there are three types of road noises that impact acoustic comfort of cars: the engine noise, the bearing noise, and the aerodynamic noise. The bearing noise varies depending on the coating and the wear of the road. The engine noise is dominating at low speed whereas the aerodynamic noise can be irritating at high speed. In her thesis, Astruc (2007) compared the hedonic evaluations of acoustic comfort of cars in two contexts: (1) the participants chose the most appropriate route to perform his/her evaluation (one route per participant) and, (2) the experimenter defined a route for its performances to reveal the acoustic properties of cars (the same route for all the participants). Results show that the communal route yields greater perceived differences between the cars, with hedonic rankings remaining the same in both contexts.

Parking

100m Fig. 20.5 Example of a driving procedure set by the researchers.

418

Context

To the question “Can we impose a route for hedonic tests?,” Astruc answers yes, drivers not only felt comfortable to be guided during the experiment but also found it normal to compare cars in a relatively controlled environment. To go further, Astruc, Blumenthal, Delarue, Danzart, and Sieffermann (2008) proposed a methodology to define a unique route from individual ones for dynamic hedonic tests of cars.

20.2.2.2 When the focus is usability In addition to the driver’s sensations, the researcher must also consider all the parameters that will interfere with the driver’s attention. Michon (1985) described a threelevel structure for subtasks involved in driving: – – –

Strategic tasks, of the highest level, such as the best route to take or the best moment to leave Tactical tasks, based on interactions between the user and the context and between the user and other users, such as making a stop, changing lanes, taking a phone call Operational tasks, dealing with the sequence of actions to perform a task planned at the higher levels, such as activating the blinkers and then turning the steering wheel

The introduction of computing and communication technologies within cars has an impact across these three levels (Forbes & Burnett, 2008). Burnett (2008) describes three types of systems: –

– –

Information-based systems, such as the navigation, the travel and traffic information, and driver alertness monitoring systems, provide relevant information to the driver about the driving environment, the car status, or even the driver him(her)self. These systems affect the strategic and tactical levels. Control-based systems, such as the adaptive cruise control, the lane keeping assistance, or the automatic parking, take the lead on the driver’s actions. These systems affect the operational level. Multimedia systems, such as music, phone call, or internet access, do not provide any functionality to support the driving task. These systems will potentially impact the three levels through distraction.

Usability is a well-established concept in the human factor domain to represent the user friendliness of a system (Maguire, 2001). Usability or “quality of use” is defined as: The extent to which a product can be used with efficiency and satisfaction by specific users to achieve specific goals in specific environments (ISO 9241 standard)

This definition clearly underlines that the perception of the product features is affected by the characteristics of the context of use. The environment or context of use is defined by the same standard as: The users, tasks and equipment (hardware, software and materials), and the physical and social environments in which the product is used”

Although this definition of context of use considers a social dimension, social theories are largely missing in the human factors/ergonomics. Farrington-Darby and

Automobiles in context

419

Wilson (2009) explain this phenomenon by the fact that social factors are explicitly part of the context or, worse, are considered as nuisance factors to be controlled rather than as focal constructs. To illustrate this point, the Context Guidelines Handbook (Macleod et al., 1993) specifies the content of the context without any mention to the social context (Table 20.1). Focusing on evaluation of in-car devices, Burnett (2008) ranks six automotive environments according to their validity and controllability (Fig. 20.6). Environments deal with laboratory versus open roads tests and with static versus dynamic tests. Here again, the social factors are missing although evidence shows, for instance, that peers’ Table 20.1 Context description User

Task

Environment

Personal details User type User role Skills and knowledge Product experience Related experience System knowledge Task experience Training Input device skills Qualifications Linguistic ability Personal attributes Age Gender Physical capabilities and limitations Cognitive capabilities and limitations Attitude and motivation

Task name Task characteristics Task goal Task frequency Task duration Task flexibility Physical and mental demands Task dependencies Task output Risk resulting from error

Organizational environment Structure (job function, work practices, assistance, interruptions) Attitudes and culture (policy on use of computers) Job design (performance monitoring and feedback, autonomy) Technical environment Configuration hardware, software, and reference materials Physical environment Workplace conditions (auditory, thermal, visual environment) Workplace design (user posture, location) Workplace safety

From Macleod, M., Thomas, C., Dillon, A., Maguire, M., Sweeney, M. Maissel, J., et al. (1993) Context guidelines handbook, Version 3. Teddington, UK: National Physical Laboratory.

Increasing confidence that data correspond to real phenomena

Real road field trials Real road test trials Test track studies Dynamic vehicle simulations Static vehicle simulations Part task/laboratory studies

Increasing control of variables and replication

Fig. 20.6 Environments for evaluation of in-car computing devices and the relationship between validity and control (Burnett, 2008).

420

Context

influence is of great importance in technology adoption (Andrews, Drennan, & Bennett, 2005; Karapanos, Zimmerman, Forlizzi, & Martens, 2009). In social marketing, researchers are focused on the way to predict and to promote some driving behaviors (Bamberg & Schmidt, 2003; Peattie & Peattie, 2009). Assessments of systems performed during driving are a huge challenge for researchers as the driver’s attention is focused on the driving activity. Researchers from the human factor domain incorporate a wide range of measures: for instance, observations to capture the use of the system during predefined tasks, time to perform a task, number of errors, and questionnaires (Burnett, 2008). Self-confrontation interviews in which the subject is commenting on the video of his past actions and perceptions allow the researcher to study at once the user perception and the context of the interaction (Cahour & Forzy, 2009). Some studies have addressed the question of cross-cultural usability. Bottom-up studies aim at developing guidelines for specific cultures and contexts from comprehensive banks of data while top-down studies investigate usability issues from the perspective of cultural models (Zhu, 2015). For instance, Son et al. (2010) analyzed the extent to which age and culture influence the driver’s attention. Parallel driving simulation sessions were conducted in the United States and Korea to measure dual-task performance. Young et al. (2012) and Khan et al. (2016) based their research on the cultural model of Hofstede (1980). Hofstede identified five cultural dimensions that can be used to characterize a culture: power distance, masculinity versus femininity, individualism versus collectivism, uncertainty avoidance, and time orientation. Findings of these studies provide clear support to the hypothesis that there are significant differences among cultures in terms of design preferences and understanding.

20.3

Characterizing the context to design relevant automotive systems’ evaluations: “Inside-product” context

Cars are often seen by nonprofessionals as a whole, with diverse services thanks to diverse features: vehicle maximum speed, autonomy, off-road capacities, number of occupants, capacity of storage, and so on. However, professionals of the car industry consider an automobile as a system composed of various subsystems in interaction. System engineering has been implemented in the new car development process to identify the various stakeholders and to manage their interactions. Although differences exist in companies’ practices, Ulrich and Eppinger (2004) describe the new product development process as a series of six stages (Fig. 20.7). In the car industry, it takes approximately 5 years from the early concept definition to the car sale. Cars’

Planning

Concept development

System-level design

Detailed design

Testing and refinement

Production ramp-up

Fig. 20.7 Main stages of the new product development process (Ulrich & Eppinger, 2004).

Automobiles in context

421

subsystems are developed separately at the same time and put together late in the development process. These subsystems are considered as products by developers and tested as is. New product development consists of problem solving. An important objective of prototyping and testing is to learn, to resolve uncertainty about technology, production, customers’ needs, and the market (Thomke, 2008). The problem-solving process is iterative, with each trial generating design alternatives. Experiments are often carried out all along the development process—Thomke (2008) underlines that companies generally underestimate the cost savings of early testing and prototyping that could result in information and team interactions—using simplified models of the eventually-intended test object and/or test environment (Thomke, Von Hippel, & Franke, 1998). Focusing on the matter of context in product evaluation, two questions arise: – –

While models and prototypes are necessary to run experiments, they do not represent reality completely. What kind of automotive models and prototypes must be used to get customers’ feedbacks during product development? Prototypes to be evaluated can be prototypes of subparts. In that case, the experimental context will be the comprehensive product itself (such as a car seat—which is the product subpart—in a car cockpit—which is the context of the experiment). How do researchers manage that kind of product context in experimental testing?

20.3.1 Various mock-ups During new product development, concepts can be presented to customers through various forms. Thomke (2008) uses the term “Fidelity” to signify the extent to which a model does represent a product, process, or service in experimentation. During new product development, prototypes cannot achieve a 100% fidelity because designers do not know all the attributes of the real situation. Besides, lower fidelity models can be useful if they are inexpensive and can be produced rapidly for “quick and dirty” feedbacks (Thomke, 2008). Clark and Wheelwright (1993) describe costs and representativeness for several automobile mock-ups: computer-assisted models, clay models as this malleable material can be easily shaped and is then widely used in the automotive industry, engineering prototypes which are usually less performing and/or less visually attractive than the final product (Ulrich & Eppinger, 2004), and pilot prototypes (Table 20.2). Computer-assisted models and clay models can be used by internal experts for technical diagnosis. They are too far from reality to be used for customers’ tests. In fact, clay models are representative regarding shape but low representative regarding materials, and novices’ perception of shapes is impacted by materials. Engineering prototypes can be used for specific satisfaction testing (Crawford & Di Benedetto, 2011). These tests do not have the objective to verify if the product under development meets customers’ expectations, but to validate the perception of specific product attributes. The non-satisfying attributes can be modified by the development team before product launch. Finally, final prototypes are interesting as they allow comparisons of the concept to competitors (Ulrich & Eppinger, 2004).

422

Context

Table 20.2 Examples of models and prototypes in the automotive industry

Cost: Materials Time Representativeness: Shape Assembly Function Materials Process

Computerassisted model

Clay model

Engineering prototype

Pilot prototype

* *

* *

** **

*** ***

** * * * *

*** ** * * *

*** *** *** *** */**

*** *** *** *** ***

*, Low; **, medium; ***high. From Clark, K. B., & Wheelwright, S. C. (1993). Managing new product development—text and cases (p. 666). Boston: Harvard Business Review Book.

Crawford and Di Benedetto (2011) distinguish between static and dynamic concepts. Static concepts are descriptions, drawings, pictures, or styling mock-ups, illustrating the shape, styling, and aesthetics of the product. Dynamic concepts are storyboard, video, computer simulation, functional mock-ups, or prototypes, illustrating the use and the function of the system. Authors underline that it is better to present dynamic concepts to customers as it is easier to imagine the real use of the product. Da Costa et al. (2011) studied a car rooftop with LED lights. During the early stages of the development, the rooftop must be evaluated without any context (left side of Fig. 20.8), a few months later it can be used in a mock-up (middle) and at last, it is fully integrated in a real car (right). Results demonstrated that presenting a “sensorial brochure”—that describes the sensory properties of the future product and complete the shortages of the mock-up—helps customers to imagine the future product. At early stages of development, customers’ assessments are not minimized by the poor representativeness (generally associated with poor aesthetics) of the mock-up. Herbeth, Charue-Duboc, and Manceau (2016) underlined the tension between the stage of the new product development process and the fidelity of the mock-ups. The

Fig. 20.8 Three ways to evaluate a rooftop during development: the rooftop alone, against a wall at the beginning of the innovation process (left), the same rooftop on a frame simulating a car so that customers can sit underneath (center) and at last, the rooftop fully integrated in a real car (right).

Automobiles in context

423

earlier the concept can be tested, the better for the designer to modify the product if required (late changes are often costly for the companies and can postpone the production and product launch—consequently companies are reluctant to make changes at this time). However, functional mock-ups or prototypes are available late in the development process, and they are often not visually representative of the final product. These prototypes can only be tested by experts, as customers would be disturbed by their poor visual attractivity. Fortunately, design methods and tools quickly evolve, hardly reducing mock-ups’ costs and production periods. For instance, 3D-printing rapidly delivers mock-ups of the product. Besides, the increase of the computing power allows, on one hand, the development of complex systems of simulation (for crash-tests for instance) and, on the other hand, the development of more and more realistic visualization systems (such as 3D screens and immersive systems) (Thomke, 2008). Naturally, the fidelity of each type of mock-up and their performance regarding customers’ tests should be questioned by the experimenter.

20.3.2 Many products of interest in one whole product Ulrich and Eppinger (2004) describe the various types of prototypes of a laser mouse with two axes (Fig. 20.9): the horizontal axis distinguishes product subparts (focused prototypes) from complete products (comprehensive prototypes). The vertical axis distinguishes physical prototypes from analytical prototypes. Physical Ball support prototype

Beta prototype Alpha prototype

Final prototype

Trackball mechanism linked to circuit simulation

Focused

Equations modeling ball support

Comprehensive

Simulation trackball circuits

Not generally feasible

Analytical

Fig. 20.9 Various types of prototypes according to Ulrich and Eppinger (2004).

424

Context

A car groups together many subsystems (seats, windows, dashboard, screens, etc.) that are put together late in the development process. These subsystems may be tested separately to be finely tuned. Designers and engineers may also require customers’ feedbacks on these subparts before their integration into the final product, or they would like to focus on the subpart’s perception without any consideration of its relationships with its environment (i.e., the final product). Consequently, the context of the evaluation can be the context of a product subpart itself (e.g., a dashboard for a screen or a cockpit for a seat). Herbeth (2014) evaluated the impact of the dashboard (i.e., the product context) on the cognitive and emotional perceptions of instrument panels (i.e., the product). She assumes that the dashboard impacts the customers’ perceptions of instrument panels because of (1) the customers’ appeal for the dashboards and (2) the customers’ perception of coherence between the instrument panel and the dashboard in which it is presented. The responses of 200 customers to six instrument panels implemented in three different dashboards were analyzed (Table 20.3). Product appraisals, emotions, and liking scores were measured on pictures.

Table 20.3 Experimental design of product context analysis: 6 instrument panels * 3 cockpits contexts (Herbeth, 2014)

MEGANE 3

MAZDA 3

SEAT LEON

KIA RIO

PEUGEOT VW GOLF 6 308

Instrument panel « OUT OF CONTEXT »

Neutral context (Dacia Logan)

Intermediate context (Renault Mégane)

Sport context (Renault Megane Floride)

Automobiles in context

425

The results showed that the instrument panels were over-graded in pleasant dashboards and under-graded in unpleasant ones, whereas the ranking of the instrument panels remained the same in each dashboard condition. Besides, a high perceived coherence between the instrument panel and the dashboard yielded an over-grading of the instrument panel. That study emphasizes that experiments that cannot be conducted with the final product context should be carried out in contexts as “neutral” as possible. In fact, the results underlined that the neutral dashboard was not either liked or disliked and that the coherence effect in the neutral condition was reduced. If product subparts can be presented unbranded, experimenters must carefully consider the idea that it will prevent customers’ assessments from the impact of brand image. Herbeth and Blumenthal (2013) analyzed the customers’ perception of 18 instrument panels’ pictures. They concluded that consumers could recognize their own instrument panel, and that the instrument panel recognized as theirs always got a significantly higher liking score. However, this result was also obtained with instrument panels consumers believed to be their own, but they were wrong. Out of 84 consumers (of 100) who reported that they recognized the instrument panel of their own car, only 69 were right (which is surprisingly low for people who are supposed to look at their instrument panel very often while driving). To be compared, unbranded subparts’ mean scores should be adjusted by this “recognition” factor. Finally, tests of car subparts will obviously focus on the subpart’s characteristics themselves. For instance, it would be incorrect to study the perception of the size of the seat without any consideration of the car cockpit.

20.4

Research perspectives

20.4.1 The need for methodologies to choose relevant contexts In terms of UX in the car industry, there is an urgent need to develop a methodology to design and select the context of evaluation, as Astruc et al. (2008) did in their studies. Gkouskos, Pettersson, Karlsson, and Chen (2015) propose a method that seems interesting with reflexive photography, contextual semi-structured interviews, and the UX curve, used to reconstruct experiences over time. To evaluate the relevance of a chosen context versus a reference context implies that researchers have validation criteria. In a substantial number of studies comparing two contexts of consumption (Central Location Test vs Home Use Test for example, Boutrolle, Delarue, Arranz, Rogeaux, & K€oster, 2007 and King, Weber, Meiselman, & Lv, 2004), researchers compare the ranking of the products (from the least appreciated to the most appreciated for example), the hedonic scores, and the ability of the products to be differentiated (under the assumption that the more the products are discriminated, the better the method is). In addition, some authors consider the repeatability of the scores (Hathaway & Simons, 2017). The comparison criteria can be questioned as the causes of a change in hedonic scores from one context to another could be numerous. For example, these differences can be a consequence of the expectations of the consumers on the location itself

426

Context

Fig. 20.10 Evaluation of dashboards in Virtual Reality CAVE (Mestre, Blumenthal, Boivin, & Pergandi, 2010).

or their intentions (K€ oster, 2003). For Boutrolle et al. (2007), the evolution of the degree of discrimination between contexts could be explained by the interaction product-context. We think that new criteria should be considered to evaluate the relevance of a chosen context and its evaluation.

20.4.2 The contribution of digital tools Context is a complex and multidimensional concept. Managing all the aspects of the context is a huge, not to say impossible, challenge for researchers. We do think that new technology and especially Virtual Reality will contribute to improved contexts in the future. We conducted several studies concerning dashboards using the Virtual Reality Cave. These tools allow experimenters to implement new subparts of the car in a controlled environment, even if these subparts exist only digitally. Besides, the context of evaluation can be more “ecological” than sensorial booths but still controlled (Fig. 20.10).

References Aggarwal, P., & McGill, A. L. (2007). Is that car smiling at me? Schema congruity as a basis for evaluating anthropomorphized products. Journal of Consumer Research, 34, 468–479. Akhtaruzzaman, M., & Shafie, A. A. (2011). Geometrical substantiation of Phi, the golden ratio and the baroque of nature, architecture, design and engineering. International Journal of Arts, 1(1), 1–22. Andrews, L., Drennan, J., & Bennett, R. (2005). Emotions in the experiential consumption of mobile phones. In: Proceedings of the 2005 Australian and New Zealand Marketing Academy Conference, (pp. 18–25). Aral, S., & Walker, D. (2011). Creating social contagion through viral product design: a randomized trial of peer influence in networks. Management Science, 57(9), 1623–1639. Astruc, C. (2007). Etude des contextes d’
evaluation pour un test h
edonique en analyse sensorielle. Application aux v
ehicules en condition de roulage – Study of the context of evaluation for a hedonic test in sensory science. A case-study on cars while driving. Thesis manuscript. Paris, France: AgroParisTech. 221 p.

Automobiles in context

427

Astruc, C., & Blumenthal, D. (2005). The understanding of a complex situation – a case-study on the roll and the lateral support of cars. In: Poster presented at the 6th Pangborn sensory science symposium, 7–11 August, Harrogate, UK. Astruc, C., Blumenthal, D., Delarue, J., Danzart, M., & Sieffermann, J. M. (2008). An original use of Pearson’s correlation to construct a unique assessment procedure from individual ones for dynamic hedonic tests of cars. In: 9th Sensometrics conference, July 20–23rd 2008, St. Catharines, ON, Canada. Bamberg, S., & Schmidt, P. (2003). Incentives, morality, or habit? Predicting students’ car use for university routes with the models of Ajzen, Schwartz, and Triandis. Environment and Behavior, 35(2), 264–285. Bergeron, F., Astruc, C., Berry, A., & Masson, P. (2010). Sound quality assessment of internal automotive road noise using sensory science. Acta Acustica United With Acustica, 96(3), 580–588. Bloch, P. (1995). Seeking the ideal form: product design and consumer response. Journal of Marketing, 59, 16–29. Blumenthal, D., Priez, A., Sieffermann, J. M., & Danzart, M. (2001). Sensory profiling of complex products- the front passenger compartment of cars. In: 4th Pangborn sensory science symposium, July 22–26 2001, Dijon, France. Blumenthal, D., Priez, A., Sieffermann, J. M., & Danzart, M. (2003). Relative impact of sensory and brand image attributes on consumers’ preferences: a case study on cars. In: 5th Pangborn sensory science symposium, July 20–24 2003, Boston, MA, USA. Blumenthal, D., Priez, A., Sieffermann, J. M., & Danzart, M. (2004). How to use stepwise regression to measure the relative impact of sensory and brand image attributes on consumer’s preferences. In: 7th Sensometrics conference, July 27–30 2004, Davis, CA, USA. Boutrolle, I., Delarue, J., Arranz, D., Rogeaux, M., & K€oster, E. P. (2007). Central location test vs. home use test: contrasting results depending on product type. Food Quality and Preference, 18(3), 490–499. Brucks, M., Zeithaml, V. A., & Naylor, G. (2000). Price and brand name as indicators of quality dimensions for consumer durables. Journal of the Academy of Marketing Science, 28(3), 359–374. Burgess, S. C., & King, A. M. (2004). The application of animal forms in automotive styling. The Design Journal, 7(3), 41–52. Burnett, G. E. (2008). Designing and evaluating in-car user-interfaces. In J. Lumsden (Ed.), Handbook of research on user-Interface design and evaluation for mobile technology: Idea Group Inc. Cahour, B., & Forzy, J. F. (2009). Does projection into use improve trust and exploration? An example with a cruise control system. Safety Science, 47(9), 1260–1270. Chater, M., & Novick, D. (2001). Accounting for domain context in evaluation. Departmental papers (CS), 6. Chitturi, R., Raghunathan, R., & Mahajan, V. (2007). Form versus function: how the intensities of specific emotions evoked in functional versus hedonic trade-offs mediate product preferences. Journal of Marketing Research, 44, 702–714. Chitturi, R., Raghunathan, R., & Mahajan, V. (2008). Delight by design: the role of hedonic vs. utilitarian benefits. Journal of Marketing, 72, 48–63. Clark, K. B., & Wheelwright, S. C. (1993). Managing new product development—Text and cases. Boston: Harvard Business Review Book. Crawford, C. M., & Di Benedetto, C. A. (2011). New products management (10th ed.). Burr Ridge, IL: Irwin-McGraw Hill. Creusen, M. E. H., & Schoormans, J. P. L. (2005). The different roles of product appearance in consumer choice. Journal of Product Innovation Management, 22, 63–81.

428

Context

Crilly, N., Moultrie, J., & Clarkson, P. J. (2004). Seeing things: consumer response to the visual domain in product design. Design Studies, 25, 547–577. Da Costa, A., Christofol, H., Blumenthal, D., Delarue, J., Sieffermann, J. M., Herbeth, N., et al. (2011). Le consommateur au sein du d
eveloppement de produit innovant. In: CONFERE’11, June 30–July 1 2011, Montb
eliard, France. Dairou, V., Priez, A., Sieffermann, J. M., & Danzart, M. (2003). An original method to predict brake feel: a combination of design of experiments and sensory science (no. 2003-010598). SAE Technical Paper. 7 p. Davis, H. L. (1976). Decision making within the household. Journal of Consumer Research, 2 (4), 241–260. Demirbilek, O., & Sener, B. (2003). Product design, semantics and emotional response. Ergonomics, 46, 1346–1360. Desmet, P. M. A., Hekkert, P., & Jacobs, J. J. (2000). When a car makes you smile: development and application of an instrument to measure product emotions. Advances in Consumer Research, 27, 111–117. Farrington-Darby, T., & Wilson, J. R. (2009). Understanding social interactions in complex work: a video ethnography. Cognition, Technology & Work, 11, 1–15. Fetscherin, M., & Toncar, M. (2010). The effects of the country of brand and the country of manufacturing of automobiles: an experimental study of consumers’ brand personality perceptions. International Marketing Review, 27(2), 164–178. Forbes, N., & Burnett, G. (2008). Investigating the contexts in which in-vehicle navigation system users have received and followed inaccurate route guidance instructions. L. Dorn (Ed.), Book of driver behavior and training (pp. 291–310). Vol. 3(pp. 291–310). Ashgate. Giboreau, A., Navarro, S., Faye, P., & Dumortier, J. (2001). Sensory evaluation of automotive fabrics: the contribution of categorization tasks and non verbal information to set-up a descriptive method of tactile properties. Food Quality and Preference, 12(5–7), 311–322. Gkouskos, D., Pettersson, I., Karlsson, M., & Chen, F. (2015). Exploring user experience in the wild: facets of the modern car. A. Marcus (Ed.), Design, user experience, and usability: Interactive experience design. DUXU 2015. Lecture notes in computer science. Vol. 9188. Cham: Springer. Grinblatt, M., Keloharju, M., & Ikaheimo, S. (2008). Social influence and consumption: evidence from the automobile purchases of neighbors. The Review of Economics and Statistics, 90(4), 735–753. Hashim, A. M., Whitfield, A. T., Jackson, S., & Effendi, R. A. A. R. (2012). The application of experimental aesthetics to a marketing research method in the automotive industry. In the Proceedings of the 2012 Conferences on Arts, Social Sciences & Technology iCast (pp. 25–29). Hassenzahl, M. (2003). The thing and I: understanding the relationship between user and product. In M. A. Blythe, A. F. Monk, K. Overbeeke, & P. C. Wright (Eds.), Funology: From usability to enjoyment (pp. 1–12). Kluwer Academic Publishers. Hassenzahl, M. (2004). The interplay of beauty, goodness, and usability in interactive products. Human-Computer Interaction, 19, 319–349. Hassenzahl, M., & Tractinsky, N. (2006). User experience—a research agenda. Behaviour & Information Technology, 25(2), 91–97. Hathaway, D., & Simons, C. T. (2017). The impact of multiple immersion levels on data quality and panelist engagement for the evaluation of cookies under a preparation-based scenario. Food Quality and Preference, 57, 114–125. Herbeth, N. (2007). Comparing customers and test drivers’ vocabulary: the case of gearshift sensations. In: Poster presented at the 7th Pangborn sensory science symposium, 12–16 August, Minneapolis, USA.

