Human Aspects of IT for the Aged Population (Lecture Notes in Computer Science) 3031348656, 9783031348655

Citation preview

LNCS 14042

Qin Gao Jia Zhou (Eds.)

Human Aspects of IT for the Aged Population 9th International Conference, ITAP 2023 Held as Part of the 25th HCI International Conference, HCII 2023 Copenhagen, Denmark, July 23–28, 2023, Proceedings, Part I

Lecture Notes in Computer Science Founding Editors Gerhard Goos Juris Hartmanis

Editorial Board Members Elisa Bertino, Purdue University, West Lafayette, IN, USA Wen Gao, Peking University, Beijing, China Bernhard Steffen , TU Dortmund University, Dortmund, Germany Moti Yung , Columbia University, New York, NY, USA

14042

The series Lecture Notes in Computer Science (LNCS), including its subseries Lecture Notes in Artificial Intelligence (LNAI) and Lecture Notes in Bioinformatics (LNBI), has established itself as a medium for the publication of new developments in computer science and information technology research, teaching, and education. LNCS enjoys close cooperation with the computer science R & D community, the series counts many renowned academics among its volume editors and paper authors, and collaborates with prestigious societies. Its mission is to serve this international community by providing an invaluable service, mainly focused on the publication of conference and workshop proceedings and postproceedings. LNCS commenced publication in 1973.

Qin Gao · Jia Zhou Editors

Human Aspects of IT for the Aged Population 9th International Conference, ITAP 2023 Held as Part of the 25th HCI International Conference, HCII 2023 Copenhagen, Denmark, July 23–28, 2023 Proceedings, Part I

Editors Qin Gao Tsinghua University Beijing, China

Jia Zhou Chongqing University Chongqing, China

ISSN 0302-9743 ISSN 1611-3349 (electronic) Lecture Notes in Computer Science ISBN 978-3-031-34865-5 ISBN 978-3-031-34866-2 (eBook) https://doi.org/10.1007/978-3-031-34866-2 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023, corrected publication 2023 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Foreword

Human-computer interaction (HCI) is acquiring an ever-increasing scientific and industrial importance, as well as having more impact on people’s everyday lives, as an ever-growing number of human activities are progressively moving from the physical to the digital world. This process, which has been ongoing for some time now, was further accelerated during the acute period of the COVID-19 pandemic. The HCI International (HCII) conference series, held annually, aims to respond to the compelling need to advance the exchange of knowledge and research and development efforts on the human aspects of design and use of computing systems. The 25th International Conference on Human-Computer Interaction, HCI International 2023 (HCII 2023), was held in the emerging post-pandemic era as a ‘hybrid’ event at the AC Bella Sky Hotel and Bella Center, Copenhagen, Denmark, during July 23–28, 2023. It incorporated the 21 thematic areas and affiliated conferences listed below. A total of 7472 individuals from academia, research institutes, industry, and government agencies from 85 countries submitted contributions, and 1578 papers and 396 posters were included in the volumes of the proceedings that were published just before the start of the conference, these are listed below. The contributions thoroughly cover the entire field of human-computer interaction, addressing major advances in knowledge and effective use of computers in a variety of application areas. These papers provide academics, researchers, engineers, scientists, practitioners and students with state-of-the-art information on the most recent advances in HCI. The HCI International (HCII) conference also offers the option of presenting ‘Late Breaking Work’, and this applies both for papers and posters, with corresponding volumes of proceedings that will be published after the conference. Full papers will be included in the ‘HCII 2023 - Late Breaking Work - Papers’ volumes of the proceedings to be published in the Springer LNCS series, while ‘Poster Extended Abstracts’ will be included as short research papers in the ‘HCII 2023 - Late Breaking Work - Posters’ volumes to be published in the Springer CCIS series. I would like to thank the Program Board Chairs and the members of the Program Boards of all thematic areas and affiliated conferences for their contribution towards the high scientific quality and overall success of the HCI International 2023 conference. Their manifold support in terms of paper reviewing (single-blind review process, with a minimum of two reviews per submission), session organization and their willingness to act as goodwill ambassadors for the conference is most highly appreciated. This conference would not have been possible without the continuous and unwavering support and advice of Gavriel Salvendy, founder, General Chair Emeritus, and Scientific Advisor. For his outstanding efforts, I would like to express my sincere appreciation to Abbas Moallem, Communications Chair and Editor of HCI International News. July 2023

Constantine Stephanidis

HCI International 2023 Thematic Areas and Affiliated Conferences

Thematic Areas • HCI: Human-Computer Interaction • HIMI: Human Interface and the Management of Information Affiliated Conferences • EPCE: 20th International Conference on Engineering Psychology and Cognitive Ergonomics • AC: 17th International Conference on Augmented Cognition • UAHCI: 17th International Conference on Universal Access in Human-Computer Interaction • CCD: 15th International Conference on Cross-Cultural Design • SCSM: 15th International Conference on Social Computing and Social Media • VAMR: 15th International Conference on Virtual, Augmented and Mixed Reality • DHM: 14th International Conference on Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management • DUXU: 12th International Conference on Design, User Experience and Usability • C&C: 11th International Conference on Culture and Computing • DAPI: 11th International Conference on Distributed, Ambient and Pervasive Interactions • HCIBGO: 10th International Conference on HCI in Business, Government and Organizations • LCT: 10th International Conference on Learning and Collaboration Technologies • ITAP: 9th International Conference on Human Aspects of IT for the Aged Population • AIS: 5th International Conference on Adaptive Instructional Systems • HCI-CPT: 5th International Conference on HCI for Cybersecurity, Privacy and Trust • HCI-Games: 5th International Conference on HCI in Games • MobiTAS: 5th International Conference on HCI in Mobility, Transport and Automotive Systems • AI-HCI: 4th International Conference on Artificial Intelligence in HCI • MOBILE: 4th International Conference on Design, Operation and Evaluation of Mobile Communications

List of Conference Proceedings Volumes Appearing Before the Conference

1. LNCS 14011, Human-Computer Interaction: Part I, edited by Masaaki Kurosu and Ayako Hashizume 2. LNCS 14012, Human-Computer Interaction: Part II, edited by Masaaki Kurosu and Ayako Hashizume 3. LNCS 14013, Human-Computer Interaction: Part III, edited by Masaaki Kurosu and Ayako Hashizume 4. LNCS 14014, Human-Computer Interaction: Part IV, edited by Masaaki Kurosu and Ayako Hashizume 5. LNCS 14015, Human Interface and the Management of Information: Part I, edited by Hirohiko Mori and Yumi Asahi 6. LNCS 14016, Human Interface and the Management of Information: Part II, edited by Hirohiko Mori and Yumi Asahi 7. LNAI 14017, Engineering Psychology and Cognitive Ergonomics: Part I, edited by Don Harris and Wen-Chin Li 8. LNAI 14018, Engineering Psychology and Cognitive Ergonomics: Part II, edited by Don Harris and Wen-Chin Li 9. LNAI 14019, Augmented Cognition, edited by Dylan D. Schmorrow and Cali M. Fidopiastis 10. LNCS 14020, Universal Access in Human-Computer Interaction: Part I, edited by Margherita Antona and Constantine Stephanidis 11. LNCS 14021, Universal Access in Human-Computer Interaction: Part II, edited by Margherita Antona and Constantine Stephanidis 12. LNCS 14022, Cross-Cultural Design: Part I, edited by Pei-Luen Patrick Rau 13. LNCS 14023, Cross-Cultural Design: Part II, edited by Pei-Luen Patrick Rau 14. LNCS 14024, Cross-Cultural Design: Part III, edited by Pei-Luen Patrick Rau 15. LNCS 14025, Social Computing and Social Media: Part I, edited by Adela Coman and Simona Vasilache 16. LNCS 14026, Social Computing and Social Media: Part II, edited by Adela Coman and Simona Vasilache 17. LNCS 14027, Virtual, Augmented and Mixed Reality, edited by Jessie Y. C. Chen and Gino Fragomeni 18. LNCS 14028, Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management: Part I, edited by Vincent G. Duffy 19. LNCS 14029, Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management: Part II, edited by Vincent G. Duffy 20. LNCS 14030, Design, User Experience, and Usability: Part I, edited by Aaron Marcus, Elizabeth Rosenzweig and Marcelo Soares 21. LNCS 14031, Design, User Experience, and Usability: Part II, edited by Aaron Marcus, Elizabeth Rosenzweig and Marcelo Soares