Automobiles in context

429

Herbeth, N. (2014). Emotion management in the new product development – A case-study in the car industry. PhD thesis. Paris, France: Polytechnique Doctoral School. 513 p. Herbeth, N., & Blumenthal, D. (2013). Product appraisal dimensions impact emotional responses and visual acceptability of instrument panels. Food Quality and Preferences, 29(1), 53–64. Herbeth, N., Charue-Duboc, F., & Manceau, D. (2016). Taking into account consumers’ emotions at the early NPD stages. In: The proceedings of the Global Innovation and Knowledge Academy conference, 25–27th June, Valencia, Spain. Herbeth, N., & Dairou, V. (2005). Development of a customer-oriented sensory lexicon. Sensory evaluation of car steering systems. In: Poster presented at the 6th Pangborn sensory science symposium, 7–11 August, Harrogate, UK. Herbeth, N., Dessalles, S., & Desmet, P. (2017). Brand and design effects on new product evaluation at the concept stage. Journal of Marketing Management, 5(1), 21–26. Hoegg, J. A., Alba, J. W., & Dahl, D. W. (2010). The good, the bad, and the ugly: influence of aesthetics on product feature judgments. Journal of Consumer Psychology, 20, 419–430. Hofstede, G. (1980). Culture’s consequences: International differences in work-related values. Beverly Hills, CA: Sage. Hoyer, W. D., & Stokburger-Sauer, N. E. (2012). The role of aesthetic taste in consumer behavior. Journal of the Academy of Marketing Science, 40(1), 167–180. Hult
en, B. (2011). Sensory marketing: the multi-sensory brand-experience concept. European Business Review, 23(3), 256–273. Karapanos, E., Zimmerman, J., Forlizzi, J., & Martens, J. B. (2009). User experience over time: an initial framework. In: Proceedings of the SIGCHI conference on human factors in computing systems (pp. 729–738). Khan, T., Pitts, M., & Williams, M. A. (2016). Cross-cuultural differences in automotive HMI design: a comparative study between UK and Indian users’ design preferences. Journal of Usability Studies, 11, 45–65. King, S. C., Weber, A. J., Meiselman, H. L., & Lv, N. (2004). The effect of meal situation, social interaction, physical environment and choice on food acceptability. Food Quality and Preference, 15(7–8), 645–653. K€ orber, M., Eichinger, A., Bengler, K., & Olaverri-Monreal, C. (2013). User experience evaluation in an automotive context. In: 2013 IEEE Intelligent Vehicles Symposium (IV), Gold Coast, QLD (pp. 13–18). K€ oster, E. P. (2003). The psychology of food choice: some often encountered fallacies. Food Quality and Preference, 14(5–6), 359–373. Krippendorff, K., & Butter, R. (1984). Product semantics: exploring the symbolic qualities of form. Innovation: The Journal of the Industrial Designers Society of America, 3(2), 4–9. Lee, J. W. (2009). Relationship between consumer personality and brand personality as selfconcept: from the case of Korean automobile brands. Academy of Marketing Studies Journal, 13(1), 25–44. Macleod, M., Thomas, C., Dillon, A., Maguire, M., Sweeney, M., Maissel, J., et al. (1993). Context guidelines handbook, Version 3. Teddington, UK: National Physical Laboratory. Maguire, M. (2001). Context of use within usability activities. International Journal of Human Computer Studies, 55, 453–483. Mahlke, S., & Th€uring, M. (2007). Studying antecedents of emotional experiences in interactive contexts. In: Proceedings of CHI 2007, San Jose, CA, USA. Mestre, D., Blumenthal, D., Boivin, L., & Pergandi, J. M. (2010). Display formats for the evaluation of automotive dashboards physical dimensions. In: 7th symposium on applied perception in graphics and visualization, July 23–24, 2010, Los Angeles, USA. Michon, J. A. (1985). A critical view of driver behavior models. In L. Evans & R. S. Schwing (Eds.), Human behavior and traffic safety. NY: Plenum Press.

430

Context

Noble, C., & Kumar, M. (2010). Exploring the appeal of product design: a grounded, valuebased model of key design elements and relationships. Journal of Product Innovation Management, 27, 640–657. Norman, D. A. (2004). Attractive things work better. In Emotional design: Why we love (or hate) everyday things. New York: Basic Books. Oliver, J. D., & Lee, S. H. (2010). Hybrid car purchase intentions: a cross-cultural analysis. Journal of Consumer Marketing, 27(2), 96–103. Page, T., Thorsteinsson, G., & Ha, J. G. (2010). Natural sections in product design. International Journal of Contents, 6(3), 71–82. Peattie, K., & Peattie, S. (2009). Social marketing: a pathway to consumption reduction? Journal of Business Research, 62(2), 260–268. Shende, V. (2014). Analysis of research in consumer behavior of automobile passenger car customer. International Journal of Scientific and Research Publications, 4(2), 1–8. Smets, G. J. F., & Overbeeke, C. J. (1995). Expressing tastes in packages. Design Studies, 16(3), 349–365. Smith, E. (1994). Good design is indeed good business. Design Management Journal, 5, 18–23. Son, J., Reimer, B., Mehler, B., Pohlmeyer, A. E., Godfrey, K. M., Orszulak, J., et al. (2010). Age and cross-cultural comparison of drivers’ cognitive workload and performance in simulated urban driving. International Journal of Automotive Technology, 11(4), 533–539. Sullivan, M. W. (1998). How brand names affect the demand for twin automobiles. Journal of Marketing Research, 154–165. Thomke, S. (2008). Learning by experimentation: Prototyping and testing. In C. Loch & S. Kavadias (Eds.), Handbook of new product development management (pp. 401–420). Routledge. Thomke, S., Von Hippel, E., & Franke, R. (1998). Modes of experimentation: an innovation process—and competitive—variable. Research Policy, 27, 315–332. Ulrich, K. T., & Eppinger, S. D. (2004). Product design and development (3rd ed.). New-York: Irwin-McGraw Hill. Wellings, T., Williams, M., & Tennant, C. (2009). Understanding customers’ holistic perception of switches in automotive human-machine interfaces. Applied Ergonomics, 41(1), 8–17. Windhager, S., Bookstein, F. L., Grammer, K., Oberzaucher, E., Said, H., Slice, D. E., et al. (2012). “Cars have their own faces”: cross-cultural ratings of car shapes in biological (stereotypical) terms. Evolution and Human Behavior, 33(2), 109–120. Yamamoto, M., & Lambert, D. R. (1994). The impact of product aesthetics on the evaluation of industrial products. Journal of Product Innovation Management, 11, 309–324. Young, K. L., Rudin-Brown, C. M., Lenn
e, M. G., & Williamson, A. R. (2012). The implications of cross-regional differences for the design of in-vehicle information systems: a comparison of Australian and Chinese drivers. Applied Ergonomics, 43, 564–573. Zhu, C. (2015). Re-examining cross-cultural user interface design indicators: An empirical study. Master’s thesis, University of Twente.

Further reading ISO (1197) ISO 9241-11: Ergonomic requirements for office work with visual display terminals. Part 11: Guidelines for specifying and measuring usability. International Standards Organisation.

The office architecture: A contextual experience with influences at the individual and group level

21

Christina Bodin Danielsson The School of Architecture, The Royal Institute of Technology (KTH), Stockholm, Sweden

21.1

Introduction

The architectural experience is contextual, where at every instant is more than the eye can see, more than the ear can hear, a setting or a view waiting to be explored as architectural theorist Lynch described it (1960, p. 1). Architecture is expressed through the shape, color, and structure used in the design and experienced in relation to its surroundings, by the sequences of events leading up to it, the memory of past experiences. The experience is a two-way process between the observer, i.e., user, and the physical environment consisting of two dimensions: (a) an emotional and physiological perception, through our senses, combined with (b) an intellectual perception, a cognitive process based on knowledge and experiences. The experience of office architecture, similar to other architectural experiences, is holistic. It is created by the combined effect of the physical characteristics of the environment, where some are architectural (Al Horr et al., 2016, p. 384). The architectural elements of an office encompasses the building layout and exterior landscaping, but also the interior environment. The latter includes plan layout, building material and detailing, placement of windows, rooms, corridors, open areas, but also furniture and equipment. How these elements are designed and configured determines the office architecture, which in turn provides the ambient conditions of the space in terms of sound, light, temperature, airflow, and air quality. All of this makes up the office environment, whose physical context is set by the office architecture, consisting of the two main dimensions: esthetic and functional together with a third symbolic dimension (Bodin Danielsson, 2015). The esthetic dimension of the physical environment concerns the nature and the appreciation of beauty in this, where esthetics is philosophically defined as subjective and sensori-emotional values (Zangwill, 2002). The role of esthetics in architecture and environmental design, and the evaluation of this is debated (e.g., Lang, 1994; Nasar, 1994). In organizational research, the esthetic experiences of organizations are recognized by the use of office design in corporate branding (Dean, Ottensmeyer, & Ramirez, 1997; Strati, 1992). The importance of the esthetic experience from an organizational perspective finds support in some Context. https://doi.org/10.1016/B978-0-12-814495-4.00021-0 Copyright © 2019 Elsevier Inc. All rights reserved.

432

Context

empirical research. For example, there are indications that beautiful and ugly spaces affect people differently over time (Maslow & Mintz, 1972), and that positive experiences of the offices architecture positively affect employees’ attitude to the workplace and organization (Bodin Danielsson, 2015). The esthetic and functional dimensions are independent of each other, although associated with each other. Esthetic dimensions of the office environment should not be disassociated from organizational goals (Strati, 1992), since these are often expressed in the esthetic dimensions of the office environment. Vilnai-Yavetz, Rafaeli, and Schneider Yaacov (2005) exemplify the relationship between the two dimensions by describing how a black leather chair may be esthetic in a manager’s office, but not in a flower shop, while it is equally functional in both environments. The functional dimension, sometimes referred to as the instrumentality of physical environments (Vilnai-Yavetz et al., 2005), concerns the usability of an environment, i.e., how well it supports the tasks and the goals of the work at the workplace. Thus, it is in human factor engineering described as instrumentality (e.g., Garling & Golledge, 1989) and can as such either support or hinder desired activities in the setting. Due to the organizational context of the office, the third dimension—the symbolic dimension—is of specific interest. This has accordingly gained the most interest from management researchers (e.g., Gagliardi, 1992; Rafaeli & Pratt, 2005). They often focus on emotions (e.g., Rafaeli & Vilnai-Yavetz, 2004), and subjective interpretations of the office environment rather than objective attributes of this, which influences employees’ task performance (Sundstrom, Herbert, & Brown, 1982). The symbolic dimension of office architecture works in tandem with the other two dimension, in affecting the users in various ways (Kupritz, 2016), e.g., interpersonal, group, and organizational communication needs in the workplace (Kupritz & Hillsman, 2011). The symbolic dimension of office architecture can either support or impede our ability to use different sensory cues through its physical properties; cues fundamental to convey and interpret messages in social interaction. Thus, office architecture can, from a communication perspective, be considered an enabler of social interaction (Vla˘duțescu, 2014). Regarding communication, the symbolic dimension of architecture is well recognized in the corporate world (Gagliardi, 1992). Today office design is often used as a strategic management tool in corporate branding, in both the external and internal branding, so-called employee branding (Bodin Danielsson, Wulff, & Westerlund, 2013). Employee branding can be defined as the process in which the brand image is driven by the messages employees receive, and that enables them to make sense of themselves (Miles & Glynn, 2004). Thus, the influence of office architecture on employees and in turn organizations comes down to its impact on the ability to communicate at the workplace (Kupritz, 2016). Nevertheless, office architecture appears to influence other aspects of relevance here, e.g., employee satisfaction and well-being (e.g., Bodin Danielsson & Bodin, 2008; Otterbring, Pareigis, W€astlund, Makrygiannis, & A., 2018). In Bodin Danielsson and Bodin (2008), it was found that there were significant differences in health and job satisfaction among employees in different office types. The best health was reported by employees working in flex- and cell-offices and the worst

The office architecture: A contextual experience

433

health in traditional open plan offices, which to some extent appeared to correlate with their job satisfaction. Regarding how the office environment influences employee well-being—both in terms of health and environmental satisfaction, the two concepts that are tightly associated with each other, are central “locus of control” and “personal control.” Their central position for employee well-being in this regard shows in the importance that employees expressing for personal control over their own physical work situation, e.g., by their ability to adopt and change their work environment in accordance with their needs whenever possible. Likewise as increased personal control over one’s own physical workspace situation leads to higher group cohesiveness and job satisfaction among office employees (Lee & Brand, 2005).

21.1.1 Personal control The concept “locus of control” refers to the degree to which a person believes to have control over the outcomes of events in life (Rotter, 1966). In a work environment context, such as an office, the associated concept of “personal control” is more relevant due to its association with the experience of stress. There are different types of personal control: (a) behavioral (direct action on the environment), (b) cognitive (the interpretation of events), and (c) decisional (having a choice among alternative courses of action) (Averill, 1973). Personal control is a key factor to understand the contextual influence of the office that operates at both an individual and an inter-personal/group level. Both levels can either increase or reduce stress, and thereby influence well-being and comfort of employees.

21.2

The spatial context of the office at an individual level

The influence of the office environment on the individual depends greatly on how much personal control the individual has in relation to the workplace, which depends on psychosocial, organizational factors, and physical factors of the work environment. The office environment enables personal control in various ways, where the proximate office environment is the most important for the employee’s satisfaction and wellbeing. Being the “home base,” from which many employees operate, it is also where most of the individual’s work is carried out when in the office. This is, however, something that depends on the individual employee’s work assignment and job rank. People that are involved in projects tend to also to engage more in teamwork, often in meeting rooms or other collaborative environments. People at higher job ranks also tend to have more meetings and are thus less at the workstation. The workplace plays an important part in many people’s daily lives, thus the office environment engages people as many emotions are tied to this (e.g., Rafaeli & VilnaiYavetz, 2004). In the office debate, open plan offices versus cell-offices is regularly discussed, which often goes back to the concept of personal control. In research, the general opinion is that employees generally prefer individual cell-offices, when they have a choice (e.g., Hedge, 1982; Kim & de Dear, 2013; Sundstrom et al., 1982). Open

434

Context

plan layouts have negative effects on personal control reflected in loss of privacy and increased acoustic and visual disturbance due to proximity to others (de Croon, Sluiter, Kuijer, & Frings-Dresen, 2005; Hedge, 1982; Kim & de Dear, 2013), but also a perceived decrease of performance (Brennan, Chugh, & Kline, 2002). However, employee attitude to sharing workspace with others depends on work assignments and needs related to this. There are indications that creative tasks are more easily carried out in a volumetric and open space, while tasks that require concentration are preferably carried out in smaller environments (e.g., Alker, Malanca, O’Brien, & Pottage, 2014). The latter is because noise disturbance negatively affects cognitive demanding assignments, e.g., calculation or writing ( Jahncke, Hygge, Halin, Green, & Dimberg, 2011; Keus van de Poll, Ljung, Odelius, & S€orqvist, 2014). However, lack of privacy appears to be less problematic in the activity-based flex-offices with nonpersonal workstations. Flex-office is an office type defined by employees and it does not have personal workstations, but good access to different work environments used on an “as-needed basis” depending on work activity. It is often used for when 24 people share workspace) –

Shared workspaces within the office

–

Flexible for organizational changes

–

Plan layout is open, based on an open flow of workspaces instead of corridor systems

–

Routine based work

–

Workstations freely arranged in the room or in rows in a larger workspace

–

Low level of interaction between employees

–

Often no amenities at workstation Continued

436

Context

Table 21.1 Continued Activity based and flexible office types: 6. Flex-office (no personal workstation, different work environments within office) –

Plan layout is open, based on an open flow of workspaces instead of corridor systems

–

Flexible for organizational changes

–

Rooms/environments for individual work and telephone calls

–

Good information communication technology (ICT) is a necessity as the common computer system is accessible from all workstations within the office

–

Different types of environments for meetings

–

Dimensioned for 20% of the work in the office not at the personal workstation

–

Back up spaces for work activities not suitable to carry out at the personal workstation

–

Sharing of common amenities in common spaces

–

Extra focus on rooms for group activities such as: project rooms (to be booked for longer periods), team rooms and meeting rooms

–

Work is both independent as well as interactive team work with colleagues in

–

The team move around in the office on an “as-needed basis” to take advantage of the wide range of common facilities

a

For further details and illustrations of office types see Bodin Danielsson (2008) and Bodin Danielsson and Bodin (2009). Source: Bodin Danielsson, C., Bodin, L., Wulff, C., & Theorell, T. (2015). The relation between office type and workplace conflict: a gender and noise perspective. Journal of Environmental Psychology, 42, 161–171. https://doi.org/ 10.1016/j.jenvp.2015.04.004; Bodin Danielsson, C., Chungkham, H. S., Wulff, C., & Westerlund, H. (2014). Office design’s impact on sick leave rates. Ergonomics, 57(2), 139–147. https://doi.org/10.1080/00140139.2013.871064.

Study 1: Satisfaction with design-related factors in different office environments In this study, we investigated differences in environmental satisfaction among employees and if this can be ascribed to the specific office type, and why we adjusted for important background factors. The analytic sample consisted of 469 office

The office architecture: A contextual experience

437

employees (men ¼ 51%, women ¼ 49%) (Bodin Danielsson & Bodin, 2009). Only part of the study about environmental satisfaction with design-related factors will be presented here. This study used the convenience-sampling method, where several offices were examined and only those that were considered to fit one of the seven identified office types in contemporary office design were asked to participate in the study. In all, 26 offices participated and were active in the following businesses: media/IT, technical professions, and personal and economic guidance. Respondents’ participation was voluntary with a response rate of 72.5%. In the sample, 34% were managers in various ranks and 60% were regular employees. The mean age was 41 years (range: 21–64 years). In contemporary office design, seven office types are identified. Studying the differences in satisfaction with their physical office environment in different office types, our study included these office types: (1) cell-office, i.e., single office room; (2) shared-room office (2–3 people share a room), traditional open plan offices; (3) small open plan office (4–9 people share a workspace); (4) medium-sized open plan office (10–24 people share a workspace); (5) large open plan office (25 and more people share a workspace), activity-based and more flexible office types; (6) flex-office (no personal workstation, different work environments within office); (7) combioffice (personal workstation, team work, different work environments within the office). For definition of the seven office types see Table 21.1 (for further details on defining features see Bodin Danielsson, 2008; Bodin Danielsson & Bodin, 2009; Bodin Danielsson et al., 2015). The satisfaction with the following three domains of design-related factors was studied: (1) Workstation design, the immediate work environment of the employee. This includes space for work material/storage, ability to personalize workstation/s, workstation’s support of work activity, and comfort/ergonomics of the workstation. (2) Workspace design, the proximate workspace/s where the individuals work. The focus was on how good the physical work environment of the workspace was, and if it contributed to job satisfaction, and supported affinity etc. (3) Office design, the office as a whole. The focus was on how well the office design reinforced interaction, had pleasant spaces for breaks and lunch areas, and was a good physical work environment in general.

The statistical analysis included univariate as well as multivariate regression analyses for dichotomous outcomes, and the Poisson regression model for summary scales. In the univariate analyses, office type was the explanatory variable. In the multivariate analyses, the office type was the main explanatory variable, but with control for the background factors of age, gender, job rank, and line of business (Bodin Danielsson, 2010; Danielsson, 2005). The univariate analysis showed the effect of office type on employee satisfaction in the three domains of design-related factors, in other words employee environmental satisfaction with these specific domains of their physical office environment. The univariate analysis also showed how the proportions of complaints/dissatisfaction were distributed between employees in different office types. Results of this analysis

438

Context

showed the highest proportion of complaints about the domain workspace design, with on average about 40% dissatisfaction with this, followed by office design with 32% dissatisfaction. Employee complaints on workstation design were less, with one exception—the ability to personalize workstations, i.e., decorate the workstation. With regard to possibility of personalization of workstations, about 56% were dissatisfied. As we look at individual office types, the analysis showed significant differences in satisfaction with design-related factors between employees in different office types. Office type had significant impact for all outcomes, with the exception for the workstation design item about general comfort (ergonomics). Results showed that cell-office employees were generally more satisfied with the two domains (workstation design and workspace design) than employees in other office types. Cell-office employees were significantly less satisfied than employees in other office types with the exception of one item of workspace space design—the workspace’s support of affinity. About 63% of the cell-office employees reported discontent with this aspect of their office design. Most dissatisfaction with both workstation and workspace design was reported in medium-sized open-plan offices (10–24 people/room). The high dissatisfaction in this office type compared to large open plan office types (>24 people/room) might come as a surprise at first, but the result is in line with other research that has found the medium-sized open plan offices in many ways to be worse for employees than large open plan office, >24 people/room people. For example, I have found that employees in medium-sized open plan offices run a higher risk of reporting both ill-health and more stress symptoms than employees in large open plan offices (Bodin Danielsson & Bodin, 2008, 2010). Also, with regard to noise disturbance, larger workspaces appear to be better. Research has found indications that it is more preferable to be exposed to more voices instead of fewer voices from both an acoustic satisfaction and workload perspective (Keus van de Poll et al., 2015). The hypothesis for this is that multiple voices mask noise disturbance better as the negative effect of background voices is removed with a large enough number of voices, possibly because of the cues to segmentation (such as abrupt changes in pitch and amplitude). Our study showed in the univariate analysis of satisfaction with design-related factors in individual office types that employees in the activity-based flex-offices, i.e., the office type with nonpersonal workstations, were most satisfied with the domain of office design. Most dissatisfaction was in medium-sized open plan offices. In the multivariate analyses, we controlled for age, gender, job rank, and line of business. Results showed that these, in no significant way, alter the results from the univariate analysis. To provide a good overview of the results, see Table 21.2 below, which is a synthesis of two different types of analyses we performed of employee satisfaction with design-related factors. The synthesis is based on: (1) the highest and lowest proportion of complaints or less satisfaction with design-related factors, and (2) the multivariate analysis of satisfaction with these factors. Satisfaction with workstation design—measured with (a) sufficiency of space for work material, (b) possibility for personalization, (c) supporting work, and (d) general comfort, ergonomics. The synthesis of the two analyses shows for this domain that cell-office employees were the most satisfied—they were significantly more satisfied

Table 21.2 Distribution of satisfaction and dissatisfaction with designated factors in different office typesa Items (self-reported complaints about or less satisfied with)

Design-related factors Workstation design Suffic. space for work material Possibility for personalization Supporting work General comfort, ergonomics Workspace design General physical work environment Contribution to job satisfaction Supporting affinity Office design Reinforcing interaction Spaces for breaks Lunch areas General physical work environment

Medium-sized open plan office (n 5 75)

Large open plan office (n 5 75)

Flexoffice (n 5 81)



●

●

●



●

Cell-office (n 5 131)

Shared-room office (n 5 26)

 

Small open plan office (n 5 56)

●

●

●



●

●



●



●



 

●



●  

Combioffice (n 5 43)

 ●

● ●

 ●

Total sample n ¼ 469 employees. , Highest relative degree of satisfaction. ●, Highest relative degree of dissatisfaction. a The synthesis is based on: (a) the highest and lowest proportion of complaints about or less satisfaction with design-relate factors (see Table 1 in Bodin Danielsson & Bodin, 2009) and (b) the multivariate analysis (see Table 2 in Bodin Danielsson & Bodin, 2009). Source: Bodin Danielsson, C., & Bodin, L. (2009). Differences in satisfaction with office environment among employees in different office types. Journal of Architectural and Planning Research, 26(3), (Autumn, 2009) 2241–2257. doi: http://www.jstor.org/stable/43030872.