x

List of Conference Proceedings Volumes Appearing Before the Conference

22. LNCS 14032, Design, User Experience, and Usability: Part III, edited by Aaron Marcus, Elizabeth Rosenzweig and Marcelo Soares 23. LNCS 14033, Design, User Experience, and Usability: Part IV, edited by Aaron Marcus, Elizabeth Rosenzweig and Marcelo Soares 24. LNCS 14034, Design, User Experience, and Usability: Part V, edited by Aaron Marcus, Elizabeth Rosenzweig and Marcelo Soares 25. LNCS 14035, Culture and Computing, edited by Matthias Rauterberg 26. LNCS 14036, Distributed, Ambient and Pervasive Interactions: Part I, edited by Norbert Streitz and Shin’ichi Konomi 27. LNCS 14037, Distributed, Ambient and Pervasive Interactions: Part II, edited by Norbert Streitz and Shin’ichi Konomi 28. LNCS 14038, HCI in Business, Government and Organizations: Part I, edited by Fiona Fui-Hoon Nah and Keng Siau 29. LNCS 14039, HCI in Business, Government and Organizations: Part II, edited by Fiona Fui-Hoon Nah and Keng Siau 30. LNCS 14040, Learning and Collaboration Technologies: Part I, edited by Panayiotis Zaphiris and Andri Ioannou 31. LNCS 14041, Learning and Collaboration Technologies: Part II, edited by Panayiotis Zaphiris and Andri Ioannou 32. LNCS 14042, Human Aspects of IT for the Aged Population: Part I, edited by Qin Gao and Jia Zhou 33. LNCS 14043, Human Aspects of IT for the Aged Population: Part II, edited by Qin Gao and Jia Zhou 34. LNCS 14044, Adaptive Instructional Systems, edited by Robert A. Sottilare and Jessica Schwarz 35. LNCS 14045, HCI for Cybersecurity, Privacy and Trust, edited by Abbas Moallem 36. LNCS 14046, HCI in Games: Part I, edited by Xiaowen Fang 37. LNCS 14047, HCI in Games: Part II, edited by Xiaowen Fang 38. LNCS 14048, HCI in Mobility, Transport and Automotive Systems: Part I, edited by Heidi Krömker 39. LNCS 14049, HCI in Mobility, Transport and Automotive Systems: Part II, edited by Heidi Krömker 40. LNAI 14050, Artificial Intelligence in HCI: Part I, edited by Helmut Degen and Stavroula Ntoa 41. LNAI 14051, Artificial Intelligence in HCI: Part II, edited by Helmut Degen and Stavroula Ntoa 42. LNCS 14052, Design, Operation and Evaluation of Mobile Communications, edited by Gavriel Salvendy and June Wei 43. CCIS 1832, HCI International 2023 Posters - Part I, edited by Constantine Stephanidis, Margherita Antona, Stavroula Ntoa and Gavriel Salvendy 44. CCIS 1833, HCI International 2023 Posters - Part II, edited by Constantine Stephanidis, Margherita Antona, Stavroula Ntoa and Gavriel Salvendy 45. CCIS 1834, HCI International 2023 Posters - Part III, edited by Constantine Stephanidis, Margherita Antona, Stavroula Ntoa and Gavriel Salvendy 46. CCIS 1835, HCI International 2023 Posters - Part IV, edited by Constantine Stephanidis, Margherita Antona, Stavroula Ntoa and Gavriel Salvendy

List of Conference Proceedings Volumes Appearing Before the Conference

xi

47. CCIS 1836, HCI International 2023 Posters - Part V, edited by Constantine Stephanidis, Margherita Antona, Stavroula Ntoa and Gavriel Salvendy

https://2023.hci.international/proceedings

Preface

The 9th International Conference on Human Aspects of IT for the Aged Population (ITAP 2023) was part of HCI International 2023. The ITAP conference addresses the design, adaptation, and use of IT technologies targeted for the use of older people in order to counterbalance ability changes due to age, support cognitive, physical, and social activities, and maintain independent living and quality of life. This year’s proceedings address a variety of topics. Researchers from all over the world shared their findings on how older people accept and use new technologies to stay informed and connected, how they develop ICT competencies and skills, and how such understanding informs the design and development of technologies to support and empower older people. In particular, a research theme that attracts much attention in this year’s proceedings is how to design immersive and playful experience–through the use of XR technologies and game design–to provide supportive functions for older people, such as promoting physical activity and maintaining cognitive functions, in addition to its entertainment value. Furthermore, an emerging theme of this year is the design of IT support for both formal and informal caregivers for older people. These changes highlight the importance of making IT applications and services an enjoyable, engaging, and integral part of older people’s lives. Two volumes of the HCII 2023 proceedings are dedicated to this year’s edition of the ITAP conference. The first part focuses on topics related to designing and assessing the older user’s experience, older people’s use of social media and its impact, design of voice assistants and chatbots, games and exergames for older people, as well as research on XR experiences for older people. The second part focuses on topics related to smart homes and other technologies supporting aging in place, eHealth applications for older people and their health literacy, IT support for caregivers, as well as socioeconomic views on ICT use and digital literacy of older people. Papers of these volumes are included for publication after a minimum of two singleblind reviews from the members of the ITAP Program Board or, in some cases, from members of the Program Boards of other affiliated conferences. We would like to thank all of them for their invaluable contribution, support, and efforts. July 2023

Qin Gao Jia Zhou

9th International Conference on Human Aspects of IT for the Aged Population (ITAP 2023)

Program Board Chairs: Qin Gao, Tsinghua University, P.R. China, and Jia Zhou, Chongqing University, P.R. China Program Board: • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •

Bessam Abdulrazak, University of Sherbrooke, Canada Inês Amaral, University of Coimbra, Portugal Ning An, Hefei University of Technology, P.R. China Maria José Brites, Lusófona University, Portugal Maurizio Caon, University of Applied Sciences and Arts of Western Switzerland (HES-SO), Switzerland Alan H.S. Chan, City University of Hong Kong, P.R. China Judith Charlton, Monash University, Australia Yue Chen, East China University of Science and Technology, P.R. China Honglin Chen, University of Eastern Finland, Finland Fausto Colombo, Università Cattolica del Sacro Cuore, Italy M. Jamal Deen, McMaster University, Canada Tova Gamliel, Bar-Ilan University, Israel Loredana Ivan, National University of Political Studies and Public Administration, Romania Yong Gu Ji, Yonsei University, South Korea Chaiwoo Lee, MIT AgeLab, USA Hai-Ning Liang, Xi’an Jiaotong-Liverpool University, P.R. China Eugene Loos, Utrecht University, The Netherlands Xinggang Luo, Hangzhou Dianzi University, P.R. China Yan Luximon, Hong Kong Polytechnic University, P.R. China Zhaoyi Ma, Nanjing University of Science & Technology, P.R. China Lisa Molnar, University of Michigan, USA Lourdes Moreno Lopez, Universidad Carlos III de Madrid, Spain Sergio F. Ochoa, University of Chile, Chile Jing Qiu, University of Electronic Science and Technology of China, P.R. China Diana Saplacan, University of Oslo, Norway Patrice Terrier, Université de Toulouse, France Wang-Chin Tsai, National Yunlin University of Science and Technology, Taiwan Ana Isabel Veloso, University of Aveiro, Portugal Nadine Vigouroux, University of Toulouse, France Tingru Zhang, Shenzhen University, P.R. China Yuxiang (Chris) Zhao, Nanjing University of Science and Technology, P.R. China Junhong Zhou, Harvard Medical School, USA

xvi

9th International Conference on Human Aspects of IT for the Aged Population

The full list with the Program Board Chairs and the members of the Program Boards of all thematic areas and affiliated conferences of HCII2023 is available online at:

http://www.hci.international/board-members-2023.php

HCI International 2024 Conference

The 26th International Conference on Human-Computer Interaction, HCI International 2024, will be held jointly with the affiliated conferences at the Washington Hilton Hotel, Washington, DC, USA, June 29 – July 4, 2024. It will cover a broad spectrum of themes related to Human-Computer Interaction, including theoretical issues, methods, tools, processes, and case studies in HCI design, as well as novel interaction techniques, interfaces, and applications. The proceedings will be published by Springer. More information will be made available on the conference website: http://2024.hci.international/. General Chair Prof. Constantine Stephanidis University of Crete and ICS-FORTH Heraklion, Crete, Greece Email: [email protected]

https://2024.hci.international/

Contents – Part I

Designing and Assessing the Older Users’ Experience Designing for Self-directed Learning: Co-creating a Demokit with Older Adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Katerina Cerna, Claudia Müller, Martin Hunker, Florian Hitpass, and David Randall Text Readability of Smartphone in Dark Mode: Effects of Font Type, Font Weight and Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Junxiang Chen, Pengyu Zhu, Jing Yun, Baixuan Tian, Yalan Yang, and Na Liu Human Factors Based New Media Design: Methodology and Assessment . . . . . Shijun Ge, Yuanbo Sun, Yin Cui, and Hao Shen Increasing Quality of Life by Playing Chess: A Blended Care Approach for Elderly People . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Matthias Haase and Jens-Martin Loebel Lessons from a COVID-Era Controlled Trial of Online Learning and Socializing with Older Participants: Benefits of Applying Consolidated Standards of Reporting Trials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Najmeh Khalili-Mahani, Berkley Petersen, and Shannon Hebblethwaite

3

23

34

45

56

Older Adults’ Perceptions and Use of Digital Engagement Channels: In the Case of Automobile Insurance Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Chaiwoo Lee, Lisa D’Ambrosio, and Joseph F. Coughlin

75

Promoting Positive Emotions in Older Adults: A Self-help Relational Savoring e-Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Elisa Pancini, Francesca Pesce, Lucia Scuzzarella, and Daniela Villani

88

The Relationship Between Older Drivers’ Cognitive Ability and Takeover Performance in Conditionally Automated Driving . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Qijia Peng, Yanbin Wu, Toshihisa Sato, and Sunao Iwaki The After Effect of COVID-19 on Colour Perception in the Elderly . . . . . . . . . . . 116 Chuanke Qin and Ao Jiang

xx

Contents – Part I

QR Codes as a Method for Older Adults to Access a Mobile Survey . . . . . . . . . . 128 Alda G. Rivas and Anthony Schulzetenberg The Effects of an Art Program on Older Adults’ Cognition and Satisfaction with Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 Fang-Wu Tung and Liang-Ming Jia College Students’ Perceptions and Preferences Regarding Intelligent Advisory Systems in Multi-device Learning Environments . . . . . . . . . . . . . . . . . . 154 Zhao Xu and Qingchuan Li Experimental Design and Design Methodology of Smart Screen Ageing Based on Emotion Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 Yan-Min Xue, Meng-Ting Xu, Bao-Xin Xu, Chang Ge, and Sui-Huai Yu Aging and Social Media Strategies for Fostering Senior Citizens’ Active Participation in Online Communities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183 Rafaela Carrilho, Liliana Vale Costa, and Ana Isabel Veloso Facebook Community Lurking and Non-participation: Sociotechnical Barriers Among Older Adults in Online Participation . . . . . . . . . . . . . . . . . . . . . . . 194 Ryan Ebardo and Merlin Teodosia Suarez Optimizing WeChat User Experience for Older Adults in China . . . . . . . . . . . . . . 205 Nan Hu I Know How but I Do not Want to Discern Falsehoods: Older Adults’ Self-Reported Inference Process to Identify and Share Short-Form Videos . . . . . 220 Wei Hu, Honglian Xiang, Jia Zhou, Wenjun Sun, and Jinjun Xia Using Media Literacy to Fight Digital Fake News in Later Life: A Mission Impossible? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Eugène Loos and Loredana Ivan Older Persons Media Usage During a One Month Election Campaign: A Bulgarian Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 Lilia Raycheva, Mariyan Tomov, Neli Velinova, and Lora Metanova Experimental Survey on Bridging the Digital Divide Through Daily Text Chat Communication with Virtual Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 Taishi Sawabe, Masayuki Kanbara, Yuichiro Fujimoto, and Hirokazu Kato