440

Context

with all variables used to measure satisfaction. However, for satisfaction with the general comfort of the workstation, employees in small open plan offices were equally satisfied as cell-office employees. Employees in all other office types were less satisfied; most dissatisfaction was in the medium-sized open plan office and the flexoffice employees were the most dissatisfied. Satisfaction with workspace design—measured with (a) general physical work environment, (b) its contribution to job satisfaction, and (c) its support of affinity. In this domain, we find a less dichotomized situation than for workstation satisfaction between different office types with most satisfaction in cell-offices—for general physical work environment and workspace design contributing to job satisfaction. Celloffice stands out negatively against the other office types in one case—for workspace design’s support of affinity at the office. Most satisfied with this are employees in small open plan offices and in flex-offices. Satisfaction with office design—measured with (a) reinforcing interaction, (b) spaces for breaks, (c) lunch areas, and (d) general physical work environment. For the domain satisfaction with office design overall, we find again a more dichotomized situation although in this case two categories of office types have more satisfied employees. In the category of small or individual office rooms, i.e., cell-offices (1 person/room) and shared-room office (2–3 people/room), they are significantly more satisfied with both lunch areas and general physical work environment. While employees in flex-offices are significantly more satisfied than others with office design’s reinforcement of interaction and spaces for breaks. Most dissatisfaction was found among employees in medium-sized open plan offices. In the study, we carried out additional multivariate analysis as well, using the Poisson regression model, to assume the relative risks (RR) for dissatisfaction with the three domains of design related-factors in the different office types. This Poisson regression analysis found a significant overall effect of office type for complaints on design-related factors. It showed that the relative risk (RR) for complaint was significantly higher in all office types in comparison to cell-offices, with the higher risks in all traditional open plan offices (small, medium-sized, and large open plan offices) (for details on relative risks and average numbers of complaints in separate office types, see Bodin Danielsson & Bodin, 2009). In summary, our study showed, based on various analyses presented in the synthesis of Table 21.2, a clear distinction in environmental satisfaction between employees in various office types for the three domains of design-related factors.

Study 2: Ownership of workstation influence on satisfaction with design-related factors This Australian study investigated the impact of having a personal workstation, i.e. fixed desk, versus not having a personal workstation for employees’ environmental satisfaction with their offices (Kim et al., 2016). It was based on a dataset collected through a short online survey (6–7 min) provided by BOSSA (Building Occupant Survey System Australia). Participants worked in 20 Australian office buildings. The survey covered information about workspace and sociodemographics of the respondent

The office architecture: A contextual experience

441

and if the employee had a fixed or nonfixed workstation. Building information was also collected (e.g., size, floor plans, heating, ventilating systems, design features, etc.). The building information is not presented here as this is outside the focus of the book chapter. Participating offices were in various businesses, e.g., building and construction, education, engineering, financial services, manufacturing, professional services, public administration etc. The total sample of the survey was of 3974 employees and the response rate was 45%. Some responses were excluded due to missing and/or incorrect information. Thus, in the final analytic sample, there were 3967 people, of which sociodemographic information from 3794 people was used. Participants had either workstations (56%) or nonfixed, i.e., nonpersonal, workstations (44%). The sample had an almost equal gender distribution, with a majority of participants in the middle-age (31–50 years) and in the following types of work: administrative, technical, professional, managerial, and other. With the Indoor Environmental Quality (IEQ) survey, employee’ satisfaction with various aspects of the physical work environment was assessed (including spatial comfort, air quality/thermal comfort, noise and visual distraction, and general comfort influencing productivity, health etc.) (for details, see, Kim et al., 2016). Focusing on environmental satisfaction, only the results about following design-related factors are presented below: (1) Spatial comfort, questions asked included immediate spatial factors and spatial factors outside the personal work zone such as space for breaks and interaction, ability to personalize workstation/s, comfort of furnishings, space for storage, amount of workspace (in normal work area). (2) Visual comfort, questions asked included lighting situation at normal work area, including both the architectural design of the office and how this supports or inhibits daylight, as well as lighting equipment. Thus, visual comfort measured lighting comfort and access to daylight. (3) General comfort, questions asked included various dimensions important for employees’ general comfort and productivity. Here only results on the satisfaction with architectural design-related factors are presented.

Different statistical analyses were performed, if adjustment for background factors was performed, it is not evident. The analysis of employee satisfaction with the design-related factors in the two employee groups is presented as mean rating scores on a bipolar scale (e.g., dissatisfied-satisfied, disagree-agree etc.) distributed between employees in flex-office and nonflex-office (see Fig. 21.1 from article by Kim et al., 2016). Results showed that the flex-office group outscored the nonflex office group (i.e., with fixed desks) with higher satisfaction on all but two design-related factors—“amount of workspace” and “storage space.” The differences between the groups were statistically significant on all items, with most differences in satisfaction for “space for breaks,” followed by “space to collaborate.” In terms of amount of workspace available, both groups were equally satisfied. Satisfaction was less in flex-offices than in nonflex offices (see Fig. 21.1), with the exception of personal

442

Context

Fig. 21.1 Comparison of mean rating scores (within the range of 1 ¼ the lowest to 7 ¼ the highest) for individual questionnaire items between desk-assigned (fixed desk) group and flexidesk (no-fixed desk) group (error bars: 95% confidence interval). Reprinted with permission from Elsevier. Source: Kim, J., Candido, C., Thomas, L., & de Dear, R. (2016). Desk ownership in the workplace: The effect of non-territorial working on employee workplace satisfaction, perceived productivity and health. Building and Environment.

storage space. In summary, the results showed more satisfaction with the designrelated factors among employees with nonpersonal workstations than employees with fixed-desks. Also, with regard to self-rated productivity and health, flex-office employees reported higher satisfaction. In addition to the quantitative study presented, a supplementary qualitative analysis was performed using open-ended comments in the questionnaire data from two of the buildings in the sample. This was performed to obtain more in-depth information about the research issues, and because it was not possible to control for confounding factors in comparison analysis across a number of workplaces that varied considerably in interior environments. Hundreds of comments were clustered into seven categories, as either being “positive” or “negative,” with neutral comments excluded (for details see, Kim et al., 2016). Only design- related comments are used in the following categories: (a) flexi-desking, (b) spatial comfort, and (c) visual comfort. The result of the qualitative analysis showed a lower environmental satisfaction than the quantitative data did. In fact, a majority of the comments were negative and generally concerned flex-desking and spatial comfort. The former accounted for a third of the total sum of comments, e.g., about nonavailable desks upon arrival at office. Comments about spatial comfort concerned various aspects such as personalization of workspace and design of the immediate working area (e.g., design and ergonomics of furniture, storage space etc.). The differences in employee satisfaction shown in this qualitative data are interesting, and could, according to the researchers, be due to those not happy with nonfixed workstations using the open-ended comments

The office architecture: A contextual experience

443

to complain. Regardless of the reason for this difference, the analysis of the smaller qualitative study showed a clear discontent with nonfixed workstations (for details see, Kim et al., 2016). Complaints concerned insufficient available desks on busy days (27%) and difficulties finding colleagues in the office (22%). It was clear that face-toface interactions was considered more efficient than electronic communication methods such as emails, messengers, and phones. Other complaints of nonterritorial workspace concerned time lost due to problems finding colleagues and arranging the workstation, lack of support in ergonomics and work efficiency, and insufficient storage. In summary, study two about the influence of ownership of workstation on satisfaction with design-related factors found in its quantitative study that environmental satisfaction was higher with design-related factors among employees with nonfixed desks. For example, their satisfaction was significantly higher with overall spatial comfort in comparison to employees with fixed workstations. The researchers suggest that this could be due to the fact that offices with nonterritorial workspaces, i.e., flexoffices, provide more additional work environments than offices with fixed-desks, e.g., more space for meeting or break-out areas. This would, according to the researchers, explain the overall higher environmental satisfaction in these offices. The hypothesis finds support in recent research that has found that employees who report low access to supportive facilities at their offices were also more dissatisfied with their offices (Bodin Danielsson & Theorell, 2018). Additionally, a contributing factor for the high satisfaction in flex-offices could be that the nonterritorial working arrangement provides a higher degree of opportunity to adapt to the local ambient conditions. According to the researchers, this may explain the better rating of indoor air quality and thermal conditions of the offices with nonfixed desks, despite identical ambient conditions in the other offices (Kim et al., 2016, p. 208). Nevertheless, nonterritorial workspaces were not perceived as all positive, as reflected in the supplementary qualitative data that highlighted negative effects on efficiency and on collaborations with team members. The fact that the different methods for interviews identified different problems and gave different pictures of employees’ opinion of nonfixed workstations is in itself interesting as debated in the discussion section of this chapter.

Study 3: Environmental satisfaction and perceived productivity in different office categories This Dutch study investigated employee environmental satisfaction and perceived productivity in offices with different workspace sizes with regard to: (a) productivity support, privacy, and concentration, (b) communication, and (c) architecture and plan layout. Two office categories were compared: smaller workspaces (including individual or shared-room offices) and activity-based office designs (including combi- and flex-offices). The sample (n ¼ 11,799) consisted of employees, with a response rate of 48% on this online questionnaire. The participants came from 26 different organizations in both the private and public sector. With a sample with a majority of male employees (m ¼ 62%, f ¼ 39%), this was also the case in all office categories. A majority of the

444

Context

participants were middle-aged (31–50 years old) and had high educational level (35% with undergraduate education, 32% with postgraduate education) (for further details see study by De Been & Beijer, 2014). With regard to work activity, more participants had mainly communication-oriented compared to concentration-oriented work (38% and 22% respectively). Participants with concentration-oriented jobs worked mainly in flex-offices (40%) and in offices with smaller workspaces (35%), i.e., most people with concentration-intensive jobs worked in flex offices. The study used a WODI Light questionnaire (29 items) with a five- or ten-point Likert response scale to assess satisfaction with work environment with regard to employee satisfaction with the environment, focused on four domains: (1) productivity support, privacy, and concentration, i.e., ability to concentrate on work when needed; (2) communication; (3) architecture and plan layout; and (4) facilities. Likert response scales were used to assess support and satisfaction with work environment (range: 1–5), and rate negative and positive evaluations of this (range: 1–10). In the statistical analysis of the four domains, a linear regression analysis was performed with a different dependent variable. Office type was used as the main predictor variable in statistical analysis and was added in block two by creating two dummy variables with the individual and shared-room office as a reference category. The following variables were used as covariates: (a) satisfaction with organization; and (b) work content; estimated work activity of (c) concentrated and (d) communication work, and traditional background factors such as (e) gender; (f ) age; and (e) educational level. Results of analysis presented in Table 21.3 showed for the first domain—productivity support, privacy, and concentration—significantly higher satisfaction among employees in the office category with smaller workspaces than in the activity-based office types (employees in combi- and flex-offices). Although the office types were significant predictors for satisfaction, it explained only slightly the variance in the model. Employee satisfaction with organization and gender were instead significant predictors for their satisfaction with this domain. Results showed for the second domain—communication—higher satisfaction among employees in the activitybased office type (combi-office) compared to the other (flex-office) and in the offices with smaller workspaces. The explained variance for the results was low and did not change with office type added as a predictor. However, two covariates did this: gender and educational level. For the third domain in focus—architecture and plan layout— the results showed that employees in offices with smaller workspaces were significantly less satisfied than those in the two activity-based office types (combi- and flex-offices). With office type as predictor in the analysis, the explained variance increases slightly. Some covariates had a similar effect on employee satisfaction with architecture and plan layout. Only educational level was a predictor for satisfaction with this domain. For the fourth and final domain in focus—facilities—employees in combi-offices were significantly more positive than those in offices with smaller workspaces (individual and shared-room offices). Employees in the other activitybased office type, flex-office, were significantly more negative toward the facilities than those in smaller workspaces. Office type had a significant effect on satisfaction with the facilities, but did not add much to the explained variance of the model, while all the covariates were significant predictors for satisfaction.

Block 1

Block 2

Organization Work Gender Age Educational level % time spent on concentrated working % time spent on communication R2 Organization Work Gender Age Educational level % time spent on concentrated working % time spent on communication Combi office Flex office R2

Productivity support, privacy and concentration (β)

Communication (β)

Architecture and lay out (β)

Indoor climate (β)

Facilities (β)

0.36** 0.09** -0.01 -0.03** 0.05** 0.06**

0.24** 0.07** -0.03** -0.03 0.06** 0.04**

0.32** 0.07** -0.03** 0.01 0.08** 0.09**

0.24** 0.08** -0.14 -0.01 0.10** 0.02

0.34** 0.06** 0.02 -0.02 0.02 0.04**

0.03

0.06**

0.07**

-0.01

-0.02

0.17 0.36** 0.08** -0.01 -0.03** 0.05** 0.10**

0.09 0.24** 0.07** -0.03** -0.03 0.06** 0.04**

0.14 0.32** 0.08** -0.02 0.01 0.09** 0.07**

0.11 0.24** 0.07** -0.14** -0.01 0.09** 0.03

0.13 0.33** 0.06** 0.02 -0.02 0.02 0.04**

0.07**

0.06**

0.04**

0.01

-0.01

-0.08** -0.19** 0.21

0.04** -0.01 0.09

0.10** 0.13** 0.15

-0.03 -0.06** 0.11

0.06** -0.06** 0.14

The office architecture: A contextual experience

Table 21.3 Results of the hierarchical regression analysis

Reprinted with permission from Emerald Group Publishing. Source: De Been, I., & Beijer, M. (2014). The influence of office type on satisfaction and perceived productivity. Journal of Facilities Management, 12(2), 142–157.

445

446

Context

In summary, this study showed differences in employees’ environmental satisfaction, but also on perceived productivity support. However, the explained variance for office type was low. Instead, the covariates explained a fair amount of variances in the environmental satisfaction and the perceived productivity support. Nevertheless, employees in smaller workspaces were more satisfied than in combi- and flex-offices with productivity support, privacy, and concentration. The greater degree of enclosure in these offices possibly explains this, as it is a factor considered important to employee satisfaction (Brill, Weidemann, Alard, Olson, & Keable, 2001; Sundstrom, Burt, & Kamp, 1980). Results also showed that the flex-office had a negative impact on satisfaction with productivity support, which indicates that the ability to choose a workstation does not compensate for the indicated negative effects of openness in the office environment. The possibility to personalize the workspaces in combi-offices may explain the high satisfaction in the other activity-based office type, combi-office, according to the researchers. Worth noting is that satisfaction with communication and social interaction was higher in both combi- and flex-offices than in smaller workspaces. It was highest in combi-offices, contrary to other research that has found satisfaction with these aspects to be highest in flex-offices (Banbury & Berry, 2005; Van der Voordt, 2004). In this study, the researchers argue that frequent communication may not always be preferred. They also suggest that the personal workstations of the combi-office may be positive for relationships between colleagues as the employee will get to know colleagues both in proximity and in sight of their own workstation. The high rates for communication and team collaboration in combi-offices is probably a result of various factors e.g., it is easier to find and communicate with colleagues with the assigned workstations in combi-offices. Thus, knowledge of where to find individual employees as well as team members is identified as crucial for contact and interaction at the workplace (Penn, Desyllas, & Vaughan, 1999). Additionally, the original combi-offices are by definition focused on teamwork, thus a high degree of communication is in accordance with this (see Table 21.1 in Study 1).

21.3

Discussion and conclusion of the three studies

These three studies focused on employees’ satisfaction with design-related factors of the office environment and they indicate partly various results, but also have partly different dimensions in focus. They do, however, when combined, indicate that personal control is a key factor to succeed with this and that different factors can enable this by different means. The first study showed that employees most satisfied with the design-related factors worked in cell-offices, i.e., individual offices. This office type enables the most personal control by allowing the individual to easily manifest this physically in the workspace. By closing or opening the door to the individual room, the employee can signal when visits are ok, and hereby also excludes visual and acoustic stimuli from the office. Cell-offices also enable personalization of the workstation, which is a way to take personal control of an area. In the office, it is used both to mark one’s territory and to assign one’s identity to an area (Brown, 2009; Sundstrom, 1986;

The office architecture: A contextual experience

447

Wells, 2000). In line with this, least satisfaction was reported in the office types with the least personal control—the traditional open plan offices. This was especially the case in medium-sized open plan offices. In these office types, most work is expected to be carried out by the individual workstation, and it is hard to find visual or acoustic privacy when needed, since these office types do not offer good access to back-up rooms or supplementary work environments (see Table 21.1 in Study 1). As formerly discussed, other research indicates that the medium-sized open plan offices are less beneficial for employees. Only with regard to satisfaction with the social dimensions of the office design did cell-office report significantly less environmental satisfaction than other office employees. Study 2 investigated what role ownership of a workstation has for employee satisfaction in terms of environmental comfort (spatial, visual, and general comfort) complements the picture of what is important for employees’ environmental satisfaction. The results showed that employees in flex-offices, i.e., with nonfixed desks, were significantly more satisfied with their offices than those with personal workstations. They were more satisfied with environmental factors such as air, temperature, or lighting, but also more content with the social dimension such as interaction with colleagues and spaces for breaks in their offices. This is in line with the results of Study 1. From a personal control perspective, which is the focus here, should the results of Study 2 be explained? According to the researchers, the better outcomes in offices with nonterritorial workspace strategies could be ascribed to the employees’ ability to adapt to different work activities these offices offer. This is difficult to achieve in open plan offices with allocated workstations, as these offices have no access to alternative work environments for this purpose, as there is less workspace area for common use such as rooms for meetings, concentrated work etc. In other words, personal control is not achieved in the traditional means, such as personalization or ownership of workstations, in these offices. Instead, it is achieved by the ability of free choice of where to work, which may explain the higher satisfaction with the indoor climate reported in offices with nonfixed workstations, as employees may choose their workstation depending on different ambient conditions during the workday e.g., daylight. Despite a notably higher environmental satisfaction among employees with nonfixed workstations in Study 2, a contrasting picture was revealed in the smaller qualitative part of the study. This is interesting to discuss why the two interview methods gave such different perspective on employee environmental satisfaction. The difference could be a consequence of the open-ended questions that were mainly answered by employees with a strong need to express their opinion for various reasons. In this specific case, it appeared that the unhappy minority of employees who did not like nonpersonal workstations chose more often to answer the open-ended questions than the majority that was satisfied. Independent of this, the qualitative part of Study 2 revealed that the standardized questionnaire did not cover all aspects important to employees’ environmental satisfaction. This raises the question of how suitable the quantitative method is in measuring contextual experiences, which by nature are contextual and not general. Using standardized questionnaires has, however, many advantages, e.g., the ability to conduct larger surveys. Thus, if quantitative methods

448

Context

are used to assess contextual experiences, it may be advisable to do qualitative prestudy before, to identify issues important to the employees. The final, large Study 3 compared environmental satisfaction and perceived productivity between employees working in smaller workspaces (individual or sharedroom offices) and employees working in activity-based office types (combi- and flex-offices). Results showed that employees in offices with smaller workspaces are more satisfied with productivity support, privacy, and concentration than the other group. The researchers suggest that this could be due to the higher extent of enclosure in the office types with smaller workspaces leading to less exposure to environmental stimuli and hereby problems with noise and visual privacy. Results also showed in line with Study 1 and 2 that the employees in the two activity-based office types had higher satisfaction with the office design’s impact on communication and social interaction. To conclude, the three studies show it is not evident which office categories are best from an environmental satisfaction perspective as this constitutes various dimensions. Personal control, a key factor for human well-being and satisfaction, appears also to be able to be achieved by different means. However, a key factor central to employee satisfaction with their offices is how well these office spaces support or inhibit work, which means various things for different work assignments and job ranks.

21.4

The spatial context of the office from a group and organizational perspective

Research tells us that the contextual influence of the office environment does not only operate at an individual level. Its influence is equally strong at a group level since the opportunity for interaction is built into and directed by the physical environment (Haner, 2005, p. 293). The influence at the group level is explained by various factors. For example, it has been found that physical proximity is important for friendships to develop between colleagues (Conrath, 1973; Szilagyi & Holland, 1980) and that support in social networks decreases with distance (Mok & Wellman, 2007). This is possibly because frequent interaction on a daily basis does not normally reach further than 18 m (59 ft) from the employees’ own workstation in the office (Sailer & Penn, 2009). Communication and interaction in and outside of the office, however, does vary with job type and with engagement in projects and teamwork, but also people in higher job ranks, e.g., people in various supervisory positions having more internal and external meetings. With regard to internal meetings, research has found that about 80% of the encounters between colleagues in the office are spontaneous, often taking place in the hallways, lounge areas, or the canteen (Backhouse & Drew, 1992). Places that generate activity are, in environmental psychology, called “activity nodes” (Bechtel, 1976). Still today, in the era of social media, spontaneous, face-to-face encounters between colleagues are crucial for transition of information, creativity, and development of knowledge, in organizations that rely on this (e.g., Hua, Loftness, Heerwagen, & Powell, 2011). Research focused on knowledge workers, i.e., workers whose main capital is knowledge, and innovative organizations, have found negative effects of distance on frequency of interaction beyond one’s own work group or team

The office architecture: A contextual experience

449

as well as frequency of interaction beyond one’s own floor (Becker, Sims, & Schoss, 2003). Consequently, isolation through office design of individual employees, teams, or departments reduces opportunities for colleagues to encounter each other, which in turn hinders vital communication for the organization (Boutellier, Ullman, Schreiber, & Naef, 2008; Davis, 1984). Also, research on innovate organizations, i.e., organizations that foster a culture of innovations, has found the close location of meeting rooms and/or meeting spaces to workstations to be important for collaboration in organizations (Hua et al., 2011). Proximity to these spaces provides employees with opportunities to carry out collaborative work and casual conversations as needed, without distracting colleagues nearby from carrying out concentrated work. Besides the described positive effects for collaboration and innovations, research has also identified that office design that supports social interactions may have other positive effects from an organizational perspective. We know from research that there is more communication in open offices (Banbury & Berry, 2005; Van der Voordt, 2004), but also that working in open plan offices may enhance the social climate at the workplace (Hedge, 1982). The positive effect of sharing a workspace with colleagues appears to operate at various occupational levels of organizations. There are indications that, at a collegiate level, plan layouts may positively affect employees’ satisfaction with other colleagues (Britner, 1992), but also for an employee-management relationship there appears to be advantages. For example, it has been found that employees’ perception of the manager’s friendliness is more positive if he/she is visible and audible in the office (Crouch & Nimran, 1989). Not surprisingly, it has also been found that in office types where employees and managers share workspaces, employees rate better relationships with their managers (Bodin Danielsson et al., 2013). One possible explanation for this could be that managers rely heavily on face-to-face, spontaneous, and unplanned meetings (Kotter, 1982), and these meetings occur easier in an office with shared workspaces. There are different types of office designs with shared workspaces, and from a communication and interaction perspective, the activity-based office types, such as combi- and flex-offices, appear to be better. Studies have found that employees in combi- and flex-offices have more face-to-face contact than those in individual or shared-room offices (Boutellier et al., 2008), and are more satisfied with communication and social interaction than the latter group of employees (De Been & Beijer, 2014). Besides the previous positive effects from a group and organizational perspective of sharing of work space, there might be other benefits. For example, spatial arrangements favoring spontaneous interaction have a positive impact on perceived productivity (Brill et al., 2001). For organizations that rely heavily on creativity and innovation, proximity between team members makes communication flow more efficiently (e.g., Allen, 1977; Becker et al., 2003). Bear in mind that distraction and disturbances are also important for perceived productivity (Haynes, 2008). For example, there are clear indications that cognitive demanding work, includes, calculating, memory capacity, or writing is influenced negatively by noise disturbance ( Jahncke et al., 2011; Keus van de Poll et al., 2014), which is a more frequent problem in open workspaces. Productivity is recognized to be hard to measure and it means different things for people’s work depending on job assignments and occupational levels. The

450

Context

difficulty to say how office design affects productivity was also clearly reflected in the three empirical studies presented herein. For example, Study 1 did not measure perception of productivity, but instead it measured employees’ perception of designrelated factors influence on support of work, affinity, and interaction where the results gave a very mixed outcome on the various outcomes. With regard to support of work itself, employees in cell-offices were most satisfied, but not about the design-related factors’ support of affinity and interaction at their offices. Instead, those in the activity-based flex-offices were most satisfied with this, which can be interpreted as different office types support various dimensions important for productivity, their internal importance for the organization’s productivity various between organizations and work sectors. Study 2 and 3 emphasize further the difficulty in assessing productivity. In Study 2 the results showed, for example, that employees working in offices with nonfixed workstations reported a higher perception of productivity. While Study 3 found that employees working in the activity-based office types (combi- and flexoffices) were significantly less satisfied with their office’s productivity support.