Contents – Part I

xxi

Credibility Judgment Against Online Health Misinformation Among Older Adults: Integrated View of Psychological Distance and Health Literacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274 Shijie Song, Xinyue Li, Jianjun Sun, and Yuxiang Zhao Judging Online Health Misinformation: Effects of Cyberchondria and Age . . . . 284 Honglian Xiang, Jia Zhou, and Mingjun Liu ICT Use and Loneliness During COVID-19 Pandemic: The Case of European Male and Female Older Workers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302 Daniele Zaccaria, Emanuela Sala, and Federica Cretazzo Voice Assistants and Chatbots Learnability Assessment of Speech-Based Intelligent Personal Assistants by Older Adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321 Muhammad Usama Islam and Beenish Moalla Chaudhry Voice Controlled Devices: A Comparative Study of Awareness, Ownership, Usage, and Reservations Between Young and Older Adults . . . . . . . 348 Dietmar Jakob, Sebastian Wilhelm, Armin Gerl, Florian Wahl, and Diane Ahrens LINE Chatbot for Recording Elderly Cognition to Screen Cognitive Impairment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366 Zi-Yi Li, Agatha Maisie Tjandra, and Chien-Hsu Chen “Hi, My Name is Robin” – Remotely Co-designing an Embodied Conversational Agent for Empathy with Older Adults . . . . . . . . . . . . . . . . . . . . . . . 376 Sefora Tunc, Femke Nijboer, Angelica M. Tinga, and Monique Tabak Learning with Pedagogical Agent: Effects of Transfer of Information Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 Linran Xu, Xingmiao Li, Naimi Tang, Hefeng Wang, Manqiu Liao, and Na Liu Games and Exergames for Older People Develop Cognitive Games for Older People Based on Dynamic Difficulty Adjustment to Motivate Them to Train . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405 Yihui Cai, Xudong Cai, Xiuhui Zheng, Xingrong Li, Yiming Wang, and Yi Ji

xxii

Contents – Part I

A Study of Interactive Design Games to Enhance the Fun of Muscle Strength Training for Older Adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420 Hui-Jiun Hu From a Research Centre to a Mall: Bringing Virtual Reality and Digital Playing to a Living Lab for Community Dwelling Older Adults . . . . . . . . . . . . . . 435 Najmeh Khalili-Mahani, Juliana Yaacoub, Adrian Calcagnotto, Julian Leclair, and Janis Timm-Bottos Research on the Design of Serious Games for the Elderly Cognitive Training Based on Augmented Reality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455 Wenjing Li and Yumiao Chen Development of Exergame-Based Frailty Assessment and Management: An Older Adult-Centric Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 467 Yuezhong Liu, Bing Xun Chia, Tan Phat Pham, Yong Joo Loh, and Yin-Leng Theng “miPlay” as a Transmedia Strategy: Co-designing a Movie-Based Digital Game for Older Adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 480 Francisco Regalado, Cláudia Pedro Ortet, and Ana Isabel Veloso Effects of Physical Functions on Mobile Casual Game Acceptance of Older Adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 496 Rita W. L. Yu, Alan H. S. Chan, and H. S. Lu XR Experiences and Aging Detecting Stress in VR 360◦ Immersive Experiences for Older Adults Through Eye Tracking and Psychophysiological Signals . . . . . . . . . . . . . . . . . . . . 511 Leonardo Angelini, Corentin Bompard, Omar Abou Khaled, and Elena Mugellini Immersive VR for Training Cognition in Seniors with Neurological Disorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 522 Mohamed-Amine Choukou A Framework to Incentivize the Use of Augmented Reality in Daily Lives of Older Adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 531 Christian Eichhorn, David A. Plecher, Daniel Bamberger, Fabian von Thun, Zong-Lin Yang, Mingyang Li, Atsushi Hiyama, and Gudrun Klinker

Contents – Part I

xxiii

Development of the Train Therapy in a Nursing Home in Switzerland . . . . . . . . . 552 Sophie Jaquenod, Daniel Ducraux, Elena Mugellini, and Leonardo Angelini Applying Virtual Reality Technology and Physical Feedback on Aging in Spatial Orientation and Memory Ability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564 Ming-Wei Kang and I-Jui Lee Research on the Development of 3D Virtual Reality Fire Hazard Factor Identification Training System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 577 Yang-Chih Lin, Chin-Feng Chen, Yu-Jen Chen, Ren-Bei Yang, and Jia-Yun Zheng Virtual Reality Hippotherapy Simulator: A Model Proposal for Senior Citizens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 592 Cláudia Pedro Ortet, Mário Vairinhos, Ana Isabel Veloso, and Liliana Vale Costa The Use of Immersive Technologies While Ageing in the Digitally Mediated Society . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 610 Cláudia Pedro Ortet, Liliana Vale Costa, and Ana Isabel Veloso Habituation to Simulator Sickness in Older Adults During a Chronic Back Pain Therapy Program with Repeated Virtual Reality Exposure . . . . . . . . . . . . . . 627 Oskar Stamm and Rebecca Dahms Investigating the Mental Workload of Experiencing Virtual Reality on People with Mild Cognitive Impairment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642 Pei-Fen Wu, Shao-Wei Yen, Kuang-Yi Fan, Wen-Fu Wang, and Feng-Chu Wu A Study of Junior High School Students’ Willingness to Learn Table Tennis Using VR Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 655 Fangfang Yang, Longfei Ren, Chao Gu, and Jinguang Wei Correction to: College Students’ Perceptions and Preferences Regarding Intelligent Advisory Systems in Multi-device Learning Environments . . . . . . . . . Zhao Xu and Qingchuan Li

C1

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 667

Contents – Part II

Smart Homes and Aging in Place An Iterative Approach to User-Centered Design of Smart Home Systems . . . . . . Lauren C. Cerino, Chaiwoo Lee, Sheng-Hung Lee, Shabnam Fakhr-Hosseini, Heesuk Son, Shen Shen, and Joseph F. Coughlin Research on Influencing Factors of Elderly User Experience of Smart Home Social Software Based on Grounded Theory . . . . . . . . . . . . . . . . . . . . . . . . . Ke Ma, Meng Gao, Francesco Ermanno Guida, and Renke He My Iliad: A Ludic Interface Using Ambient Assistive Technology to Promote Aging in Place . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hubert Ngankam, Célia Lignon, Maxime Lussier, Aline Aboujaoudé, Renée-pier Filiou, Hélène Pigot, Sébastien Gaboury, Kevin Bouchard, Guy Paré, Carolina Bottari, Mélanie Couture, Nathalie Bier, and Sylvain Giroux

3

17

31

Visual Ambient Assisted Living Technologies for Different Daily Activities: Users’ Requirements and Data Handling Preferences . . . . . . . . . . . . . . Julia Offermann, Caterina Maidhof, and Martina Ziefle

47

Smart Home for the Elderly - A Comparative Study on Interaction Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Monika Schak, Isabell Bürkner, Rainer Blum, and Birgit Bomsdorf

66

Living in a Networked Home: Older Adults’ Expectations and Attitudes . . . . . . . Yash Trivedi, Shabnam Fakhr-Hosseini, Chaiwoo Lee, Sheng-Hung Lee, Lauren C. Cerino, and Joseph F. Coughlin Data Privacy and Smart Home Technology Adoption: Older Adults’ Attitudes and Beliefs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Manasi Vaidya, Chaiwoo Lee, Lisa D’Ambrosio, Sophia Ashebir, and Joseph F. Coughlin

80

96

eHealth Applications for Older People and Health Literacy Health Education Mobile Applications: Evaluation of Persuasion Strategies and Impact on Older Users . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109 Shushan Fan and Yongyan Guo

xxvi

Contents – Part II

Designing Intelligent Decision Assistants to Assist Seniors with Medicare Plan Decision-Making: An Application of Cognitive Response Theories . . . . . . . 121 Ebenezer Nana Banyin Harrison, Wi-Suk Kwon, Xiao Huang, Nick McCormick, and Salisa Westrick Aging in Place Virtual Care Technology from the User Experience Perspective . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 Irina Kondratova, Helene Fournier, and Fateme Rajabiyazdi Privacy, Technology and Telehealth: Canadian Older Adults Voice Their Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145 Constance Lafontaine, Marie-Ève Ducharme, and Kim Sawchuk Paving the Way: Trust in Healthcare Systems as a Prerequisite for Technology Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160 Sophia Otten, Julia Offermann, and Martina Ziefle Trust in Health Information Among Older Adults in Iceland . . . . . . . . . . . . . . . . . 173 Ágústa Pálsdóttir Tamamon: Designing a Gamified Medication Education Solution for Older Adults Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186 Tan Phat Pham, Yin-Leng Theng, Keng Teng Tan, Jie Chong Lim, Su Qi Hong, Nur Atiqah Binte Mohamad, and Tng Shu Ting Valerie The Age Factor and the Acceptance: A Rhetorical Study on the Older Audience of the COVID-19 Popular Science Animated Films . . . . . . . . . . . . . . . . 197 Yi Su Using Persona Development to Design a Smartphone Application for Older and Younger Diabetes Patients – A Methodological Approach for Persona Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214 Fatima Varzgani, Soussan Djamasbi, and Bengisu Tulu Effects of Health Science Popularization Platform on Older Adults’ eHealth Literacy: From the Perspective of Digital Empowerment . . . . . . . . . . . . . 228 Yixuan Wang, Xiaoting Xu, and Qinghua Zhu Behavioral Intention Model for Online Consultation at Internet Hospitals Among Older People at Home . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246 Mingwen Zhang, Liangchen Jing, Qizhi Wei, and Cong Cao