21.5

Concluding remarks

This book chapter has focused on office design from both the individual employee and the group and organizational perspective, and the reason for this is that the experience of the office and its influence operates simultaneously at these levels. As this review of office research shows, the influence of office design at these levels stand sometimes in opposition to one another. For example, enclosure and ability for privacy at the office is identified to be important for employees’ environmental satisfaction. At the same time, a high degree of enclosure and privacy does not support interaction and collaboration between colleagues - factors important for interaction and collaboration that in turn are positive for both employee job satisfaction and organizational success. The office experience, like all environmental experiences, to a great extent, is contextual by nature. A context that in the office is set by the architecture and the factors it influences, but also set by the organizational culture that owns and controls the office environment (Mazumdar, 1992). However, despite the contextual character of the office experience, due to the strong ties to the office architecture, some general conclusions can be drawn about offices’ environmental influences on employees. For example, we know offices with a shared, open plan office design tend to encourage interaction and communication (Banbury & Berry, 2005; Van der Voordt, 2004), but also lead to more visual and acoustic disturbances (e.g., Kim & de Dear, 2013; Kristiansen et al., 2009; Kupritz, 1998). The office experience is not only contextual, but also holistic. Like all environments, the office environment exerts a holistic influence on the individual. It has been found to influence employees’ health and wellbeing (Bodin Danielsson & Bodin, 2008; Bodin Danielsson, Chungkham, Wulff, & Westerlund, 2014; Meijer, Frings-Dresen, & Sluiter, 2009; Pejtersen, Feveile, Christensen, & Burr, 2011), and employees’ perception of the workplace and attitudes to the organization as a whole (Bodin Danielsson, 2015).

The office architecture: A contextual experience

451

Based on this, I believe there is a need of a holistic office design that recognizes the different needs related to office work that is founded in the contextual experience of the office. As such, it must handle the contrasting needs of concentration and stimulation—a balancing act between distraction and stimulation, central for office employees’ well-being and productivity. This belief is founded in the insight that people are different with diverse experiences and background, but needs also vary over time and with work activities. A supportive office design must relate to this and to the importance of the context.

Acknowledgments This research was supported by Formas, the Swedish Research Council for Sustainable Development (Young Mobility Grant 259-2011-1580).

References Al Horr, Y., Arif, M., Kaushik, A., Mazroei, A., Kaytafygiotou, M., & Elsarrag, E. (2016). Occupant productivity and office indoor environment quality: a review of the literature. Building and Environment, 105, 1–21. https://doi.org/10.1016/j.buildenv.2016.06.001. Alker, J., Malanca, M., O’Brien, R., & Pottage, C. (2014). Health, wellbeing & productivity in offices. The next chapter for green building. World Green Building Council. Allen, T. J. (1977). Managing the flow of technology. Cambridge, MA: MIT Press. Averill, J. R. (1973). Personal control over aversive stimuli and its relationship to stress. Psychological Bulletin, 80(4), 286–303. https://doi.org/10.1037/h0034845. Backhouse, A., & Drew, P. (1992). The design implications of social interaction in a workplace setting. Environment and Planning B: Planning and Design, 19(5), 573–584. Banbury, S., & Berry, D. (2005). Office noise and employee concentration: identifying causes of disruption and potential improvements. Ergonomics, 48(1), 25–37. Bechtel, R. (1976). Enclosing behavior. Stroudsburg, PA: Dowden, Hutchinson & Ross. Becker, F., Sims, W., & Schoss, J. (2003). Interaction, identity and collocation: what value is a corporate campus? Retrieved from Ithaca, New York. Bodin Danielsson, C. (2008). Office experiences. In H. Schifferstein & P. Hekkert (Eds.), Product experience (pp. 605–628). San Diego, CA: Elsevier Scientific Publications, Netherlands. Bodin Danielsson, C. (2010). The office—An explorative study. Architectural design’s impact on health, job satisfaction & well-being. (Doctoral degree Doctoral thesis) Stockholm: KTH (Royal Institute of Technology). Bodin Danielsson, C. (2015). Aesthetics versus function in office architecture: employees’ perception of the workplace. Nordic Journal of Architectural Research, 2, 11–40. Bodin Danielsson, C., & Bodin, L. (2008). Office-type in relation to health, well-being and job satisfaction among employees. Environment and Behavior, 40(5), 636–668. https://doi. org/10.1177/0013916507307459. Bodin Danielsson, C., & Bodin, L. (2009). Differences in satisfaction with office environment among employees in different office types. Journal of Architectural and Planning Research, 26(3), 2241–2257. (Autumn, 2009) http://www.jstor.org/stable/43030872. Bodin Danielsson, C., & Bodin, L. (2010). Office design’s influence on employees’ stress levels. In: Paper presented at the ARCC/EAAE 2010 international conference on architectural

452

Context

research, June 2010, Washington DC, USA http://www.aia.org/aiaucmp/groups/aia/docu ments/pdf/aiab087211.pdf. Bodin Danielsson, C., Bodin, L., Wulff, C., & Theorell, T. (2015). The relation between office type and workplace conflict: a gender and noise perspective. Journal of Environmental Psychology, 42, 161–171. https://doi.org/10.1016/j.jenvp.2015.04.004. Bodin Danielsson, C., Chungkham, H. S., Wulff, C., & Westerlund, H. (2014). Office design’s impact on sick leave rates. Ergonomics, 57(2), 139–147. https://doi.org/10.1080/ 00140139.2013.871064. Bodin Danielsson, C., & Theorell, T. (2018). Office employees’ perception of workspace contribution: a gender and office design perspective. Environment and Behavior, 1–32. https:// doi.org/10.1177/0013916518759146. Bodin Danielsson, C., Wulff, C., & Westerlund, H. (2013). Is perception of leadership influenced by office environment? Journal of Corporate Real Estate, 15(3/4), 194–212. https://doi.org/10.1108/JCRE-03-2013-0008. Boutellier, R., Ullman, F., Schreiber, J., & Naef, R. (2008). Impact of office layout on communication in a science-driven business. R&D Management, 38(4), 372–391. Brennan, A., Chugh, J., & Kline, T. (2002). Traditional versus open office design: a longitudinal field study. Environment and Behavior, 34(3), 279–299. Brill, M., Weidemann, S., Alard, L., Olson, J., & Keable, E. (2001). Disproving widespread myths about workplace design. Jasper, IN: Kimball International. Britner, J. M. (1992). The impact of physical surroundings on customers and employees. Journal of Marketing, 56(2), 57–71. Brown, G. (2009). Claiming a corner at work: measuring employee territoriality in their workspaces. Journal of Environmental Psychology, 29, 44–52. Conrath, C. W. (1973). Communication patterns, organizational structure, and man: some relationships. Human Factors, 15(5), 459–470. https://doi.org/10.1177/001872087301500503. Crouch, A., & Nimran, U. (1989). Office design and the behavior of senior managers. Human Relations, 42(2), 139–155. https://doi.org/10.1177/001872678904200203. Danielsson, C. (2005). Office environment, health & job satisfaction. An explorative study of office design’s influence. Stockholm: KTH (Royal Institute of Technology). Davis, T. (1984). The influence o physical environment in offices. Academy of Management Review, 9(2), 271–283. De Been, I., & Beijer, M. (2014). The influence of office type on satisfaction and preceived productivity. Journal of Facilities Management, 12(2), 142–157. de Croon, E., Sluiter, J., Kuijer, P. P., & Frings-Dresen, M. (2005). The effect of office concepts on worker health and performance: a systematic review of the literature. Ergonomics, 48(2), 119–134. Dean, J. W., Ottensmeyer, E., & Ramirez, R. (1997). An aesthetic perspective on organizations. In C. L. Cooper & S. Jackson (Eds.), Creating tomorrow’s organizations: Handbook for future research in organizational behavior (pp. 419–437). Chichester: John Wiley and Son. Gagliardi, P. (Ed.), (1992). Symbols and artifacts: View from the corporate landscape. New York: Aldine de Gruyter. Garling, T., & Golledge, R. G. (1989). Environmental perception and cognition. In E. H. Zube & G. T. Moore (Eds.), Advances in environment, behavior, and design (pp. 203–236). New York: Plenum. Haner, U.-E. (2005). Spaces for creativity and innovation in two established organisations. Creativity and Innovation Management, 15, 288–298.

The office architecture: A contextual experience

453

Haynes, B. P. (2008). Impact of workplace connectivity on office productivity. Journal of Corporate Real Estate, 10(4), 286–302. Hedge, A. (1982). The open plan office. A systematic investigation of employee reactions to their work environment. Environment and Behavior, 14(5), 519–542. https://doi.org/ 10.1177/0013916582145002. Hua, Y., Loftness, J. H., Heerwagen, J., & Powell, K. M. (2011). Relationship between workplace spatial settings and occupant-perceived support for collaboration. Environment and Behavior, 43(6), 807–826. Jahncke, H., Hygge, S., Halin, N., Green, A. M., & Dimberg, K. (2011). Open-plan office noise: cognitive performance and restoration. Journal of Enviromental Psychology, 31, 373–382. https://doi.org/10.1016/j.jenvp.2011.07.002. Keus van de Poll, M., Ljung, R., Odelius, J., Schlittmeier, S. J., Sundin, G., & S€ orqvist, P. (2015). Unmasking the effects of masking on performance: the potential of multiple-voice masking in the office environment. Journal of the Acoustical Society of America, 138(2), 807–816. https://doi.org/10.1121/1.4926904. Keus van de Poll, M., Ljung, R., Odelius, J., & S€orqvist, P. (2014). Disruption of writing by background speech: the role of speech transmission index. Applied Acoustics, 81(81), 15–18. Kim, J., Candido, C., Thomas, L., & de Dear, R. (2016). Desk ownership in the workplace: the effect of non-erritorial working on employee workplace satisfaction, percieved productivity and health. Building and Environment, 103, 203–214. https://doi.org/10.1016/ j.buildenv.2016.04.015. Kim, J., & de Dear, R. (2013). Workspace satisfaction: the privacy-communication trade-off in open-plan offices. Journal of Enviromental Psychology, 36, 18–26. Kotter, J. (1982). What effective general managers really do. Harvard Business Review, 60(2), 157–169. ˚ . M., Shibuya, H., Munch Petersen, H., Kristiansen, J., Mathiesen, L., Kofoed Nilsen, P., Hansen, A et al. (2009). Stress reactions to cognitively demanding tasks and open-plan office noise. International Archives of Occupational and Environmental Health, 82, 631–641. Kupritz, V. W. (1998). Privacy in the work place: the impact of building design. Journal of Enviromental Psychology, 18, 341–356. https://doi.org/10.1006/jevp.1998.0081. Kupritz, V. W. (2016). The communicative nature of space in organizations. In G. Koc¸, M. Claes, & B. Christiansen (Eds.), Handbook of research on cultural influences in architecture (pp. 58–79). Hershey, PA: IGI Global. Kupritz, V. W., & Hillsman, T. (2011). The impact of supervisory communication skills transfer. International Journal of Business Communation, 48(2), 148–185. https://doi.org/ 10.1177/0021943610397269. Lang, J. (1994). Symbolic aesthetics in architecture. In J. L. Naser (Ed.), Environmental aesthetics: Theory, research and application (pp. 11–26). Cambridge: Cambridge University Press. Lee, Y., & Brand, J. L. (2005). Effects of control over office workspace on perceptions of the work environment and work outcomes. Journal of Enviromental Psychology, 25(3), 323–333. https://doi.org/10.1016/j.jenvp.2005.08.001. Lynch, K. (1960). The image of the city. (Sixth Paperback Printing ed.) Cambridge, MA: MIT Press. Maslow, A. H., & Mintz, N. L. (1972). Effects of aesthetic surroundings: Initial short-term effects of three aesthetic conditions upon perceiving “energy” and “well-being” in face. In R. Gutman (Ed.), People and buildings (pp. 212–219). New York: Basic Books.

454

Context

Mazumdar, S. (1992). Sir, please do not take away my cubicle: the phenomenon of environmental deprivation. Environment and Behavior, 24(6), 691–722. Meijer, E. M., Frings-Dresen, M. H., & Sluiter, J. (2009). Effects of office innovation on office workers’ health and performance. Ergonomics, 52(9), 1027–1038. https://doi.org/10.1080/ 00140130902842752. Miles, S. J., & Glynn, M. (2004). A conceptualization of the employee branding process. Journal of Relationship Marketing, 3(23), 65–87. https://doi.org/10.1300/J366v03n02_05. Mok, D., & Wellman, B. (2007). Did distance matter before the Internet? Interpersonal contact and support in the 1970s. Social Networks, 29(3), 430–461. https://doi.org/10.1016/ j.socnet.2007.01.009. Nasar, J. L. (1994). Urban design aesthetics: the evaluative qualities of building exteriors. Environment and Behavior, 26(3), 377–401. Otterbring, T., Pareigis, J., W€astlund, E., Makrygiannis, A. L., & A. (2018). The relationship between office type and job satisfaction: testing a multiple mediation model through ease of interaction and well-being. Scandinavian Journal of Work, Environment & Health, 44(3), 330–334. https://doi.org/10.5271/sjweh.3707. Pejtersen, J. H., Feveile, H., Christensen, K. B., & Burr, H. (2011). Sickness absence associated with shared and open-plan offices—a national cross sectional questionnaire survey. Scandinavian Journal of Work, Environment and Health, 37(5), 376–382. https://doi.org/ 10.5271./sjweh.3167. Penn, A., Desyllas, J., & Vaughan, L. (1999). The space of innovation: interaction and communication in the work environment. Environment and Planning B: Planning and Design, 26 (2), 193–218. Rafaeli, A., & Pratt, M. (Eds.), (2005). Artifacts and organizations: Beyond mere symbolism. Mahwah, NJ: Lawrence Erlbaum. Rafaeli, A., & Vilnai-Yavetz, I. (2004). Emotions as a connection of physical artifacts and organizations. Organization Science, 15(6), 671–686. Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General & Applied, 80(1), 1–28. https://doi.org/ 10.1037/h0092976. Sailer, K., & Penn, A. (2009). Spatiality and transpatiality in workplace environments. In: Paper presented at the 7th international space syntax symposium, Stockholm, Sweden. Strati, A. (1992). Aesthetic understanding of organizational life. Academy of Management Review, 17(3), 568–581. Sundstrom, E. (1986). Work places: The psychology of the physical environment in offices and factories. New York: Cambridge University Press. Sundstrom, E., Burt, R., & Kamp, D. (1980). Privacy at work: architectural correlates of job satisfaction and job performance. Academy of Management Journal, 23(1), 101–117. Sundstrom, E., Herbert, R., & Brown, D. (1982). Privacy and communication in an open-plan office: a case study. Environment and Behavior, 14, 379–392. Szilagyi, A., & Holland, W. (1980). Changes in social density: relationships with functional interaction and perceptions of job characteristics, role stress, and work satisfaction. Journal of Applied Psychology, 65(1), 28–33. https://doi.org/10.1037/0021-9010.65.1.28. Van der Voordt, D. J. M. (2004). Productivity and employee satisfaction in flexible workplaces. Journal of Corporate Real Estate, 6(2), 133–148. https://doi.org/10.1108/ 14630010410812306. Vilnai-Yavetz, I., Rafaeli, A., & Schneider Yaacov, C. (2005). Instrumentality, aesthetics, and symbolism of office design. Environment and Behavior, 37(4), 533–551.

The office architecture: A contextual experience

455

Vla˘duțescu, S. (2014). Communication environment: context/situation/framework. Journal of Sustainable Development Studies, 6(1), 193–204. Wells, M. (2000). Office clutter or meaningful personal displays. The role of office personalization in employee and organizational well-being. Journal of Enviromental Psychology, 20, 239–255. Zangwill, N. (2002). Aesthetic judgment. In Stanford encyclopedia of philosophy. Stanford, CA: Stanford University. https://plato.stanford.edu/entries/aesthetic-judgment/.

This page intentionally left blank

Conducting contextualized and real-life product tests: Benefits and experimental challenges

22

Julien Delarue*, Thierry Lageat† *AgroParisTech, INRA, Universit
e Paris-Saclay, Massy, France, †EUROSYN, Villebon-surYvette, France

22.1

Testing contexts and product experience

22.1.1 Beyond CLT and HUT In the everyday practice of market research, hedonic tests of products are traditionally divided in two context-wise categories: Central Location Tests (CLTs) and Home Use Tests (HUTs) (Delarue & Boutrolle, 2010). CLTs are tests conducted under controlled conditions in rooms that are generally equipped with several tables independent of each other (or individual booths in the case of a sensory laboratory). Such a configuration allows interviewing several persons at the same time as it prevents participant distraction by the other participants being interviewed at the same time. Participants may be prerecruited in advance thanks to an access panel or they may be intercepted and selected on the spot throughout the day by the investigators. The evaluation session usually does not exceed 20 min. CLTs are thus easy to set up and can usually be completed within a few days. The controlled evaluation conditions in CLTs ensure that all participants test all products in the same way. They also enable the evaluation of multiple products in a monadic sequential way if desired. CLTs thus allow one to obtain precise data in a relatively short time and with a reasonable budget. For these reasons, CLTs are very popular, especially for testing food. However, by setting up tests under such standardized and hence artificial conditions, we completely overlook the determining contribution of context in the evaluation of food (K€ oster, 2003, 2009; Meiselman, 1996). Often opposed to CLTs, HUTs aim to evaluate foods under normal conditions of consumption. Testing the products at home implies that participants are prerecruited from databases and are therefore frequently compensated. Once products are distributed, participants are free to eat/use them during a few days and their responses are then collected, usually online. The product distribution mode is generally dictated by the budget and geographic scope of the study. Products may be dropped off at the participant’s home, shipped by mail, or simply collected by the participant in a central facility. When several products are to be tested, practices may vary. Products are generally distributed one after the other with a variable time interval between each Context. https://doi.org/10.1016/B978-0-12-814495-4.00022-2 Copyright © 2019 Elsevier Inc. All rights reserved.

458

Context

evaluation (from a few days to a few weeks). This sequential monadic product presentation is time consuming and thus a limited number of different samples are usually tested in home tests (at most 2 or 3). With its more naturalistic conditions, HUTs are believed to be more predictive of reality than CLTs. The downside of HUTs is the lack of control of the testing conditions. It is indeed impossible to truly know how people taste (or use) the product in their home, or if they even taste it at all. In addition, testing products at home requires more logistics for the survey institutes (e.g., when samples must be shipped to participants or when several products are to be tested over several days). Under such circumstances, HUTs may thus be more expensive. As can be seen, the advantages of one method are generally the counterpoint of the limitations of the other. This contrast between the precision of CLTs and the supposed validity of HUTs thus represents a “conflicting desiderata” (Brinberg & McGrath, 1985). With the aim of getting closer to the real world, nearly 30 years ago researchers in the field called for testing real people, eating real foods in real situations (Meiselman, 1992). This has given rise to much discussion about the methods we should use (Mela, Rogers, Shepherd, & MacFie, 1992; Pliner, 1992; Rolls & Shide, 1992) and whether contextual variables can be considered separately or if context should be taken as a whole (K€ oster, 2003). Eventually, most sensory and consumer scientists would now agree on the necessity to combine realism and control in order to obtain better and more reliable data (Rolls & Shide, 1992).

22.1.2 Lessons from “nonfood” studies and product design Interestingly, for food sensory and consumer scientists, testing products in a real or realistic context is seen as a plus. That is to say, a way to increase the validity of collected data (Delarue & Boutrolle, 2010; Jaeger & Porcherot, 2017; Meiselman, 2013). However, in many fields of applications other than food, evaluating products in real life settings is essential. One reason for this is that many nonfood products directly interact with the physical context in which they are used and may thus behave differently from one context to another. Typically, the perception of a car, or its elements or functions, while driving (e.g., seat comfort, steering, braking, etc.) strongly depend on the type of road, on the route, and traffic conditions (see Chapter 20: Herbeth & Blumenthal). Likewise, the perception of the cushioning of sport shoes largely depends on the nature of the practice ground. For many cosmetic products (skin care, hair care, makeup) perception (and behavior) will change depending on ambient temperature and humidity. For this reason, cosmetic companies sometimes use climate chambers for sensory evaluation of products that are sensitive to these variations. Accordingly, research methods should be able to accommodate dynamic changes in experiences that occur over time, including in the field of food science (Delarue & Blumenthal, 2015). Food design in the future is thus likely to regard the physical “food product” as just one of several elements shaping the intended consumer experience (Schifferstein, 2015).

Conducting contextualized and real-life product tests: Benefits and experimental challenges

459

Another major reason for the need to test nonfood products in real or realistic contexts relates to the need to assess the product’s functionality. Indeed, a major characteristic of many nonfood products is that they have one (sometimes several) functionality that is the main reason why people use them. One thus needs to interact with the product in order to evaluate that functionality. Most often, functionality relates to product performance or efficiency (for example, a hand razor must first ensure proper nick-free shaving. Pleasure, smoothness, and sensations of glide come next; a lipstick should last all day long; a soap should cleanse; a styling gel should allow for the shaping of hair, etc.). This means that the assessment process necessarily includes the evaluation of the functionality in addition to sensory and pleasurable aspects. Moreover, this means that to understand consumers’ hedonic responses, functionality aspects have to be evaluated (Delarue, Masson, & Blumenthal, 2018). In many cases, functionality cannot be assessed in a standard CLT. This could be because of the lack of time (for example, a lipstick needs to be tested over one full day before a consumer could evaluate its lasting effect). The need for room or for appropriate equipment may also be critical (for example, for testing shower gel). Overall, CLT conditions do not allow the consumer to experience the product in the multiple conditions of everyday life (e.g., testing wet grip for tires), when performing different activities (e.g., wearing an antiperspirant T-shirt at work, when exercising, in public transportation) and interacting with different people. As a result, for many nonfood products—and this is especially true for cosmetics—HUTs, or more generally “Use Tests” are more frequent than CLTs. However, Hekkert and Schifferstein (2008) invite us to see beyond the utilitarian function of products and to tackle the subjective “product experience”. That is to say, to develop an awareness of the psychological effects elicited by the interaction with a product, including the degree to which all our senses are stimulated, the meanings and values we attach to the product, and the feelings and emotions that are elicited. Hekkert and Schifferstein actually define the field of “product experience” as a research area that develops an understanding of people’s subjective experiences that result from interacting with products. In the case of food, given the importance of the symbolic dimension of eating behavior, product experience seems like an obvious thing to take into account. In this regard, the development of new food products should take the usage situation and the consumption context into account. Interestingly, a growing number of food companies now adopt this design thinking in the development of new products and pay much more attention to product experience and hence to looking for appropriate testing contexts. As a result of cross-fertilization with other fields (notably design engineering), a growing number of researchers and practitioners are increasingly aware of the interest of testing products in real life settings and have started to learn from other fields. In addition to this, methodological development and evolution of digital technologies, generalization of social networks, etc., have been conducive to an evolution of practices. Researchers have seized the opportunity offered by these evolutions to create more scope for context.

460

22.2

Context

Measuring consumer responses in contextualized and real-life environments

22.2.1 Challenges and added value of recreated contexts Given the need for designers to understand the product
user
context interaction, it comes as no surprise that researchers in design engineering pioneered the use of living labs. Although there is no clear and consensual definition for living labs, they are reallife or realistic experimentation environments where users (consumers) are co-involved in the innovation process (Bergvall-Ka˚reborn, Eriksson, Sta˚hlbr€ost, & Svensson, 2009; Dell’Era & Landoni, 2014). Drawing inspiration from these usercentered design approaches, some consumer researchers in the area of food design have started using home laboratories and living labs in the view of making the laboratory a real-life situation (Giboreau, 2018). At a larger scale, the University of Wageningen (The Netherlands) uses an experimental restaurant, “The Restaurant of the Future”, for studying food behavior in a meal context (Hinton et al., 2013; Zeinstra, Koelen, Kok, & de Graaf, 2010). These attempts to get closer to real food contexts using physical means is however not limited to restaurants. For example the same team has recreated an aircraft context in order to test food dishes during international flights (Holthuysen, Vrijhof, de Wijk, & Kremer, 2017). Likewise, cosmetic companies now use test bathrooms to test personal care products, but also makeup and toiletries. In a totally different field of application, one may recreate retail stores, fitting rooms, or even car showrooms, in order to get closer to situations where consumers may compare products and make purchase decisions (Fig. 22.1). As in a movie studio, one aims to recreate a complete environment. Such platforms are well adapted to test products or equipment that can usually be experienced in public places (bars, caf
e terrace, gym, etc.).