Contents – Part II

xxvii

IT Support for Caregivers A Study for Estimating Caregiving Contexts Based on Extracting Nonverbal Information from Elderly People at Home . . . . . . . . . . . . . . . . . . . . . . . 259 Sinan Chen, Masahide Nakamura, and Kiyoshi Yasuda Caregiver Attitudes Toward Internet-Enabled Sources of Information . . . . . . . . . 269 Adam Felts, Alexa Balmuth, Chaiwoo Lee, Lisa D’Ambrosio, and Joseph F. Coughlin What to Do Next? An Activity Scheduling Schema for Social Assistance Robots for Older Adults . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 David Gollasch and Gerhard Weber Technologies and Services to Support Care: Family Caregivers’ Experiences and Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307 Chaiwoo Lee, Alexa Balmuth, Lisa D’Ambrosio, Adam Felts, and Joseph F. Coughlin Observations of Caregivers of Persons with Dementia: A Qualitative Study to Assess the Feasibility of Behavior Recognition Using AI for Supporting At-Home Care . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331 Wilson Lozano, Sayde King, and Tempestt Neal Engagement as a Goal and Process for Improving Support for Informal Caregivers: The Cremona Beside Caregiver Project . . . . . . . . . . . . . . . . . . . . . . . . . 345 Michele Paleologo, Eleonora Gheduzzi, Rita Bichi, Maria Grazia Cappelli, Matteo Donelli, Niccolò Morelli, Cristina Masella, and Guendalina Graffigna Live Classification of Similar Arm Motion Sequences Using Smartwatches . . . . 357 Sergio Staab, Lukas Bröning, Johannes Luderschmidt, and Ludger Martin Technology Service Design for the Older Adults with Dementia . . . . . . . . . . . . . . 377 Wang-Chin Tsai, Chia-Fen Chi, and Yu-Hsing Huang Effectiveness of Support Programmes for (in)Formal Caregivers of Older Dependent People to Design Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392 Vanessa Zorrilla-Muñoz, Alberto Veira-Ramos, María Silveria Agulló-Tomás, Nicolas Garcia-Aracil, and Eduardo Fernandez

xxviii

Contents – Part II

Aging, ICT Use and Digital Literacy Intergenerational Solidarity: Perceptions of Young Adults . . . . . . . . . . . . . . . . . . 413 Inês Amaral, Ana Marta Flores, and Eduardo Antunes Understanding ICTs in Older Life: A Scope Review of ‘The Gerontologist’ and ‘Research on Aging’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 426 Simone Carlo and Sara Nanetti Applying PBL Model to Technological Learning Strategies for Older Adults . . . 443 Kuei-Yuan Chang and Chang-Franw Lee ICTs in Later Life and Post-pandemic Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . 457 Fausto Colombo, Piermarco Aroldi, and Simone Carlo Digital Media Use and the Role of Internet Self-efficacy Among Older Technology Volunteers – A Baseline Study from the Project “DiBiWohn” . . . . . 472 Michael Doh, Joshua L. Schlichting, David Leopold, Linda Göbl, and Mario R. Jokisch First and Second-Level Digital Divides and Cultural Capital: Framing Digital Lives of Seniors in Portugal and Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . 489 Tiago Lapa, Teresa Martinho, and Célia Reis Influence of Cognitive Function and the Mental Model on ICT Use – Examined by TAM Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505 Taiga Nohara, Kaoru Takagi, Ikuko Sugawara, Yasuyuki Gondo, and Misato Nihei A Social Justice-Oriented Perspective on Older Adults Technology Use in HCI: Three Opportunities for Societal Inclusion . . . . . . . . . . . . . . . . . . . . . . . . . 519 Linnea Öhlund Value Co-creation Perspectives on Digital Literacy Training for Older Adults: A Call to Action Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 533 Yuxiang (Chris) Zhao Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543

Designing and Assessing the Older Users’ Experience

Designing for Self-directed Learning: Co-creating a Demokit with Older Adults Katerina Cerna1(B) , Claudia Müller2 , Martin Hunker2 , Florian Hitpass2 , and David Randall2 1 Applied IT Department, University of Gothenburg, 41756 Gothenburg, Sweden

[email protected]