Fig. 22.1 Example of a recreated car showroom in CLT conditions in the Eurosyn facilities.

Conducting contextualized and real-life product tests: Benefits and experimental challenges

461

These approaches to make CLTs closer to real life situations—also called “situational CLTs”—may also be implemented using simpler means. For example, Kim, Lee, and Kim (2016) generated slight differences in hedonic scores when simulating a caf
e using colored tablecloths, decorative items (pictures of a caf
e, mug cups, whole coffee beans in glass bottles, books, etc.) and dressing the experimenter as a waiter. Context variables can thus be adjusted, and various degrees of realism can be reached, depending on the budget, on the objectives of the study, and its stakes. One may even attempt to make the sensory booth context more appropriate for the evaluation of a given product. For example, in a study intended to measure the satiating effect of breakfast biscuits, Talbot et al. observed that providing a choice of hot drinks and magazine doubled the time spent by participants in sensory booths and resulted in an increase of 30% of eaten biscuits (Fig. 22.2) (Talbot et al., 2009). However, the added value of such physical improvements of context may be limited if restricted to decoration of sensory booths (Petit & Sieffermann, 2007). Besides, Garcı´a-Segovia, Harrington, and Seo (2015) have shown that playing with elements of context should be consistent with the location of the test. They indeed found that participants liked the appearance of food served in plastic tray settings significantly more

Fig. 22.2 Effect of contextualizing a regular sensory booth on intake and time spent in the booths to evaluate biscuits (N ¼ 71, within-subject design; mean data + SEM) (Talbot, Delarue, & Sieffermann, 2009).

462

Context

in sensory booths than in realistic contexts and that conversely, they liked the appearance of the food served in gourmet table setting significantly more in realistic contexts (i.e., in a restaurant) than in sensory booths. This stresses the fact that playing with context variables should be done in regard of hypotheses on the role of context, and testing conditions should be properly thought out and adapted to each product type and research question. In the case of food, contextualizing the evaluation of a given product may imply serving a full meal (Parizel et al., 2016; Saint-Eve et al., 2016) (Fig. 22.3).

22.2.1.1 Advantages of recreated contexts/contextualized CLTs Contextualized or situational CLTs thus allow for providing elements of contexts to participants while controlling the experimental conditions. This has the advantage of being able to deliver the same experience to several participants, possibly simultaneously and possibly with social interactions. In addition to this, participants’ behavior can be easily observed and recorded using video cameras. Working this way also offers the possibility to use environmental variables as experimental parameters. Some researchers have indeed played with environmental variables such as the lightning and furniture in order to immerse test participants in different ambiances and to measure their effects on choice (Sester et al., 2013). Eventually, situational CLTs naturally benefit from the practical advantages of standard CLTs. Logistics remain relatively simple. Several products can be tested in a monadic sequential way if desired. Also, it is usually easier to recruit participants, especially when working with an access panel.

Fig. 22.3 Example of CLT of desserts implying participation to a full meal (Parizel et al., 2016).

Conducting contextualized and real-life product tests: Benefits and experimental challenges

463

22.2.1.2 Challenges and limits of situational CLTs In spite of their potential added value for understanding product experience and for improving participants’ engagement, testing products in situational CLTs also has some drawbacks. First considerations pertain to the realism of the settings. It is indeed not always possible to mimic natural conditions in a realistic way. Reasons for this are multiple and may lie in details such as the general appearance of the building where the test is conducted, the reception desk, the scope for choice, duration of the sessions, etc. In fact, because they come to a central facility, consumers are not fooled when they participate in such experiments. In our opinion, this doesn’t entail the interest of such test designs, but experimenters should be aware of this. Even in most realistic facilities such as living labs and experimental restaurants, participants have expectations that are associated with those places (having dinner at the Institute Paul Bocuse research center is not neutral). A closely related consideration is that one situational CLT design represents only one context. Organizing such situational CLTs requires resources (large rooms with modular space, budget for equipment, decoration, and so on). Also, playing with physical means makes the implementation less flexible and it is difficult to switch from one context to another. As a result, only one recreated context is usually tested, or at least only one context at a time, whereas a given product may be used in multiple contexts and with different uses. Memories and associations may thus vary a lot from one context to another. Besides, recreating a context, even in a fully realistic and functional way as in a living lab, has its own limitation because it is standardized, while in their daily lives, consumers may experience different versions of that context. For example, an experimental kitchen would certainly place people in a kitchen where they can prepare, manipulate, and eat food as they would normally do (Fig. 22.4). But there is little chance that this kitchen would look like their own kitchen (and actually like anyone’s kitchen). In other words, setting up situational CLTs doesn’t leave much scope for customization of the testing context. As a result of this limitation, several researchers have proposed the use of evocations rather than physical means to immerse participants into a relevant context, using either predefined scenarios (Delarue & Boutrolle, 2010; Hersleth, Monteleone, Segtnan, & Næs, 2015) or autobiographic recall (Hein, Hamid, Jaeger, & Delahunty, 2010; Jaeger et al., 2017).

22.2.2 Challenges and added value of real-life studies 22.2.2.1 Product experience and need for real-life testing Numerous situations cannot be tested under standardized conditions and yet they would provide extremely useful and salient information to product developers. Testing products in a central location gives only a partial view of reality whereas stepping out of the laboratory and experiencing real life for an extended duration is sometimes essential. Some aspects of context are difficult to recreate or even to anticipate. Again, this is particularly striking when testing nonfood products for which the notions of

464

Context

Fig. 22.4 Test kitchen of the AgroParisTech living lab.

product performance and product experience are essential. Consider for example these questions: How easy is it to apply sunscreen when sunbathing on a beach and in contact with sand? How easy is it to open snack bar packaging with only one hand when hiking or biking? How is the sensation of glide provided by a surfboard perceived? Social context may also play a role, because many products are worn or displayed in front of others and convey an image of self. For all these reasons, when running a test, it is important to allow consumers to experience the product in different situations, different ways of interacting with the product depending on the context, the intended use, the presence of others, etc. Hence, contrary to the standards of sensory evaluation of food, where efforts are made to separate participants in order to collect individual and independent data, social interactions are sometimes interesting to include in the test design. Interactions among participants may even be necessary when evaluating products such as sport equipment because many sports are played collectively or at least need a partner (e.g., tennis) (Fig. 22.5). In fact, even in the case of food, research on eating behavior has largely demonstrated the social influences on food intake (see for example Higgs & Thomas, 2016). Thus, depending on the product type and on the objectives of the study, food may also be tested in meal contexts that would allow social interactions (King, Weber, Meiselman, & Lv, 2004). One often underestimated element of context is duration. When testing food products for example, CLTs usually imply relatively short sessions (about 15–20 min and up to 1 h in a meal condition). The time span allocated to experiencing a product is thus

Conducting contextualized and real-life product tests: Benefits and experimental challenges

465

Fig. 22.5 Examples of testing contexts for sport equipment.

limited. For some products that are used or worn all day long, CLTs are thus very restrictive and only part of the product experience can be assessed, including for some foods or cosmetics. How to test a long-lasting lipstick in a 20 min session? How can you anticipate makeup interaction with objects in everyday life (e.g., glasses, cloths, food)? More generally, there is often a need to evaluate the evaluation of product performance with time, and to consider wear-off (on the product side) and fatigue (on the human side). In such cases, manufacturers need to run extended use tests that are conducted for a long period of time and can go beyond just the initial use of a product. These tests naturally imply testing in real life conditions.

22.2.2.2 Examples of real-life tests of products other than food In addition to what has been exposed, real-life tests of nonfood products can be conducted to address two important types of issues: measuring the impact of outdoor context on product performance and product perception by consumers, and evaluating the evolution of product performance with time. Below are some typical examples of tests implemented by Eurosyn where real-life testing conditions were necessary. Outdoor contexts—These tests are implemented to evaluate the impact of the natural environment on product performance under various climactic conditions: -

D
ecathlon, a French manufacturer of sport equipment, develops products that must guarantee performance and wellbeing of its customers during training and practice. D
ecathlon tests about 3000 prototypes every year. Naturally, many instrumental tests are conducted, but consumers’ perception of performance, comfort, and wellbeing must also be measured. For some sports, practice conditions may be quite extreme and equipment needs to be tested in Arctic conditions (Lapland, Fig. 22.6). Indeed, some perceptions may only be experienced at 20°C. This explains why it is necessary to organize such expeditions to obtain relevant information. The test lasts a week and participants should live the full experience. They hike, set up a camp, cook their food, etc. This is straining and fatigue must be taken into account. “Under such extreme conditions, you cannot improvise and product performance must be particularly well-adapted. In a way, your sleeping bag is your house. You should be able to eat, to keep your clothes and your shoes warm! Otherwise everything is frozen the next day. A zip that breaks or that cannot be tightened easily may have critical consequences”,

466

Context

Fig. 22.6 Example of a product test setup by Eurosyn team in extreme conditions in an outdoor context (Lapland).

-

-

says Karim Zameziati, product manager for the Quechua division of D
ecathlon, who organizes treks with consumers to evaluate prototypes of clothes and material (tents, sleeping bags, etc.). Consequently, in order to follow participants, to take notes and record information in an efficient manner, experimenters should not only be trained appropriately, but they should also prepare and anticipate those specific material conditions. Another manufacturer who wants to check that its sunglasses provide mountain bike riders a proper and comfortable vision must anticipate all practice situations and all-weather conditions. In such a case, a field study is necessary to compare the performance of different tints and different technologies. This means that the experimenter must design routes that allow participants to practice biking with varying contexts (trails, hills and bumps, undergrowth, etc.). The routes must provide experience with specific conditions, with some degrees of freedom and yet being standardized enough to allow comparison of responses over different participants and over different sessions for within-subject comparisons of materials. Using a similar approach to study consumer appraisal of cars while driving, Astruc, Sieffermann, Delarue, Danzart, and Blumenthal (2007) designed a route intended to be representative of the various types of roads and events (e.g. roundabouts, speed bumps, etc.) that a driver may encounter. This way, researchers could test cars under real-life and yet reproducible conditions. Testing material for snorkeling has been another challenge. The objective was not only to test the primary functionality of a new system (i.e. breathing underwater) in the quiet context of a swimming pool (Fig. 22.5), but also to measure consumers’ perceptions in natural environments, in the sea, in cold or warm water, with or without waves, etc. In such a case, the critical point was to manage to replicate tests with a common protocol in locations that are wisely selected so that the material is comfortable in all the circumstances mentioned above.

22.2.2.3 Challenges of real-life tests As could be expected, implementation of a real-life test is more demanding than a CLT, even a contextualized one. There is a clear need for time and resources. However, with proper organization, impact of testing in real-life conditions on costs remains limited. In addition, contrary to common beliefs, real-life tests are not restricted to qualitative studies or to small panels. For instance, we have tested bikes or surfboards in natural environments, with more than 120 participants in each case.

Conducting contextualized and real-life product tests: Benefits and experimental challenges

467

The most critical points for implementing such tests are anticipation and logistics. Everything should be carefully prepared: from testing and recording material (including waterproof cameras, sensors, etc.) to insurances; not to mention the training of experimenters and technical supervision with experts and/or medical staff. Of course, when planning a test under specific environmental conditions, in specific places, possibly in foreign countries, good connection with local teams is essential. It should be noted that, in our experience, these tests reveal such a great diversity of behaviors that there is a risk of being overwhelmed by data, if the test is not focused on precise objectives and if the typology of behaviors or responses is not properly anticipated. One should distinguish between supervised tests in real-life conditions from purely observational field studies. In the first case, the consumers interact with a product in a natural context, but do so under the experimental design of the researcher. Participants may thus follow a specific protocol and answer to precise questions. In addition, participants are specially recruited to test the products in that given context, according to criteria that are defined with stakeholders. On the contrary, in purely observational studies, as underlined by Payne and Wansink (2010), it is not possible to systematically test variables thought to influence behavior. Participants may not even be aware of the goal of the study. These tests are the most realistic and they provide rich data, but conclusions may suffer from patchy data. Besides, experimenters should collect data systematically in order to check for possible confounding effects.

22.2.2.4 Other options A middle ground between standardized CLT and real-life tests is to test products in rented facilities such as houses, retail stores, movie theaters, car showrooms, etc. In such cases, the advantages and limits of testing products this way are similar to those of recreated contexts and living labs, the main difference being that investment is less important and that chances to reach a good degree of realism are higher. Naturally, renting such facilities may still represent high additional costs, compared to a standard CLT.

22.2.3 Challenges and added value of immersive technologies In the view of bringing context to the laboratory, consumer and market researchers have seized the opportunities offered by the boom of digital technologies to attempt more immersive approaches, using either virtual reality (VR) or multi-sensory immersive rooms. Actually, fully immersive virtual reality caves have been used for many years to help product design or to train operators in aerospace and armaments industry and more recently in the car industry. Such structures are however extremely expensive and only very large companies or research universities could afford them. It thus comes with no surprise that more affordable, agile, and responsive devices (like VR headsets) raise huge expectations. These technologies spark a real enthusiasm, as they are expected to overcome some limitations of recreated contexts using physical

468

Context

means only. Many teams are thus experimenting with several approaches. Naturally, part of the excitement comes from the attraction to new technologies, but we think that they truly offer interesting perspectives.

22.2.3.1 Immersive rooms First, in line with immersive experiences provided by 3D movies or attractions in theme parks, the use of large high definition video screens is now affordable to most research teams and can be quite easily implemented to test products. Pictures or movies of natural usage environments can thus be displayed, possibly in combination with other sensory stimuli (smell, temperature, lighting, breeze, ambient sounds or music). Applying this approach for the first time, Bangcuyo et al. (2015) found that hedonic data collected in a virtual coffeehouse were more discriminating and a more reliable predictor of future coffee liking than those collected in traditional sensory booths. Using the same settings to evaluate the liking for cookies in a virtual home kitchen, Hathaway and Simons (2017) further showed that the discriminability and reliability of consumer acceptance data improves when using more complete level of immersion. Following similar approaches, many teams are now experimenting with such immersive technologies in the view of obtaining better quality data (Sinesio et al., 2018). However, the opportunities and limitations associated with the use of immersive technologies in product testing remain largely unexplored. For example, to date researchers have not taken full advantage of the flexibility allowed by these techniques and few have compared various virtual contexts. In a study using a multisensory immersive room to test flavored alcohol-free beers, Brasset, Gachet, Abiven, and Delarue (2017) found that some products would better match a given context and drinking experience (i.e., on a beach or in a night club) than others, hence providing insights for better market positioning. In addition, providing information in a dynamic and interactive way using recorded actors or avatars has been barely investigated so far (Liu, Hooker, Parasidis, & Simons, 2017). Another perspective is the use of these techniques to replicate a given context simultaneously in different locations, possibly different countries. This would give way to new approaches to cross-cultural studies. Eventually and perhaps most important, in their first experiment, Simons and his team have shown that participants’ engagement in the test significantly improved in the immersive condition, an outcome that likely also contributed to improved data quality (Bangcuyo et al., 2015). Interestingly, this finding has raised interest among sensory and consumer scientists for consumers’ mindset when they participate in our tests and we may hope that future research will further address this issue.

22.2.3.2 Virtual reality Another technology that triggers a strong interest among consumers and market researchers is virtual reality. Popularized by the videogame industry, VR headsets are now available to anyone and are also being experimented with quite extensively as a way to contextualize product testing (see for example: Ischer et al., 2014;

Conducting contextualized and real-life product tests: Benefits and experimental challenges

469

Porcherot et al., 2018). This will be developed in the next chapter and thus won’t be detailed here. Nevertheless, we would like to underline a few points one should be aware of when implementing this type of technology. First and foremost, VR is amazingly immersive, which makes it so popular for videogames. This goes to the point that some people cannot stand it and feel dizzy when wearing the headset, which could be one limitation with the technology currently available. Besides, it should also be noted that, to date, its use for providing context in routine tests is limited for two reasons: first, because designing a virtual context requires advanced computer artist skills and talented people rather work for entertainment industries than for consumer research; second limitation is that realism of VR environments remains pretty basic and unnatural. In other words, immersion should not be confused with realism. In addition to this, current headset devices are still heavy and make it difficult to track what you are eating, which thus limits its application to testing food. This being said, the extremely fast evolution of technologies may provide solutions in a very near future and, in spite of these reservations, we can see interesting perspectives for the use of VR to test products. VR indeed gives more flexibility and makes it potentially very easy to switch from one context to another. In theory, VR is totally configurable and would thus allow designing an infinity of virtual environments. Besides, as for immersive rooms, VR offers perspectives for testing products with the same virtual context in different locations. In high tech and heavy industry, VR is also used to allow somehow experiencing future products that are expensive to make or even to model. Digital modeling is thus a way to test various design factors, but also to orchestrate the experience of products that do not exist yet. Beside these uses of VR, one should also note that the emergence of self-driving and connected cars is leading the automotive industry to develop enhanced reality media to deliver new types of information to product users. This may also be conducive to new ways to test products, integrating context in a completely new manner.

22.3

Conclusions

This chapter intended to provide a review of various ways to conduct consumer tests in real-life conditions (or to make test context closer to real life) with the view of getting the most of the consumer-product-context interaction. We aimed to present the lessons we drew from 20 years of experience in implementing a variety of tests in diverse conditions, from sensory labs to Lapland, with many different product types and partners. Note that we do not consider one approach to be intrinsically better than another. Rather, we intended to share the view that each type of method has advantages and limits, as summarized in Table 22.1. To a certain extent, new technologies break up the usual trade-off between control and realism. Several possibilities are thus offered and constitute a portfolio in which consumers and market researchers will have to choose their method depending on their goals and constraints. The most important point in this respect is to be aware of what one wants to learn from the test.

470

Context

Table 22.1 Summary of the main advantages and limits of three contextualizing approaches Main advantages

Main limits

Recreated contexts, Contextualized CLTs

-

Contextualized yet controlled conditions More affordable Easier to recruit Social interactions are possible

-

Real-life testing, Field studies

-

Most realistic, participants can truly experience the product Wealth of information for designers and developers Access to the diversity of behaviors Participants truly enjoy the tests and are really engaged Very immersive

-

-

VR and Immersive techniques

-

-

Possible to test nonexisting objects or situations (disruptive innovation) Flexible Easy to replicate in different locations

-

-

-

Participants still come to a test facility; expectations and satisficing biases Participants are not “fooled” Short sessions Not all elements of context can be provided Requires important logistical undertaking Diversity of situations and behaviors may be overwhelming Requires agility and flexibility in face of unexpected events

-

Difficult to implement (technological constraints)

-

Not always realistic Some people cannot stand it (VR)

For this reason, we have emphasized product experience. In line with Hekkert and van Dijk (2011), we share the view that products obtain meaning through interaction with people, and that the context determines the appropriateness of any interaction. This means that to understand the relationship between the product and its user, one should leave room for product experience. This can only be achieved in proper context and this may be especially relevant for evaluating innovative products or products in the early stages of development. To this end, many consumer research methods are available (see for example van Kleef, van Trijp, & Luning, 2005). Our point is that, regardless of the method that is used, either quantitative or qualitative for instance, selecting a context that allow consumers to experience the products is essential. In the very first stages of a new product development project, observing behaviors in various and realistic contexts provides critical insights. It also gives designers, developers and all project teams a “solfeggio”, a common grid for analyzing product appraisal. When correctly conducted, this becomes an invaluable tool for further stages of development, as all tests and analyses could then be conducted using that grid. When more advanced prototypes or versions of the products are being tested, quantitative tests may be usefully completed by qualitative interviews in order to get further understanding of what works or not.

Conducting contextualized and real-life product tests: Benefits and experimental challenges

471

In addition to this, it must be stressed that implementing tests in realistic context gives an opportunity to film the interaction between the participants with the product which is not only a great way to record and to code behaviors and expressions, but also represents a powerful tool for communicating to stakeholders. Seeing people using— and in some cases misusing—a product is much more eloquent than reading data reports. When discussing with top-management, it may be the reason for choosing this test type.

References Astruc, C., Sieffermann, J. M., Delarue, J., Danzart, M., & Blumenthal, D. (2007). The influence of assessment context on the consumer responses in a driving situation. In Paper presented at the 7th Pangborn sensory science symposium. Minneapolis: MN. Bangcuyo, R. G., Smith, K. J., Zumach, J. L., Pierce, A. M., Guttman, G. A., & Simons, C. T. (2015). The use of immersive technologies to improve consumer testing: The role of ecological validity, context and engagement in evaluating coffee. Food Quality and Preference, 41, 84–95. https://doi.org/10.1016/j.foodqual.2014.11.017. Bergvall-Ka˚reborn, B., Eriksson, C. I., Sta˚hlbr€ost, A., & Svensson, J. (2009). A milieu for innovation: defining living labs. In: Paper presented at the ISPIM innovation symposium: 06–09 December 2009. http://urn.kb.se/resolve?urn¼urn:nbn:se:ltu:diva-31540. Brasset, A.-C., Gachet, J., Abiven, F., & Delarue, J. (2017). Taking control of product testing context thanks to a multi-sensory immersive room. A case study on alcohol-free beer. In: Paper presented at the 12th Pangborn sensory science symposium, Providence, RI. Brinberg, D., & McGrath, J. E. (1985). Validity and the research process. Beverly Hills, CA: Sage Publications. Delarue, J., & Blumenthal, D. (2015). Temporal aspects of consumer preferences. Current Opinion in Food Science, 3, 41–46. https://doi.org/10.1016/j.cofs.2014.12.001. Delarue, J., & Boutrolle, I. (2010). The effects of context on liking: Implications for hedonic measurements in new product development. In S. R. Jaeger & H. MacFie (Eds.), Consumer-driven innovation in food and personal care products (pp. 175–218). Cambridge: Woodhead Publishing. Delarue, J., Masson, M., & Blumenthal, D. (2018). Consumer evaluation of nonfood products. In: G. Ares & P. Varela (Eds.), vol. 2. Methods in consumer research (pp. 411–426). Cambridge, UK: Woodhead Publishing. Dell’Era, C., & Landoni, P. (2014). Living lab: A methodology between user-centred design and participatory design. Creativity and Innovation Management, 23(2), 137–154. https://doi. org/10.1111/caim.12061. Garcı´a-Segovia, P., Harrington, R. J., & Seo, H.-S. (2015). Influences of table setting and eating location on food acceptance and intake. Food Quality and Preference, 39, 1–7. https://doi. org/10.1016/j.foodqual.2014.06.004. Giboreau, A. (2018). Situational factors and the design of in situ evaluations. In: G. Ares & P. Varela (Eds.), vol. 2. Methods in consumer research (pp. 109–116): Elsevier. Hathaway, D., & Simons, C. T. (2017). The impact of multiple immersion levels on data quality and panelist engagement for the evaluation of cookies under a preparation-based scenario. Food Quality and Preference, 57, 114–125. https://doi.org/10.1016/j.foodqual.2016.12.009. Hein, K. A., Hamid, N., Jaeger, S. R., & Delahunty, C. M. (2010). Application of a written scenario to evoke a consumption context in a laboratory setting: Effects on hedonic ratings. Food Quality and Preference, 21(4), 410–416. https://doi.org/10.1016/j.foodqual.2009.10.003.

472

Context

Hekkert, P., & Schifferstein, H. N. J. (2008). Introducing product experience. In Product experience (pp. 1–8). San Diego: Elsevier. Hekkert, P., & van Dijk, M. (2011). VIP vision in design. A guidebook for innovators. BIS Publishers. Hersleth, M., Monteleone, E., Segtnan, A., & Næs, T. (2015). Effects of evoked meal contexts on consumers’ responses to intrinsic and extrinsic product attributes in dry-cured ham. Food Quality and Preference, 40, 191–198. https://doi.org/10.1016/j.foodqual.2014.10.002. Higgs, S., & Thomas, J. (2016). Social influences on eating. Current Opinion in Behavioral Sciences, 9, 1–6. https://doi.org/10.1016/j.cobeha.2015.10.005. Hinton, E. C., Brunstrom, J. M., Fay, S. H., Wilkinson, L. L., Ferriday, D., Rogers, P. J., et al. (2013). Using photography in ‘the restaurant of the future’. A useful way to assess portion selection and plate cleaning? Appetite, 63, 31–35. https://doi.org/10.1016/ j.appet.2012.12.008. Holthuysen, N. T. E., Vrijhof, M. N., de Wijk, R. A., & Kremer, S. (2017). “Welcome on board”: Overall liking and just-about-right ratings of airplane meals in three different consumption contexts—Laboratory, re-created airplane, and actual airplane. Journal of Sensory Studies. 32(2), e12254. https://doi.org/10.1111/joss.12254. Ischer, M., Baron, N., Mermoud, C., Cayeux, I., Porcherot, C., Sander, D., et al. (2014). How incorporation of scents could enhance immersive virtual experiences. [Methods]. Frontiers in Psychology. 5(736). https://doi.org/10.3389/fpsyg.2014.00736. Jaeger, S. R., Fiszman, S., Reis, F., Chheang, S. L., Kam, K., & Pineau, B. (2017). Influence of evoked contexts on hedonic product discrimination and sensory characterizations using CATA questions. Food Quality and Preference, 56, 138–148. https://doi.org/10.1016/j. foodqual.2016.10.003. Jaeger, S. R., & Porcherot, C. (2017). Consumption context in consumer research: Methodological perspectives. Current Opinion in Food Science, 15(Suppl C), 30–37. https://doi.org/10.1016/j.cofs.2017.05.001. Kim, S.-E., Lee, S. M., & Kim, K.-O. (2016). Consumer acceptability of coffee as affected by situational conditions and involvement. Food Quality and Preference, 52, 124–132. https://doi.org/10.1016/j.foodqual.2016.04.008. King, S. C., Weber, A. J., Meiselman, H. L., & Lv, N. (2004). The effect of meal situation, social interaction, physical environment and choice on food acceptability. Food Quality and Preference, 15(7–8), 645–653. K€ oster, E. P. (2003). The psychology of food choice: Some often encountered fallacies. Food Quality and Preference, 14(5–6), 359–373. https://doi.org/10.1016/s0950-3293 (03)00017-x. K€ oster, E. P. (2009). Diversity in the determinants of food choice: A psychological perspective. Food Quality and Preference, 20(2), 70–82. https://doi.org/10.1016/j.foodqual.2007.11.002. Liu, R., Hooker, N. H., Parasidis, E., & Simons, C. T. (2017). A natural experiment: Using immersive technologies to study the impact of “all-natural” labeling on perceived food quality, nutritional content, and liking. Journal of Food Science, 82(3), 825–833. https://doi.org/10.1111/1750-3841.13639. Meiselman, H. L. (1992). Methodology and theory in human eating research. Appetite, 19(1), 49–55. https://doi.org/10.1016/0195-6663(92)90235-X. Meiselman, H. L. (1996). The contextual basis for food acceptance, food choice and food intake: The food, the situation and the individual. In H. L. Meiselman & H. J. H. MacFie (Eds.), Food choice, acceptance and consumption (pp. 239–263). Boston, MA: Springer US.

Conducting contextualized and real-life product tests: Benefits and experimental challenges

473

Meiselman, H. L. (2013). The future in sensory/consumer research: Evolving to a better science. Food Quality and Preference, 27(2), 208–214. https://doi.org/10.1016/ j.foodqual.2012.03.002. Mela, D. J., Rogers, P. J., Shepherd, R., & MacFie, H. J. H. (1992). Real people, real foods, real eating situations: Real problems and real advantages. Appetite, 19(1), 69–73. https://doi. org/10.1016/0195-6663(92)90238-2. Parizel, O., Sulmont-Ross
e, C., Fromentin, G., Delarue, J., Labour
e, H., Benamouzig, R., et al. (2016). The structure of a food product assortment modulates the effect of providing choice on food intake. Appetite, 104, 44–51. https://doi.org/10.1016/j.appet.2015.11.018. Payne, C. R., & Wansink, B. (2010). Doing consumer research in the field. In S. R. Jaeger & H. MacFie (Eds.), Consumer-driven innovation in food and personal care products (pp. 358–385). Cambridge, UK: Woodhead Publishing. Petit, C., & Sieffermann, J. M. (2007). Testing consumer preferences for iced-coffee: Does the drinking environment have any influence? Food Quality and Preference, 18, 161–172. Pliner, P. (1992). Let’s not throw out the barley with the dishwater: Comments on Meiselman’s “methodology and theory in human eating research” Appetite, 19(1), 74–75. https://doi. org/10.1016/0195-6663(92)90239-3. Porcherot, C., Delplanque, S., Gaudreau, N., Ischer, M., De Marles, A., & Cayeux, I. (2018). Immersive techniques and virtual reality. In: G. Ares & P. Varela (Eds.), vol. 2. Methods in consumer research (pp. 69–83): Woodhead Publishing. Rolls, B. J., & Shide, D. J. (1992). Both naturalistic and laboratory-based studies contribute to the understanding of human eating behavior. Appetite, 19(1), 76–77. https://doi.org/ 10.1016/0195-6663(92)90240-7. Saint-Eve, A., Leclercq, H., Berthelo, S., Saulnier, B., Oettgen, W., & Delarue, J. (2016). How much sugar do consumers add to plain yogurts? Insights from a study examining French consumer behavior and self-reported habits. Appetite, 99, 277–284. https://doi.org/ 10.1016/j.appet.2016.01.032. Schifferstein, H. N. J. (2015). Employing consumer research for creating new and engaging food experiences in a changing world. Current Opinion in Food Science, 3, 27–32. https://doi.org/10.1016/j.cofs.2014.11.004. Sester, C., Deroy, O., Sutan, A., Galia, F., Desmarchelier, J.-F., Valentin, D., et al. (2013). “Having a drink in a bar”: An immersive approach to explore the effects of context on drink choice. Food Quality and Preference, 28(1), 23–31. https://doi.org/10.1016/j. foodqual.2012.07.006. Sinesio, F., Saba, A., Peparaio, M., Saggia Civitelli, E., Paoletti, F., & Moneta, E. (2018). Capturing consumer perception of vegetable freshness in a simulated real-life taste situation. Food Research International, 105, 764–771. https://doi.org/10.1016/ j.foodres.2017.11.073. Talbot, L., Delarue, J., & Sieffermann, J.-M. (2009). Taking context into account for the study of satiation and liking for biscuits. In: Paper presented at the 8th Pangborn sensory science symposium, Florence, Italy, 26–30 July. van Kleef, E., van Trijp, H. C. M., & Luning, P. (2005). Consumer research in the early stages of new product development: A critical review of methods and techniques. Food Quality and Preference, 16(3), 181–201. https://doi.org/10.1016/j.foodqual.2004.05.012. Zeinstra, G. G., Koelen, M. A., Kok, F. J., & de Graaf, C. (2010). The influence of preparation method on children’s liking for vegetables. Food Quality and Preference, 21(8), 906–914. https://doi.org/10.1016/j.foodqual.2009.12.006.

This page intentionally left blank

Inducing context with immersive technologies in sensory consumer testing

23

Patrick Hehn*, Dariah Lutsch†, Frank Pessel‡ *Marketing and Consumer Psychology, Harz University of Applied Sciences, Wernigerode, Germany, †Sensory & Consumer Insights, Symrise AG, Holzminden, Germany, ‡Fragrance Development, Henkel AG & Co. KGaA, D€usseldorf, Germany

23.1

Fundamentals of immersive technologies

Beyond the first chapter’s definitions, we define immersive technologies as methods and devices that induce targeted behavior in individuals by creating an effect of identification with immersive media (LaValle, 2017, p. 1; Schart & Tschanz, 2018, p. 20) through sensory stimuli. Immersive means “providing, involving, or characterized by deep absorption or immersion in something (such as an activity or a real or artificial environment)” (Merriam-Webster.com). Immersion is the technological quality of media and the created effect is called presence (psychological experience of being there) (Cummings & Bailenson, 2015, p. 2; Slater & Wilbur, 1997). Targeted behavior means that the user “is having an ‘experience’ that was designed by the creator. Examples include flying, walking, exploring, watching a movie” (LaValle, 2017, p. 1). Sensory stimulation is done by artificial or captured stimuli instead of ordinary input. The above definition of immersive technologies includes augmented and virtual reality as well as 360° videos. Although the same high-end headsets (loaded with tracking sensors that the simple phone-based platforms don’t yet have) can be used to show VR and 360° content, the main difference between both is that 360° videos lack interactivity or the viewer’s free movements in a virtual world (BBC R&D, 2018). The only “interaction” with 360° media is that users can move their head around within a captured spherical space or—at best—they can “interact” by clicking overlaid hyperlinks or questionnaires with a built-in or separate controller. The phonebased or simple headsets currently track the head’s orientation but not its position in space (Smith, 2015) so that users can only view the captured environment from the camera’s perspective. VR, on the other hand, allows users to pick up and use virtual objects, to open virtual doors, and change rooms etc. by the use of integrated and external sensor-based tracking systems and controllers (Aukstakalnis, 2017, pp. 195 ff., 214 ff). Thus, immersive systems can be distinguished at least by the dimensions interactivity and immersion, which both lead to a more or less intense feeling of presence.

Context. https://doi.org/10.1016/B978-0-12-814495-4.00023-4 Copyright © 2019 Elsevier Inc. All rights reserved.

476

Context

The most comprehensible approach to distinguish between AR and VR was introduced by Milgram, Takemura, Utsumi, and Kishino (1994) and refined by Milgram and Colquhoun Jr. (1999). The initial Reality-Virtuality Continuum places the real environment on the left side and the fully virtual environment on the right side (see Fig. 23.1). Everything in between can be called mixed or merged reality. The real environment disappears completely in the virtual environment (virtual reality) while in augmented reality, the portion of the real environment predominates. As the virtual environment is completely modeled at the computer while the real environment is unmodelled, the Extent of World Knowledge (EWK) Continuum was added later in the shape of a parallel line to the Milgram Continuum. EWK refers “to the extent of knowledge present within the computer about the world being presented” (Milgram & Colquhoun, 1999, p. 6) so that any environment on the Milgram Continuum corresponds to a location along the parallel EWK Continuum. If real and graphic images are combined, the world is partially modeled having less EWK as is the case with AR. VR display technologies that provide the users with the sensation of presence can be worn (head-mounted) or fixed (large semi-immersive or fully projection-based systems, e.g., computer-assisted virtual environments (CAVEs) and domes) (Aukstakalnis, 2017, p. 8). AR, on the other hand, “is a general term applied to a variety of display technologies capable of overlaying or combining alphanumeric, symbolic, or graphical information with a user’s view of the real world” (Aukstakalnis, 2017, p. 1). AR is well-known from head-up displays (HUDs), smart glasses, and AR apps for mobile devices. In this chapter, we focus on 360° media as well as augmented and virtual reality. All systems that are used to display immersive environments are visual, most of them also have integrated audio. Later we will also introduce some recent developments that enhance the more or less immersive environment by addressing the other senses. With this, the basic idea of virtual contexts from 1951 may become realizable within the next years: In his short story “The Veldt”, Ray Bradbury (1981, p. 10) described a virtual multisensory scene with 360° vision, sound, smell, temperature, and wind (touch) that sounds like a present day experience. Although the author described the idea of virtual reality, the technological term was introduced by Jaron Lanier in the 1980s. The term itself may date back to German philosopher Immanuel Kant (LaValle, 2017, p. 5), who said that we don’t perceive the things around us themselves (what Kant calls “local presence”) but we perceive things just as mental objects (what he calls “virtual presence”) (von Soemmerring, 1796, p. 82).

Fig. 23.1 Simplified representation of the Reality-Virtuality Continuum (Milgram et al., 1994, p. 283).

Inducing context with immersive technologies

477

The usage of immersive technologies in product tests and beyond basically needs two steps: media creation and media presentation. Creating media means to take a 360° video or to program augmented and virtual objects and environments that can be presented afterwards with appropriate devices. Both will be considered in the following description of selected available technologies. Besides creating a test environment, immersive devices can also be used to present the test products themselves.

23.1.1 Audio-visual devices The simplest way to create and present an immersive environment would be to shoot a life scene from one or more perspectives with one or more common video cameras and to use video projectors in order to create a CAVE-like atmosphere. The “CAVE” can additionally be equipped with real furnishings from the original environment (see part 3 of this chapter for examples). CAVEs and domes can also be used to present VR content. Both are classified as semi-immersive (Onyesolu and Eze, 2011, pp. 57f). CAVEs are based on multisided, rear-projection, or flat panel displays while domes use large-format hemispherical projections (Aukstakalnis, 2017, pp. 115–119). Another immersive level would be 360° technologies. Consumer-oriented devices are affordable and enjoy increasing popularity. Photographs and videos can be captured with 360° cameras that are usually equipped with two 180° lenses and a microphone. Common devices are Samsung Gear 360, Ricoh Theta, and Insta360. Having a tripod socket, they are usually used as stand-alone cameras being controlled by an app. Many Insta360 models can also be connected directly with the smartphone’s plug socket. Most manufacturers deliver their cameras with editing software that is needed to stitch both 180° videos, to cut, and save them in the required video format. Some software also allows sound editing. The resulting video or picture can be presented on a non-immersive desktop computer or smartphone display whereby the mouse or the touch display is used to look around in the captured environment. It can also be presented with a smartphone that is attached to a VR headset (e.g., Samsung Gear VR headset) or with PC driven head-mounted displays (e.g., HTC Vive, Oculus Rift) that also allow for directional audio. The user looks around by moving his head. PC driven devices are the number one choice for presenting VR content because such systems use additional controllers for navigation and interaction as well as tracking position and motion. For details of different types of fully immersive displays see Aukstakalnis (2017, pp. 103–120). Unfortunately, the creation of virtual space or test products needs specialists with experience in computer-aided design and geometric modeling software (Aukstakalnis, 2017, p. 15). Augmented content is also programmed by an engineer or software designer. The presentation ranges from smartphone apps to binocular (e.g., Epson Moverio, Microsoft Hololens, Sony SmartEyeglass) and monocular (e.g., Google Glasses) augmenting displays (for details see Aukstakalnis, 2017, pp. 77–102). Augmenting displays are stand-alone devices with transparent glasses or eye shields (optical see-through display) that are used for projections. Most AR displays currently address mainly developer and business solutions. Smartphone AR apps usually use the

478

Context

phone’s camera and add virtual overlays (information, hyperlinks, shapes, objects, colors, textures etc.) to the picture.

23.1.2 Tactile and kinesthetic devices Devices that address the sense of touch simulate kinesthetic and tactile cues and allow interaction with the immersive environment by grabbing virtual objects. There is an increasing number of interfaces that enable the user to perceive mechanical stimuli (pressure, flutter, textures, vibration) addressing the tactile sense as well as tension and other muscular forces addressing the kinesthetic sense (Aukstakalnis, 2017, pp. 179f). Tactile feedback devices use small vibration motors to produce the expected sensation. The most promising device is the Gloveone interface. This lightweight glove can be used with a head-mounted display or a computer screen and allows detailed tactile interaction with virtual objects by the use of integrated vibrotactile actuators and sensors. The sensors are needed to track finger movements, hand rotation, and device position while the 10 actuators let the users feel virtual objects, differentiate between textures, interact with buttons and elements, as well as trigger actions like a smart controller.

23.1.3 Olfactory devices The sense of smell (see Chapter 18) can be addressed for olfactory enhancement of the immersive context but with smell it might be useful to provide test fragrances by scent emitting devices. Environmental smell could be used in virtual restaurants or coffee shops or to emit new product smells of cars, wooden furniture etc. As many products such as detergents or toiletries are perfumed, it would be possible to use olfactory devices rather than real test products if required. In combination with other VR and AR devices, one could simulate the multisensory use of home appliances in virtual environments, e.g., sniff test while virtually doing the laundry. A device called Vaqso VR attaches to the bottom of any VR headset and is capable of emitting, muting, and switching between up to five scents instantly. Feelreal is a sensory mask that creates various physical stimulations such as temperature, wind, water mist, vibration, and smell. The mask is mounted to a VR headset and its odor generator includes nine removable smell cartridges. The biggest challenge with olfactory devices is to simulate the product smell (quality, intensity of the perfume in combination with the other ingredients) as realistically as possible.

23.1.4 Other devices The sense of taste is not relevant in creating an immersive context but it may be useful to address when it comes to test a virtual product’s taste, aroma, or mouthfeel. The Project Nourished merges the physicality of molecular gastronomy with VR. Its purpose is mainly health oriented and supports weight loss, allergy and diabetic management, eating therapy, and elder care. “3D printed food” serves as a vehicle for simulating taste, texture, and consistency, a “VR headset” provides visual simulation

Inducing context with immersive technologies

479

of aesthetics and environment, the “aromatic diffuser” uses ultrasonic waves and heat to dissipate the food smell, a “bone conduction transducer” mimics chewing sounds, a “virtual cocktail glass” with built-in sensors is a vessel for beverages and finally the “gyroscopic utensil” translates the diner’s physical hand movements into VR by simulating food intake (www.projectnourished.com). This system may also be suitable for taste, aroma, and texture testing. Walking around in a virtual environment is usually tracked by external sensors in the room. With Virtuix Omni a VR motion platform is available that enables virtual movements by real movement on the spot. Similar devices are the Roto motorized interactive VR chair that uses touch pedals for walking whilst seated and the Icaros that combines fitness and VR by letting users fly in a virtual environment. Its gyroscopic design allows three-dimensional movements within virtual worlds.

23.1.5 Conclusion The future challenge is to provide a realistic integrated experience that synchronously addresses all required senses in the immersive environment and additionally to reproduce test products virtually if appropriate. The current audio-visual systems are still in the infant stage (Aukstakalnis, 2017, p. 356), so multisensory integration should require years.

23.2

Immersive applications

In general, there are a lot of commercial application areas for immersive media. Aukstakalnis (2017, pp. 226–329) lists gaming and entertainment, architecture and construction, science and engineering, health and medicine, aerospace and defense, skills education and knowledge acquisition, and telerobotics and telepresence. With relation to product design, the Swedish furniture store IKEA introduced AR with its 2014 catalogue. People placed the printed IKEA catalogue where they wanted to put the furniture in the room. The IKEA app activates the smartphone camera, which shows a live video of the room. The selected furniture is automatically laid over the catalogue so that respondents get an impression of how the furniture would look in their room. A similar feature called “AR view” was launched at the end of 2017 as part of the Amazon app. Based on the same live video principle as the IKEA app, AR view allows you to view thousands of products in the consumers’ home before they buy them. In 2015/16, L’Oreal launched its Makeup Genius, an AR app that allows users to virtually try decorative cosmetics using nothing but their phone. The app scans the users’ face and allows them to select from the huge L’Oreal cosmetics range either using the app catalogue or by scanning the product barcode at the point of sale. The selected product is virtually applied on the users’ face (lips, cheek, eyes) so that they can evaluate many different existing products on their own face within a short time. A similar app was introduced by Schwarzkopf Professional at the CES 2018. The Schwarzkopf SalonLab Consultant App includes AR technology that enables

480

Context

the hairdresser to virtually color the client’s hair on a tablet computer, which helps to select suitable hair coloration (www.schwarzkopfpro.com/salonlab/). In terms of mass customization, shoppers can design their own shoes at the Nike flagship store in Paris since 2017. The user takes a white blank of the desired model and puts it into the AR tool, which is used to project different colors and textures onto the blank. Projections are controlled by a tablet computer. The advantage for the company is that they can gather a lot of information about the consumer preferences. A similar technique is offered by agencies that use laser and video projections to augment product blanks for product development and visual evaluation. They visualize the appearance of athletic shoes, car seats, etc., which is useful for market research. This approach needs just one prototype instead of different designed test products. As consumers use and finally assess products in a natural context, researchers want to know if the natural context can be simulated in an as much as possible realistic virtual or augmented reality as this would allow a realistic test setting with controlled conditions. For example, the German railroad company Deutsche Bahn tested future design alternatives for train interiors. Respondents were able to move through the virtual prototypes with a VR headset (Hellwig, Girard-Reydet, & Scheckenbach, 2015). Immersive technologies can provide almost unlimited environments that can be used for design and product evaluation, they can augment reality by adding virtual objects to the real world and they can also simulate certain test products whose characteristics are the object of interest. But there are still limits due to time and costs of extensive programming of virtual worlds and objects, due to technical limits of interaction and a lack of multisensory impressions. Due to these restrictions, the following studies from the field of sensory product research used rather low-immersive techniques but reached acceptable levels of presence to gather first insights.

23.3

Benefits and restrictions of testing in context induced by immersive media—Learnings from different case studies

Immersive media as an upcoming technology is attracting increasing attention in traditional market and sensory research studies (e.g., Bangcuyo et al., 2015; Kim, Lee, & Kim, 2016). Originally, sensory research was predominately characterized by laboratory-based product tests. It was argued that standardized conditions minimize the so called “noise”, allowing consumers to focus on the test products themselves. This makes it easier to identify even subtle sensory differences between the test products (e.g., Lawless & Heymann, 2010; Meiselman, 1992). Hence, a high degree of internal validity instead of realism was the ultimate goal. However, as illustrated by the various articles in this book, lately there has been an increasing trend of considering more contextual/environmental effects because consumers’ product choices and consumptions do not happen in a neutral environment but are influenced by various contextual cues (e.g., Boutrolle, Delarue, Arranz,

Inducing context with immersive technologies

481

Rogeaux, & K€ oster, 2007; Meiselman, 1992; Spence & Zampini, 2007). Many companies therefore often use in-home-use tests in which consumers test products in their natural environment. The downside of this approach is the considerably higher costs compared with central location tests (CLT). Also, as research regarding multi-modal Point of Sale or package design shows, product preferences can be highly influenced by ambient/contextual cues such as ambient acoustics (e.g., music) or ambient odor (e.g., Michon & Chebat, 2004, 2005; Salzmann, 2007) which further highlights the need of testing in the right context. A possibility to test products at a central location to keep costs down and the degree of standardization up and still to consider contextual cues is the use of immersive media in consumer testing. In order to learn and better understand the opportunities and limitations of using immersive media in sensory research, we conducted several case studies that varied different parameters such as tested product (food, beverages, and non-food products), degree of standardization (no standardization in in-home use tests to highly standardized in laboratory tests) as well as test setup (e.g., laboratory versus real life condition versus immersion setup) and the considered dependent variables (e.g., overall acceptance and consumer engagement). Six different case studies are presented below.

23.3.1 Beverage testing in three different setups In the first two case studies—a cappuccino and a beer tasting—the objective was to understand whether testing in different setups (sensory laboratory versus CAVE-like projection versus real life scenario) results in a different liking order of the different products. By liking order, the authors mean the order of the products when sorting the products according their overall liking mean scores, starting with the highest mean value.

23.3.1.1 Cappuccino study In this study, 104 consumers tested four different cappuccinos in three different settings: in a sensory laboratory, an immersive setup, and a real coffee shop (Glassl et al., 2016; see Fig. 23.2). The sample preparation was identical in all three setups. Although the same respondents participated in all setups, they were not aware of the fact that they tasted the same products. All products were rated on a 9-point hedonic scale, ranging from 1 “dislike very much” to 9 “like very much”. To understand the test setup main effect on product preference, the mean score across all four products was calculated per setup: sensory laboratory (mean 6.6) and coffee shop (mean 6.5) scored on a comparable level, with immersion (mean 6.2) significantly lower (ANOVA, Post-Hoc: Duncan, P < .05). More interesting, however, is the missing product by setup interaction effect (general linear model (GLM) with products and setups as fixed factors and respondent’s ID as random

482

Context

Sensory laboratory

Immersion

Set-up: controlled laboratory situation

Set-up: imitation of a natural coffee shop atmosphere while still guaranteeing high degree of standardization (e.g., standardized preparation (Barista) and environment) Contextual cues: more realistic optical and acoustical cues by depicting a coffee shop and typical “coffee shop noise”, no odor stimuli

Contextual cues: no effect of conextual cues by ensuring constant temperature, ligth etc. and no impairment by possible background noise or odor

Standardization

Coffee shop Set-up: real coffee shop with no standardization

Contextual cues: various environmental influences that could all affect the tasting such as odor impairment, nonstandardized background noise etc.

Real life

Fig. 23.2 Test setup description in the cappuccino study.

factor, P < .05), i.e., the rank order of the samples remains the same for all setups. This reflects a robust evaluation of the samples across all environments, which is in line with past research (e.g., Meiselman, 1992). This is a strong indication that although on different levels (natural environment likings are mostly on a higher level than the analogous scores gathered in laboratory testing) product order stays the same and makes the more expensive testing in natural environments unnecessary, especially when considering that most companies use benchmarks that are based on CLT results that are good indicators for the future product success. In a last step, the explained variance R2 (calculated in the previously mentioned GLM) for the fixed factor “setup” was calculated, which quantifies the degree of the product differences and is thus an indicator for the discriminatory power of each setup. Due to the absence of standardization of the coffee shop tasting, the discriminatory power is lower here than in the more standardized setups (see Fig. 23.3). So, when a high discriminatory power is desired, more standardized setups such as a sensory laboratory and the virtual reality setup should be chosen.

23.3.1.2 Beer study In the beer study, four different beer samples were tested in three different settings by the same respondents according an experimental within-design: in a sensory laboratory, a CAVE-like setup, and a real bar by 119 consumers in March/April 2016 (see Fig. 23.4). Again, the sample preparation was kept identical and a blind testing approach was used to mask that the same products were tested in all three setups.

Inducing context with immersive technologies

483

Overall liking per setting (1 = dislike very much // 9 = like very much; different letters stand for sign. different means; P < .05) Sensory laboratory

Immersion

Coffee shop

Mean: 6.2 (B)

Mean: 6.6 (A)

Mean: 6.5 (A)

Product 1

a

7.4

a

7.0

a

7.1

Product 2

a

7.3

a

7.0

a

7.2

Product 3

b

Product 4

c 1

R2

2

6.2

5.7 3 16%

4

5

6

7

5.7

b

5.1

c 8

9

1

2

3

4

5

6

7

18%

8

9

1

b

5.9

b

5.9

2

3

4

5

6

7

8

9

11%

Means | n = 104 per situation

Fig. 23.3 Results cappuccino study.

Fig. 23.4 Test setup description for the beer study.

Products were rated on a 9-point hedonic scale, ranging from 1 “dislike very much” to 9 “like very much”. To analyze the product, setup main and interaction effects, a GLM was carried out using products and setups as fixed factors and respondent’s ID as random factor. In this study, neither a setup main effect (P
.05) nor a product setup interaction effect (P
.05) was observed. Especially for very familiar products where consumers have a

484

Context

fixed view of the product, context main effects are often missing (see Cardello & Meiselman, 2016; Go, Kim, & Chung, 2017; Hersleth, Uelanda, Allaina, & Næs, 2005). Surprisingly, in this test the discriminatory power per test setup was overall quite low—with the highest R2 in the bar setup, which could be due to the familiarity of the product category (see Fig. 23.5).

23.3.1.3 Conclusion for the cappuccino and beer study When only considering the insights of these two studies, there are no strong indications in favor of testing in an immersive or even real-life situation since no product by setup interactions were identified. Also, the overall product liking level of laboratory testing and testing in natural environments were not significantly different in both studies, which deviates from other context studies in which laboratory testing compared to real-life testing often underestimates product acceptance (see Boutrolle et al., 2007; Boutrolle, Arranz, Rogeaux, & Delarue, 2005; Cardello & Meiselman, 2016). However, in the beer study, the discriminatory power of the real-life setup was higher than in the laboratory or CAVE situation whereas it was lower in the cappuccino study and this needs further clarification. Furthermore, it raises the question of which factors, besides standardization, influence the discriminatory power of a test setup.

23.3.2 Spread testing in five different setups The objective of this study was to understand the effect of test standardization and context cues on product liking, consumer engagement, and discriminatory power. In this spreads study, 800 respondents tested two spread products in five different setups in 2016. There were 160 respondents per setup. To understand the effect of Overall liking per setting (1 = dislike very much // 9 = like very much; different letters stand for sign. different means; P < .05) Sensory laboratory

Immersion

Bar Mean: 5.8 (A)

Mean: 5.5 (A)

Mean: 5.7 (A)

a

a

Product 1

a

Product 2

b

5.7

a

5.6

bc

5.6

Product 3

b

5.7

a

5.6

b

5.7

Product 4

b 1

R2

2

6.3

5.2 3 3%

4

5

6

5.9

5.1

b 7

8

9

1 2%

2

3

4

5

6

6.5

5.2

c 7

8

9

1

2

3

4

5

6

7

8

9

6%

Means | n = 119 per situation

Fig. 23.5 Results for the beer study.

Inducing context with immersive technologies

485

standardization and contextual cues on product liking and consumer engagement, we incrementally increased the external validity—which goes hand in hand with decreasing standardization and thus internal validity. CAVE-studies are generally not less standardized and thus have a lower internal validity. The special aspect of this study is that in the immersive CAVE setup, not only contextual cues such as visual and acoustical elements were integrated, but the respondents could also interact with each other, which of course lowers the degree of standardization. The sensory laboratory test in test booths had the highest degree of standardization followed by the “Foodie”, which was carried out in small groups at one table without separation. The third test took place in groups in a CAVE-like kitchen, which maintained a high degree of standardization but preserved the contextual cues. Closest to reality were the two in-home tests—one with detailed usage instructions and one without (see Fig. 23.6). Products were rated on a 7-point hedonic scale, ranging from 1 “very poor” to 7 “excellent”. Consumer engagement was measured in each of the five environments using a consumer engagement scale used by Bangcuyo et al. (2015). For the present study, two experienced researchers translated the questionnaire items into German before an independent native speaker conducted a back-translation (Table 23.3 in the Appendix shows the items used in the study). In this study, a GLM was carried out to analyze setup and product main as well as interaction effects. Neither a setup main effect (P > .05) nor a product by setup interaction effect (P > .05) was observed. Surprisingly, in this test the discriminatory power per test setup was overall quite low—with the highest R2 in the Foodie and CAVE-like setup and rather no discrimination in the two HUTs. Little or no Sensory laboratory

Foodie

Immersion

Set-up: controlled laboratory situation

Set-up: testing in one room with tables for four respondents, products were served under standardized conditions

Contextual cues: no effect of contextual cues by ensuring constant temperature, light etc. and no impairment by possible background noise or odor

Contextual cues: more realistic testing by sitting at tabels; neither odor nor acoustic stimuli

Set-up: imitation of a kitchen atmosphere while still guaranteeing high degree of standardization (e.g., standardized preparation & environment) Contextual cues: testing in one room with nicely prepared desks; a kitchen was displayed on 2 walls; respondents got their products themselves from two fridges; background music

HUT with instructions

HUT w/o instructions

Set-up: testing at the consumers’ home with detailed instructions given how to store and consume products

Set-up: testing at the consumers’ home with no standardizes procedure

Contextual cues: various environmental influences that could all affect the tasting such as odor impairment, non standardized background noise etc.

Contextual cues: various environmental influences that could all affect the tasting such as odor impairment, nonstandardized background noise, interaction with other people etc.

Instructions

Standardization

Fig. 23.6 Test setup description for the spreads study.

Real life

486

Context

discrimination in the HUTs (see Fig. 23.7) can be explained by a low degree of standardization. This results in a lot of background noise that covers product effects— which were rather small but still detectable by a sensory panel. Although having highest standardization in the sensory laboratory setup, we still see a greater amount of discrimination in the Foodie and CAVE-like setup. This can be explained by a higher consumer engagement. There is significantly higher consumer engagement in more natural testing environments than in artificial ones (see Fig. 23.8). This is reflected by the three dimensions of consumer engagement, which are derived from a pre-validated inventory (Bangcuyo et al., 2015). To calculate the different engagement scores (1CE “Involvement & Engagement”, 2CE “Impairment by unusual testing situation”, and 3CE “Aesthetics & support by test setup”) we ran a Overall liking per setting (1 = very poor// 9 = excellent; different letters stand for sign. different means; P < .05) Sensory laboratory Mean: 5.5 (C)

5.6

5.4

a

b

Product 1 Product 2

R2

Foodie

Immersion

HUT with instructions HUT w/o instructions

Mean: 5.3 (D)

Mean: 5.4 (D)

Mean: 5.9 (A)

5.4

a

b

b

a

Product 1 Product 2

3%

5.9

5.2

a

Product 1 Product 2

2%

5.9

5.5

5.2

Mean: 5.7 (B)

a

Product 1 Product 2

3%

5.7

5.7

a

a

Product 1 Product 2

1%

1%

Means | n = 160 per situation

Fig. 23.7 Results for the spreads study. Engagement score per setting (different letters stand for sign. different means between set-ups; P < .05) Sensory laboratory

Immersion

Foodie

1.05 a

0.99 ab

HUT with instructions

0.91 b

0.03 b a

b

a

HUT w/o instructions

b

0.27 a

0.12 b

a

0.22 a

0.07 b

a

a

a

a

–0.07 c –0.48 e

–0.35 c

–0.31 d

–0.30 c –0.61 c

CE1: Involvement & Engagement

CE2: Impairment by unnatural testing situation

–0.69

c

CE3: Aesthetics & support by test set-up

Means | n = 160 per situation

Fig. 23.8 Results of consumer engagement in the spreads study.

Inducing context with immersive technologies

487

factor analysis (principal component analysis with varimax rotation) and used the factor scores as engagement scores. 1CE refers to the degree the respondents were involved and immersed into the testing situation reflected by items such as “I lost track of time” or “being aware of events occurring in the real world”. 2CE expresses the degree of how much the respondents got distracted by the test environment. 3CE comprises items such as “the testing environment assisted my evaluations of the samples” or “the testing environment was appealing”. The different factors of consumer engagement do not only have a positive, or for 2CE negative main effect on product liking, 1CE and 3CE also influence the discriminatory power of each setup positively, resulting in highest discrimination in the Foodie and CAVE setups. Discrimination is operationalized using the difference between the two mean scores of the samples across all test setups. These two setups thus represent the best trade-off between rather low impairment by unnatural testing situation and good support by test setup (see Table 23.1).

23.3.2.1 Conclusion for the spreads study Overall, we can conclude for the spreads study that more standardized methods allowed for more insightful product discrimination whereas more realistic scenarios increased consumer engagement. Foodie and CAVE—being a good trade-off between standardization and still enabling a relatively high consumer engagement—provided the highest level of discrimination (although overall on a relatively low level). HUTs had high consumer engagement but too much background noise that covered product differences. Small effects were only detectable in the laboratory conditions. Now, one might argue that if respondents do not notice a difference in a HUT, those differences might not be relevant for the manufacturer. However, in a ranking task included in the Table 23.1 Impact of consumer engagement on liking and discrimination (discrimination between the products is calculated as absolute difference between mean score of both test products across all test setups; different letters indicate sign. different means)

1CE: Involvement & Engagement 2CE: Impairment by Unnatural Testing Situation 3CE: Aesthetics & Support by Test Setup

Product 1

Product 2

Discrimination between products

low

5.49 a

5.37 a

0.12

0%

high low

5.83 a 5.89 a

5.65 b 5.82 a

0.18 0.07

2%

high low

5.50 a 5.46 a

5.32 b 5.35 a

0.18 0.11

1%

high

5.86 a

5.67 b

0.20

Consumers were divided into two groups per engagement factor using median split.

R2

488

Context

HUTs, product 1 (the better liked one in the (more) standardized testing setups) scored significantly better than product 2—meaning when it comes down to a forced decision, minor effects could tip the scale and this is relevant to the business.

23.3.3 Yoghurt testing in five different setups In the spreads study we could see that a high degree of standardization went hand in hand with high consumer engagement, which resulted in a better discriminatory power of the setup. Hence, the purpose of this study was to find out whether bringing mixed reality into the laboratory is a possible substitution for testing in a natural environment or whether by a standardized setup in a room with VR components, i.e., by introducing mixed reality into the laboratory, we can achieve a high discrimination and high external validity at the same time. There were 167 consumers who tested four different yoghurts in two of the four different settings in December 2015/January 2016: in a sensory laboratory, in a sensory laboratory with mixed reality/immersion, in a CLT room with mixed reality/ immersion, and a canteen as a natural environment. Each respondent participated in two setups with differently coded samples (see Fig. 23.9). Four strawberry yoghurts were rated on a 9-point hedonic scale, ranging from 1 “very poor” to 9 “very good”. Consumer engagement was measured with the same pre-validated inventory that was used in the spread study (see Appendix). All respondents were regular consumers of strawberry yoghurt. Sensory laboratory

Lab + Mixed reality

Set-up:controlled laboratory Set-up: testing in a situation controlled lab situation while adding immersive elements; still guaranteeing high degree of standardization (e.g., standardized preparation & environment) Contextual cues: no effect Contextual cues: video of contextual cues by screen in booth displaying a ensuring constant canteen with according temperature, light etc. and ambient sound; constant no impairment by possible temperature, ligth etc. and background noise or odor no impairment by possible odor

CLT + Mixed reality

Natural environment

Set-up: imitation of a canteen atmosphere while still guaranteeing high degree of standardization (e.g., standardized preparation & environment)

Set-up: real canteen with no standardization but a standardized sample preparation

Contextual cues: testing in one room with nicely prepared desks; a canteen was displayed on one wall with according ambient sound

Contextual cues: various environmental influences that could all affect the tasting such as odor impairment, non standardized background noise etc.

Standardization

Fig. 23.9 Test setup description for the yoghurt study.

Real life

Inducing context with immersive technologies

489

Overall liking per setting (1 = dislike very much // 9 = like very much; different letters stand for sign. different means; P < .05) Sensory laboratory

Lab + Mixed reality

CLT+ Mixed reality

Natural environment

n = 91 Mean: 5.7 (A)

n = 97 Mean: 5.6 (A)

n = 80 Mean: 5.8 (A)

n = 66 Mean: 5.7 (A)

Product 1

a

6.4

a

Product 2

a

6.4

a

Product 3

b

Product 4

c

5.3

4.7

1 2 3 4 5 6 7 8 9 R2

14%

b

c

6.8

6.5

5.0

4.1

a

6.9

a

6.8

a

6.7

a

6.6

b

b

5.1

4.4

1 2 3 4 5 6 7 8 9

1 2 3 4 5 6 7 8 9

31%

27%

b

b

5.2

4.2

1 2 3 4 5 6 7 8 9 29%

Means | n varies

Fig. 23.10 Results of the yoghurt study—Overall liking.

In this study, there was neither a setup main effect (P
.05) nor a product by setup interaction effect (P
.05). In contrast to the spread study, in this test the discriminatory power per test setup was quite high overall—with the highest R2 in the laboratory + mixed reality (see Fig. 23.10). Interestingly, the sensory laboratory test revealed the lowest discriminatory power. This finding suggests—and is in line with the previous results—that a high discriminatory power is not only the result of a high degree of standardization (due to less noise) but is also influenced by other factor(s). As we could see in the spreads study, the consumer engagement positively drives discriminatory power—hence, this hypothesis needs to be further clarified. Unexpectedly, the natural environment had the lowest engagement scores (see Fig. 23.11); the CLT with mixed reality and not the canteen (natural environment) resulted in the highest consumer engagement scores. Furthermore, no positive correlation between consumer engagement and discriminatory power can be identified (r ¼  .041). Hence, in this study the hypothesis that consumer engagement is positively driving the discriminatory power of a test setup cannot be confirmed.

23.3.3.1 Conclusion for the yoghurt study As a key insight of the yoghurt study we can conclude that standardized setups with mixed reality elements generate the highest discriminatory power. Surprisingly, not only the discriminatory power of the highly standardized sensory laboratory test is—not as expected—lower than the natural environment setup but also the consumer engagement is higher than the natural environment. A reason for a high consumer engagement in the laboratory setup could be due to the fact that for students, testing in an unknown laboratory/CLT setup might have been more exciting than testing in the well-known canteen and thus have caused a higher engagement.

490

Context

Consumer engagement per setting (mean | n=334 | ANOVA, Post hoc: Duncan, sign. 0.05)

a CLT + Mixed reality (n = 80)

a

Sensory laboratory (n = 91)

ab

Lab + Mixed reality (n = 97)

b

6.7

4.0

Natural environment (n = 66)

b

6.6

4.9

1

2

3.9

8.4

4.0

7.3

3

4

5

6

7

8

9

10 11 12 13 14 15 16

Means | n varies

Fig. 23.11 Results of the yoghurt study—Consumer engagement.

23.3.4 Toilet rim blocks in three different setups Some perfumed products are preferably tested in special test facilities. These include toilet rim blocks that the perfume and home care industry usually evaluate in test toilet facilities. Solid toilet rim blocks are very common in Western Europe but not in the USA because the product needs a washdown flushing system instead of siphoning. The product hangs down in the bowl and water passes through the rim block basket with each flush (see Fig. 23.12). The solid rim block slowly dissolves in water and releases a pleasant fragrance that also diffuses into the restroom. Some varieties additionally use disinfectants or bleach. As market research institutes rarely have such test toilet facilities, the idea of doing a sniff test with toilet rim blocks in a “virtual” lavatory was born. A restroom video with sound (toilet flush that shows the product in action) was produced with a Samsung Gear 360 camera. The 30 s video was presented on a Samsung S6 smartphone that was fixed to the Samsung Gear VR glasses. There were 187 consumers that tested two different toilet rim blocks in three different settings: regular sniff test in a sensory laboratory where respondents sniffed the diluted products (n ¼ 60); test studio conditions showing the 360° lavatory video as a context prime while they sniffed the diluted products (n ¼ 60); in-home use test with 67 respondents who used the rim blocks at home (see Fig. 23.12). Products were tested one after another in varied order. In the laboratory and VR setting, we showed a picture of the test product at the beginning of the questionnaire because respondents at home saw the test product itself and we wanted to have the same start conditions in all settings. The restroom video was only presented to the VR group. In the laboratory and VR setting, respondents sniffed the diluted rim blocks from coded glass jars. At home, the fragrance was assessed from the original product in the respondents’ bath or restroom. These settings reflect the feasibilities of most sensory product research institutes.

Inducing context with immersive technologies

Sensory laboratory • Controlled lab situation • Elimination of environmental influences • Constant temperature, light etc.

491

360° device • VR glasses showing a 360° video • Higher realism • Still standardized preparation and constant environment

Inhome use test • Products were used at the respondent’s home • Various environmental influences that could affect scent evaluation

Standardization

Real life

Fig. 23.12 Test setup description for the toilet rim block study.

Odor liking of both test products was rated on a 9-point hedonic scale, ranging from 1 “don’t like at all” to 9 “like very much”. In odor liking we observed a significant main effect but there was no significant interaction between the fragrances and the test settings. Fragrance 1 was always liked more than fragrance 2. Thus, the discrimination between samples was given in each setting although the liking was lower in the in-home use test. Odor intensity was measured on a 5 point “just right” scale. The results in the laboratory and VR setting were rather unremarkable. But it is striking that the intensity of fragrance 2 was perceived as too weak at home while fragrance 1 slightly polarized at home. Thus, in terms of intensity, the field performance of fragrance 2 was lower than the performance under controlled conditions (see Fig. 23.13). At the end of the questionnaire, participants were asked about the study itself. The overall study liking did not differ significantly between test conditions. The 360° setting made respondents feel like being in a real lavatory and it significantly stimulated their sensory awareness compared to both other settings, both with P < .05 (see Fig. 23.14; scales on the basis of Bangcuyo et al., 2015; O’Brien & Toms, 2010; Witmer & Singer, 1998).

23.3.4.1 Conclusion for the toilet rim block study In overall liking, the same odor was preferred in all three settings. Thus, the discrimination between fragrances was given in each setting although the level of liking was lower in the in-home use test. But the 360° video proved to be a serious method if it is important to simulate a real usage context for certain products. Of course, for pure fragrance selection, the sensory laboratory condition is as good as the video setting. But the VR glasses enhance the

492

Context Odor intensity (5point just right scale)

Fragrance 1 Sensory Laboratory

360°

Inhome use-test

25

13

25

sum “too weak”

57

Fragrance 2 18

15

17

17

70

45

just right

30

62

53

67

23

30

28

sum “too strong”

% share | n varies

Fig. 23.13 Results for the toilet rim block study—Odor intensity.

sensory awareness, environmental aesthetics, immersion, and novelty perception, which may also result in a higher motivation of participants during the test. The in-home use test has some weaknesses in fragrance selection particularly in terms of internal validity and distractions. Nevertheless, in-home use tests are important to gather information about the product performance in real life before the product is launched. The weaker intensity of fragrance 2 could also arise with a winning fragrance. So, it is advisable to do both: first a fragrance selection test and second a fragrance performance test.

23.3.5 Potato chip testing in virtual and augmented reality setups In all research studies described so far, we either used optical (e.g., beaming a bar, a kitchen, or a canteen to one or two walls) and acoustical cues (e.g., sound of a canteen, radio etc.) to create an immersive environment or used VR glasses with acoustical input to put the respondents into a virtual situation. With the chips study, we wanted to learn more about the pros and cons of using VR glasses versus AR glasses. There were 77 consumers who tested one potato chip sample in two of the three different settings in July/August 2017: either in the sensory laboratory and AR setup or in the sensory laboratory and VR setup with differently coded samples (see Fig. 23.15). In this study, no setup main effect (P
.05) was identified. The test product received similarly good liking scores in all three setups, which is in line with the findings of the previous studies (see Fig. 23.16). However, as expected VR and AR involve and engage the consumers more into the test scenario (see Fig. 23.17). Overall, VR tends to provide even more immersive experience than AR, disconnecting respondents from the actual studio environment (P < .1). However, the advantage of AR versus VR is that consumers still have the possibility to see the test product, which does not only facilitate its consumption but allows an evaluation of its appearance.

4

Inducing context with immersive technologies

493

Sensory laboratory I took particular care in sniffing the samples.

A

The test situation stimulated my senses.

B

The test situation piqued my curiosity.

A

The test situation distracted me from the sniff test.

B 1.2

A

B 1.3

A

The sniff test was fun.

A

The test situation made me feel like being in a lavatory.

B

3.3

A

I lost track of the time during the test.

A

3.3

A

1.6

A

4.5

1

2

3

4

4.7

3

1.6 1.9

B

4.3 3.7

A

3.4 2

2.7

B

4.0

5 1

Don’t agree at all

2.8

B

4.0

B 1.3

B

4.1

B

3.6

A

3.7

B

4.4

A

3.2

The sniff test was boring.

* Different characters = sign. difference between test settings (p25.000 tours without distinct overalls color changes up to 7000 tours grade  3 up to 5000 tours grade  3–4 Trousers

Table 25.3 Test criteria, test method, and requirements of the care properties according the Hohenstein Quality Standard 702 workwear (Hohenstein, 2016) Test criteria Dimensional stability after 5 care cycles DIN EN ISO 15797 Crease recovery surface area after 3 care cycles DIN EN ISO 15797

Test method DIN EN ISO 3759, DIN EN 25077 DIN EN ISO 15487

Requirement PES/CO CO

  2.5%   3.0% 3–4

There are many standards available to determine the antimicrobial effect of antimicrobial textiles. The choice of test method depends on the textile, the antimicrobial agent, the expected use, and the test conditions. The efficacy of the antimicrobial function should remain for the whole lifetime of the garment. The durability of the antimicrobial effect depends on the antimicrobial agent, the application method, and the concentration. The antimicrobial efficacy must be tested after typical use conditions and washing cycles. Antimicrobial functionalization must be durable during all processes of the reprocessing of the workwear. Numerous washing and dry cleaning cycles, as well as curing and the hot pressing should not have an effect on the antimicrobial function. Antimicrobial fabrics should be safe for use on skin and not lead to irritation or other health problems. Also, the antimicrobial finishing does not restrict the usual textile processes such as coloring and applying other finishes. The antimicrobial treatment should also not affect the comfort of the fabric.

Assessment of the comfort of workwear for the food industry

25.4

527

Comfort of clothing

The wearing comfort of clothing is nowadays more and more important for workwear and personal protective clothing. Comfort not only affects the wellbeing of the wearer, but also the wearer’s performance and efficiency. For wearers who perform high levels of physical activity, it is important that the workwear provides protection and good heat and moisture management. Comfort is a complex, highly subjective quality and is often defined as the absence of discomfort. Workwear should support the wearer during his activities and must fulfill the needs created by its intended use. The physiological demands can vary depending on the type of clothing, the type of work, and the working environment. Frequently, comfort is thought only to be specific to each person, but this is incorrect: a high work load or ambient temperature will raise the body’s core temperature. To counteract this, humans sweat and if this sweat cannot evaporate, it accumulates on the skin. When a certain amount of the skin’s area is covered by sweat, humans start to feel uncomfortable. With increasing area covered by sweat, discomfort increases and can become unbearable (Bartels & Umbach, 2001b). Clothing physiology has to consider the wearer’s thermoregulation and energy balance. During all activities, humans produce heat within the body and the amount of heat varies greatly with the level of activity. The four important aspects of comfort in clothing are thermophysiological comfort, skin sensorial comfort, ergonomic comfort, and psychological comfort (Mecheels, 1998). Thermophysiological comfort deals with the interaction between the body and the clothing and involves the transport of heat and moisture from the body through the clothing into the environment. Thermophysiological sensations include coolness, warmth, chilling, and sweating. The body reacts to temperature changes with changes in energy. Properly designed clothing must assist human thermoregulation in such a way that the body core temperature remains at a steady value between 36.5°C and 37.5°C. The human body generates heat energy at a steady state, i.e., “metabolic rate”. It varies from 80 W while sleeping to 800 W or more in very high levels of physical activity. To maintain the body core temperature at about 37°C within a limit of only +/ 2°C at varying metabolic rates, the human body has its own thermoregulatory mechanism as shown in Fig. 25.1. The heat transport from the body is affected by a “dry” heat flux, a “latent” heat flux created by sweat evaporation, as well as respiration. In the body, 10% of the heat produced is lost by respiration. About 90% of the heat has to leave the body via the skin and clothing. The heat loss from the skin is affected by a dry heat flux caused by conduction, convection, and radiation. The heat loss depends on the difference between skin temperature and ambient temperature and on the thermal insulation or heat resistance of clothing. If less heat is removed from the body caused by a high thermal insulation from clothing, excess heat is stored within the body. In this case, the human’s thermoregulation counteracts this by sweating. The actual aim of sweating is to remove excessive heat and to cool the body by effecting a “latent” or evaporative heat flux from the skin via evaporation of liquid sweat on the skin’s surface. Vasoconstriction and vasodilatation during sweating is the most effective body mechanism in

528

Context

Fig. 25.1 Thermoregulatory mechanism of the human body (Hohenstein).

keeping human’s energy equilibrium balanced. How well the sweat can evaporate is dependent on the humidity of the ambient air and the moisture transport properties or the moisture resistance of clothing. Additionally, controlling the humidity in the microclimate i.e., in the air layer within the clothing next to the skin is extremely important to maintain values low enough to be subjectively perceived as comfortable. In addition to thermophysiological comfort, the skin sensorial sensation, also called skin sensory comfort, describes the comfort for textile that is worn in contact with the skin and the contact and interaction between textile and the skin. Skin sensory comfort is determined by the mechanical sensation a textile causes through the direct contact with the skin. It includes pleasant perceptions such as smoothness or softness as well as unpleasant perceptions such as scratchiness, stiffness, or clinginess. Textiles with poor skin sensorial wear comfort may even lead to mechanically induced skin irritations. Ergonomic comfort deals with the fit of clothing and the freedom of movement it allows. Essential properties for ergonomic comfort are the garment’s pattern and the elasticity of its materials. Psychological comfort is affected by fashion, personal preferences, ideology, etc. The psychological aspect should not be undervalued: who would feel comfortable in clothing of a color he or she dislikes? Wear comfort is a complex phenomenon that cannot be properly judged by the customer through simply trying the garment on nor can it be defined by sales representatives. However, wear comfort can be measured because it is not entirely an undefined, purely subjective individual sensation. Wear comfort is a quantifiable consequence of the body-climate-clothing interaction.

Assessment of the comfort of workwear for the food industry

25.5

529

Testing of wear comfort

In the beginning of the science of clothing physiology, the only reliable method to evaluate the comfort characteristics of clothing consisted of wear trials with human subjects. Today, the physiological function of textiles and whole garment systems can be measured by a set of laboratory test methods. In the early days of comfort testing, fabrics were field-tested during wearer trials with human subjects because no objective test method was available. These human subject trials evaluated the physiological impact of the clothing on human bodies. However, human subject trials can show a high variability, are time-consuming, and more expensive than the objective test methods. Laboratory test methods are fast and can easily determine different aspects of comfort with high reproducibility. Results obtained with testing devices have less variability than those measured on human beings. However, the results of laboratory test methods must be correlated with human perception in wearer trials because the laboratory test methods cannot directly measure comfort. This physiological validation is crucial because only then will test results of wear comfort be meaningful. As a result, mathematical correlations between the wearer’s perception of comfort and results provided by laboratory devices have to be established (Bartels, 2011). However, recent clothing physiological research has led to a system with which most of the time consuming and expensive in vivo tests can be replaced by short-time laboratory measurements with biophysical apparatus. This system consists of the 5 levels of clothing physiological tests (see Fig. 25.2). The biological analysis of level 1 is performed with the skin model and a materialspecific test method. The biophysical analysis of level 2 is performed with a thermal

Fig. 25.2 The 5 levels of clothing physiological tests (Umbach, 1983a).

530

Context

manikin and yields specific quantities of textiles and clothing, which are directly correlated to their actual wear performance. These tests are product tests and ready-made clothing systems are necessary. The results of the test with the skin model (level 1) and the manikin (level 2) are used in predictive calculations to calculate the wear comfort of clothing that is highly consistent with practical experiences (Umbach, 1983b). The predictive calculations are based on the results of a large number of fundamental wearer trials with human subjects. Their accuracy is so high that the wearer trials of Level 3–5 of the system of analysis are in most cases expendable for the manufacturer or the user of garments.

25.5.1 Thermophysiological comfort testing on the material level On the level of the textile material, the thermophysiological comfort can be determined with the sweating guarded hot plate (the skin model), a thermoregulatory model of the human skin that has been internationally standardized (ISO 11092, 2014). The skin model simulates the way the skin emits heat and moisture. It consists of an electrically heated porous metal plate with a water supply. The metal plate is placed in a climatic cabinet with adjustable temperature, air humidity, and air movement (Fig. 25.3). Water vapor and water can be released in a controlled manner; thus, simulating perspiration of human skin and different wear situations with different levels of sweat production. The measurements with the skin model are highly reproducible. Normal wear situations with only moderate sweating can be simulated, as well as situations with increased or heavy sweating. With the skin model, practically all different wear situations occurring under practical use of textiles can be simulated, ranging from normal/negligible sweating up to very heavy sweating with a high amount of

Skin model Climatized air stream

Measuring head Textile sample

Textile sample Heating

Guard

Guard

Water supply

© Hohenstein Institute

Fig. 25.3 Hohenstein Skin Model, a sweating guarded hot plate (Hohenstein©).

Assessment of the comfort of workwear for the food industry

531

sweat on the skin. A series of parameters characterizing the thermophysiological quality of a textile material can be determined. The normal wear situation with low levels of perspiration can be described with the thermal insulation (Rct, thermal resistance [m2K/W]), the “breathability” Ret, (water vapor resistance [m2Pa/W]) and the short time water vapor absorbency Fi [g/m2]. With increasing physical activities, moderate perspiration starts, and this is described by the sweat buffering Fd. Under heavy physical activity, heavy perspiration starts, and the important parameters are the sweat transport F [g/m2h], the water retention ΔG [%], the sweat buffering Kf, and the drying time Δt [min]. All these parameters can be determined with the Hohenstein Skin Model (see Table 25.4). The material testing can be made with single layer or multilayered textiles. The values of the multilayer combination are influenced by the air layers and the connection of the layers. It is not possible to measure only the single layer of a multilayer garment and make a calculation of the breathability.

25.5.2 Thermophysiological comfort of the product level To assess the thermophysiological properties of ready-made garments, a life-size thermal manikin is used to evaluate the performance of clothing systems. Workwear and protective clothing are often made with specialized materials and unique designs. The garments are often worn with auxiliary equipment such as a tool belt or respirator. It is important to evaluate the comfort and performance of the entire protective clothing system—not just the properties of component materials (McCullough, 2005). Thermal manikins have been used since 1949 to measure the thermal resistance and

Table 25.4 Important thermophysiological parameters measured with the skin model Normal wear situation 5 insensible perspiration Test criteria

Abbreviation

Unit

Thermal insulation Water vapor resistance Water vapor absorbency index Short time water vapor absorbency

Rct Ret imt Fi g/m2

m2K/W m2Pa/W – g/m2

Increased physical activity 5 moderate sensible perspiration Buffering of water vapor

Fd

–

Heavy physical activity 5 strong sensible perspiration Sweat transport Water retention Sweat buffering Drying time

F ΔG Kf Δt

g/m2h % – min

532

Context

evaporative resistance of clothing systems (Fan, 2006; Holm
er, 2006). The firstgeneration manikins are standing manikins. Different thermal manikin models differ in the number of segments corresponding to different parts of the body. The segments are separately controlled by a computer and allow the simulation of characteristic temperature gradient at the body surface. In the past, thermal manikins were often built for research activities in universities or research organizations and showed varying sizes and shapes that caused differences in the results. The second-generation manikins are movable/able to walk. The walking action is useful for the realistic judgment of comfort because the design and ventilation effects associated with movement have a considerable influence on the physiological function. The third-generation manikins are movable/able to walk and/or perspire and these are called sweating manikins with different sweating techniques (Mandal, Annaheim, Camenzind, & Rossi, 2017). Different standards can be used for manikin measurements (e.g., ISO 15831, 2004; ISO 9920, ASTM F 1291, ASTM F 1720, ASTM F 2370) to evaluate the thermal and evaporative resistance of clothing. The studies of the last year show that the ambient environmental conditions, the fabric features, and/or the manikins body postures affected the heat and moisture transfer through the clothing (Mandal et al., 2017). A trend in the past has been the development of specialized heated body parts such as a head, hand, and foot (Kuklane, Nilsson, & Holmer, 1997). Manikins do not simulate the human body physiologically. They are thermal measuring devices in the size and shape of a human being that are heated so that their surface temperatures simulate the local and/or mean skin temperature of a human being. They do not respond to changes in the environment or clothing like the human body does (McCullough, 2005). At Hohenstein, the thermal manikin “Charlie” was developed. The thermal movable manikin Charlie 4 consists of 16 segments, which can be heated and controlled separately in the same way that heat is generated by humans (Fig. 25.4). The manikin is placed in a climatic chamber to control the environmental conditions (air temperature, relative humidity, air velocity). The environmental condition can be adapted to the realistic user scenario for the investigated clothing. The assessments made using the thermally segmented manikins are an important complement to those made using the skin model, because the influence of the way the item or garment is made (e.g., fit, elasticized cuffs, turtlenecks etc.) can be taken into consideration. The quantities measured with the manikin can be applied as input into predictive calculations that yield the clothing ensemble’s temperature range of utility where, with a given physical activity, the wearer is not yet suffering from hypo- or hyperthermia, respectively. On the other hand, with a specific scenario of climate and activity conditions, the wearer’s degree of physiological comfort can be predicted. The advantage of manikin testing is that it provides a realistic simulation of heat transfer from the body through the clothing into the environment. The wear properties of garments are not only determined by the different fabrics included and their interaction but also by the air volumes trapped inside the clothing due to the garments’ design or adhering to their outer surface. Manikins can simulate different body

Assessment of the comfort of workwear for the food industry

533

Fig. 25.4 The thermal manikin “Charlie 4”.

postures relevant to the clothing’s actual use, that is standing, walking, sitting, and lying down. This allows evaluation of the influence of convection and ventilation in the garment’s microclimate caused by wearer’s body movement on their physiological function. The results of manikin tests and the sweating guarded hot plates are objective physical data. But without correlation to wearer trials, the comparison with the human perception is not possible. To be able to translate the data obtained using the thermal manikin into the comfort perceived by humans, a huge number of wear tests with human subjects were carried out with different garments.

25.5.3 Wearer trials Human trials are the most commonly used method to evaluate the comfort characteristics of clothing. In the early days of comfort testing, fabrics were field-tested during wearer trials with human subjects because no objective test method was available. Wearer trials with human test subjects are performed in field tests or in a climate chamber under controlled conditions. In field tests, the environmental condition i.e., air temperature, air humidity, wind speed can vary between individual tests and so a huge number of participants is necessary to obtain a statistically significant result. Wearer trials as field tests can show a high variability, are time-consuming, and more expensive than the objective test methods. Wearer trials in a climatic chamber have the advantages that the environmental condition can be easily controlled and also the test subjects’ activity. For the evaluation of the clothing, it’s important to simulate a realistic user scenario in the climatic chamber. The selection of the test setup is very

534

Context

important to be able to judge and distinguish different clothing systems because the environmental test condition shows a high influence on the results. Several subjective and objective parameters are assessed during the investigation. Objective data can be recorded via various sensors attached to the subject’s body to measure the body functions, i.e., the heart rate, skin temperature, humidity of the microclimate, and body core temperature. Mean skin temperature is usually assessed as the weighted average measurement at specific points on the skin. Subjective data are selected by usage of scales to determine how the subjects feel before, during, and after the trial. During the trials, the test subjects rate their microclimate comfort regarding temperature and humidity on a rating box that is installed near the test subjects. The subjective temperature is rated by a modified seven-point Bedford scale (Bedford, 1936): 1 ¼ too cold, 2 ¼ cold, 3 ¼ slightly cold, 4 ¼ neutral, 5 ¼ warm, 6 ¼ hot, 7 ¼ too hot. Humidity is not perceived as detailed as the temperature perception, so a four-level scale is used for the subjective humidity rating: 1 ¼ dry, 2 ¼ slightly moist, 3 ¼ moist, and 4 ¼ wet. Additionally, a questionnaire is used during and after the trials to collect information on the feelings of the subjects. The questionnaire must be prepared carefully to avoid unclear or suggestive questions. Body mass loss is also one important parameter for the assessment of the thermal strain of the clothing. The body mass loss is mainly due to sweat loss. The human subject and all garments worn during the wearer trials are weighed before and after the test. The results of wearer trials in climatic chambers have a high reproducibility and repeatability in comparison to the wearer trials in field tests. However, wearer trials are time consuming and expensive, while laboratory tests can provide results at the material and clothing level for product development (Fig. 25.5).

25.5.4 Skin sensorial comfort In addition to thermophysiological wear comfort, the skin sensory wear comfort is a key determinant of comprehensive wear comfort for textiles that are worn in direct contact with the skin (Schmidt, Paul, Classen, Morlock, & Beringer, 2016). In his first contact with the fabric, the consumer examines the fabric properties by touching it with his hand to select a good clothing material according to his feeling and experience (Kawabata, 1980). Touching with the hand or fingers does not provide the same perception as when a fabric is worn in direct contact with the human body. Clothes worn next to the skin have to offer particularly good sensorial wear comfort to be accepted by the wearer (Bartels & Umbach, 2001a). For this propose, the Hohenstein Institutes in Germany have developed and improved a complex system of skin sensory test devices and assessment formula over several decades (Bartels & Umbach, 2002; Mecheels, 1982). The research results show that the following parameters are important: the stiffness s, the sorption index iB, the surface index iO, the number of contact points nK, and the wet cling index iK. The skin sensitivity to mechanical irritation becomes stronger as

Assessment of the comfort of workwear for the food industry

535

Fig. 25.5 Example of a subject trial of workwear in the climatic chamber under wind conditions on a treadmill.

moisture increases. Therefore, the textile must promote sweat transport from the skin through the textile layer into the environment. The stiffness s of a textile is an indicator of how well a textile will adapt to the body shape (Bartels & Umbach, 2002; Mecheels, 1982). A textile can feel comfortably smooth or soft, or it may be unpleasantly stiff, or too limp. It is measured in a special device using a laser beam to calculate the bending angle of a sample strip draped over a thin stick. The results of these test methods were correlated to the sensation of human subjects in different research projects. Therefore, defined criteria for different products and areas of use were developed, which ensure maximum wear comfort and prevent any mechanical skin irritation caused by excessive stiffness. By definition, s can assume values between 0 (completely floppy) and 90 (completely rigid). For sensorial comfort, it is important to keep the body’s surface as dry as possible, even under high activity levels (Bartels & Umbach, 2002; Mecheels, 1982). Skin wetness leads to a faster abrasion of the upper part of the body, a dilution of skin fats and decrease of their protective function, a change in the skin pH-value, and a higher penetration of chemicals or allergenic substances through the skin. Wet skin is much more sensitive and can be irritated more easily than dry skin. Therefore, rapid sweat transport through the textile leads to better thermophysiological comfort and higher skin

536

Context

sensorial comfort. To measure how quickly a water droplet is absorbed by the textile, the sorption index iB uses time and contact angles. As the skin becomes moister, it becomes more sensitive to mechanical irritation. It is important for a textile material to transport sweat away from the skin as quickly as possible. The sorption index indicates the speed at which a drop water on the textile is absorbed by it. To measure this, a drop of water with a defined volume is applied to the textile sample and observed by a video camera. The contact angle between the drop of water and the surface of the textile is measured continuously to establish how quickly the textile materials absorbs the water. The surface index iO expresses the hairiness or roughness/smoothness of a textile. Here, a camera records a microscopic image of a cross section of the textile. The number and size of the protruding fibers in the fabric can then be calculated. The surface index can be used to judge whether a textile will scratch or feel too smooth. The number of contact points between the textile and skin nK states how fast a textile will be sensed as clammy or damp. This number is determined optically with a topograph, which gives a 3-dimensional picture of the textile surface. Image analysis systems connected to a surface scanner show the number of contact points and the surface structure of textiles. These can be used as a scale for working out what area of the textile material is in contact with the skin. The wet cling index iK indicates how likely a textile is to adhere to perspiration moistened skin. The textile sample is put on a sintered glass plate that simulates the sweating human skin. The force required to move the textile over the glass plate is determined (Fig. 25.6).

Fig. 25.6 Five skin sensorial test devices (Hohenstein©).

Assessment of the comfort of workwear for the food industry

537

Ergonomic comfort of clothing is especially important for protective clothing and workwear. Ergonomics is a scientific discipline concerned with understanding the interaction between humans and other elements of a system. Ergonomics applies theory, principles, data, and methods to design in a way that optimizes human wellbeing and overall system performance (IEA, 2017). Proper ergonomic design is necessary to prevent repetitive strain injuries and other musculoskeletal disorders that may develop over time and can lead to long-term disabilities. Ergonomics is concerned with the “Fit” between user, his/her equipment, and their environment. If a garment fits the human body, the wearer can easily move in the garment. Ergonomic comfort, sometimes also called aesthetic comfort, can be tested with different objective methods (Sinclair, 2014). One traditional method measures the space between the body and the clothing and calculates the fitting index. The ease of motion can be assessed through wearer trials. In these wearer trials, wearers are required to perform a series of activities, which normally occur in practice. The wearer rates the ease of body movement on a scale, e.g., the Likert scale. To judge the ergonomic comfort of a whole garment, different assessments are made for different parts of the body and different activities. Photographs can be taken for additional subsequent visual assessment. Another method is the symmetric dot pattern technology (Sinclair, 2014). This new method is based on changes in dot patterns and imaging technologies that allow the capture and analysis of garment images. However, the garment surface may be folded or wrinkled. In this case, the space between the body and garment cannot easily be measured with high precision and efficiency. Additionally, 3D-scanning technology makes it possible to identify spaces between the body and the garment. Three-dimensional body data for both nude and dressed human subjects are necessary to build up the spaces between the body and the garment. With 3D-scanning technology, the body is described in a static way and motion of the body is not involved. With innovative 4D-scanning technology, there is now the possibility to give a more realistic picture of the user scenario by including movements of the body.

25.5.5 Psychological comfort testing Psychological comfort represents how the individual feels in the clothing. It is influenced by different factors such as styling, color, fashion, and aesthetics. Psychological comfort is highly subjective because of the personal nature of positive association and because the personal experiences of each subject are different. Therefore, expectations differ, resulting in a significant effect on their response. Subjective perceptions involve psychological processes in which all relevant sensory perceptions are formulated, weighted, combined, and evaluated against past experiences and present desires. This type of comfort is qualitative and cannot be measured (Classen, 2018).

538

Context

25.6

Quality of workwear

To make an overall statement regarding the wear comfort of a textile, Umbach combined the individual parameters and extended the measuring methods used in clothing physiology as well as biophysical evaluation models since 1976 (Hohenstein, 2012; Umbach, 1983a, 1983b). In 1988, clothing physiological specification values were incorporated in different standards for personal protective clothing, e.g., for “weatherproof clothing” specific values were first introduced in the German standard DIN 61539 in 1988, which was replaced by the European standard EN 343 in 1998. As a result, since this time, specification values for wear comfort have been included in numerous other standards for personal protective clothing, e.g., in EN 469 (Protective clothing for fire fighters), in EN 20471 (High-visibility warning clothing), and in DIN 10524 (Workwear in the food industry). Protective clothing must satisfy these values in order to affect good wear comfort, depending on the conditions under which they are used. Therefore, a lot of wearer trials were carried out with a wide variety of materials and a range of climatic conditions and physical activity levels. In the end, an overall wear comfort vote was developed and introduced for fabrics to give producers, retailers, and consumers a simple system to judge and compare the thermophysiological and skin sensorial wear comfort of fabrics. This wear comfort can be calculated from the results of the measurements on the skin model and the five skin sensorial test methods.

25.6.1 Sensorial comfort vote With the different skin sensorial wear comfort parameters, a comprehensive skin sensorial skin wear comfort vote WCs can be predicted. WCs is given as a 6-point scale ranging from 1 (“very good”) to 6 (“unsatisfactory”). With the sensorial comfort vote, an understandable measurement is given to compare the skin sensorial wear comfort in different textile constructions. The determination of WCs slightly differs for knitwear and for woven fabrics, because wearers expect knitwear to be softer or smoother than a woven fabric. The skin sensorial comfort of workwear WCS for woven fabrics is calculated by the following formula: WCS ¼ α1 imt + α2 ik + α3 iB + α4 ½9  ιΟ  + α5 nΚ + α6 s + β where imt¼ water vapor permeability index ik ¼ wet cling index iB ¼ sorption index iO ¼ surface index nK¼ number of contact points

(25.1)

Assessment of the comfort of workwear for the food industry

539

s ¼ stiffness α1, α2, α3, α4, α5, α6, β constants.

With the above formula, the skin sensorial wear comfort can be predicted.

25.6.2 Comfort vote for workwear in the food industry The skin sensorial wear comfort vote WCs can be combined with the thermophysiological properties of a textile to obtain an overall comfort vote WC. Different mathematical formulae were developed that provide the wear comfort rating. These formulae differentiate the type of clothing and the wearer’s level of activity and are based on the textile parameters measured using the skin model and skin sensory devices. The wear comfort vote can predict the perceived wear comfort in practice. Differences of 0.5 or more can be regarded as perceivable. The thermophysiological wear comfort vote WCT for woven fabrics in occupational clothing (workwear) or trousers is calculated by the following formula: WCT ¼ α Ret + β Fd + χ Kf + δ F1 + ε ΔG + ϕ

(25.2)

where Ret ¼ water vapor resistance [m”Pa/W] Fd ¼ buffering capacity [g/m2] F1¼ sweat transport [g/m2h] ΔG ¼ water retention [%] α, β, χ, δ, ε, Δ + ϕ are constants

In different research projects at the Hohenstein Institute, it was found out that 66% of the subjective comfort perceived with a woven fabric is caused by the thermophysiological characteristics and 34% by its sensorial properties, with the fabrics total resulting comfort WC determined out of the thermophysiological and sensorial comfort votes using the following formula: WC ¼ 0:66 WCT + 0:34 WCS

(25.3)

where WCT ¼ thermophysiological comfort vote WCS ¼ skin sensorial comfort vote

This wear comfort rating ranges from 1 (¼ excellent) to 6 (¼ inadequate) following the German school grading system and provides a quantitative assessment of the physiological quality of a textile product, as well as making it possible for those who know little about textiles to make direct product comparisons based on wear comfort when making a purchase. Wearer trials with different garments have shown that the comfort vote WC thus predicted for a fabric is in very good agreement with the comfort sensation actually perceived by customers.

540

Context

Table 25.5 Comfort parameters for workwear for the food industry Thermophysiological parameters Unit Thermal resistance Rct Water vapor resistance Ret Water vapor absorbency index imt Buffering of water vapor Fd Sweat buffering of the liquid phase Kf Sweat transport F

0.01–0.02 m2K/W 0.15 >0.4 >0.78 >765 g/m2 h

Sensory parameters Wet cling index iK Sorption index iB Surface index iO Number of contact points nK Stiffness S