2 Information Systems and New Media Department, University of Siegen,

57068 Siegen, Germany {claudia.mueller,martin.hunker,florian.hitpass, david.randall}@usi.de

Abstract. Supporting older adults in developing their ability to learn using digital tools in a self-directed way holds the potential to make them more autonomous. A promising space for that is in participatory design. However, currently we lack the tools to support their learning in a sustainable way during as well as beyond a single project. Hence, in our study conducted in Germany, we explore development of a demokit, a set of learning resources to various degrees co-created with our older participants. Through a year and a half long study taking place between December 2019 and September 2021, we have engaged with a group of 17 older adults in activities oriented towards self-directed learning as well as tested our findings during a public event. With our findings we illustrate the collaborative and situated aspect of self-directed learning when participants explored our cocreated artifacts. We conclude by discussing autonomy in relation to self-directed learning of older adults and collaborative aspects of their learning activities. Keywords: Participatory design · Demokit · Self-directed learning · Older adults

1 Introduction The notion of human autonomy is hardly new. It can be traced back, at very least, to Spinoza’s arguments concerning ‘the good life’ (see e.g. [1, 2]) and resonate with Maslow’s ‘hierarchy of needs’ (see [3]). More recently, such concerns have been manifest in a wide variety of contexts. They include colonialism and decolonialism ([4, 5]), healthcare (e.g. [6]), human-machine interaction and AI (e.g. [7–10]), marginal groups (e.g. [11–13]), and moral and political philosophy (e.g. [14, 15]) and centrally for our purposes, interaction with, by and for older adults (see e.g. [16]). A consistent theme running through these diverse literatures is that human autonomy is not something to be ascribed, but something to be achieved, a view we share. In what follows, we make a modest contribution to the prospect of autonomy of older adults through attention to selfdirected learning. Self-directed learning, the ability to selforganize what is needed and © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 3–22, 2023. https://doi.org/10.1007/978-3-031-34866-2_1

4

K. Cerna et al.

wanted in an autonomous way rather than receiving it as a ‘gift’ from outside agencies is hence of key importance in this process [17]. This type of learning is viewed as critical to “survival and prosperity in a world of continuous personal, community, and societal changes” ([18], p.32) This is especially relevant for older adults when it comes to the digital landscape of the current global world. Despite the fact that that there are many efforts to provide the older adults with access to learning resources, for example via NGOs, these are typically not well-resourced and many older adults still struggle with digital devices use on a daily basis (see e.g. [19]). Fostering self-directed learning of older adults holds the potential to help them overcome these barriers in a sustainable way. The concept of autonomy briefly outlined above suggests that autonomy cannot be considered merely as an outcome. It is a defeasible achievement as well. We argue that self-directed learning of older adults can be fostered and achieved through sustainable participatory design (PD). There is an important distinction to be made here between the outcomes of a PD process and what participants to that process might learn from it. We would argue that the existing literature tells us much more about the former than it does the latter. If PD is to be sustainable then participants must, by definition, be able to transfer the skills and capabilities they have learned. As Di Salvo et al. [20] have pointed out, ‘The goal of the learning sciences is not only to understand the phenomenon of learning but also impact educational practices.and enable more effective learning.” They go on to say, “Until now, however, the learning sciences community has not focused on the design of artifacts for supporting learning… Nowhere is this oversight more evident than with regard to engaging stakeholders actively in the design process.” To which we might add, such communities have also paid little attention to the design and development of sustainable skills and expertises of the participants. This is even more true for communities which historically lack the experience of using digital tools, such as older adults. The notable exception is a co-design study highlighting that learning can become a useful result of the co-design sessions with older adults [21]. However, the same study also points out that learning is not an automatic consequence of the PD process and that the design space needs to be configured in a certain way. Our approach aims to address this problem by co-creating a mobile demokit together with our older participants. Our main research aim is to understand how to organize participatory design projects with older adults in a sustainable way that fosters their own learning to use digital tools. More specifically, the following research questions have guided our study: • How can we understand self-directed learning of older adults when older adults and younger researchers collaborate within participatory design? • How do older adults overcome the challenges connected to self-directed learning in a learning space? Through a year and a half long study from December 2019 to September 2021, we have engaged with a group of 17 older adults in PD activities oriented towards selfdirected learning. We further evaluated our findings during a public event in a small town in Germany, which this study reports from. Co-creation of the demo-kit has allowed us to study self-directed learning in two ways: during the actual PD process but also later when evaluating it. We have reported about the actual PD workshops elsewhere [22–25], and this study focuses on the final step of the process, the evaluation of the demokit during a public event. We argue that to achieve sustainable PD with older

Designing for Self-directed Learning

5

adults, we need to foster their self-directed learning by enabling for different modes of collaboration, providing suitable support in guiding their attention and collaboratively creating connections between older adults’ lives and novel technologies. A vital feature of the learning process in this context, and one which our research addresses, is the ongoing and rapid changes in the digital landscape. We contribute with the following: • We articulate a more nuanced understanding of the concept of self-directed learning to improve the PD of for and with older adults, in particular in co-creation of demokits for sustainability. An advanced understanding of the dimensionality of self-directed learning should be useful for the general domain of sustainable PD. • Rich empirical material that shows the facets of older adults’ self-directed learning • We contribute to the HCI literature by reflecting over lessons learned evaluating our mobile demokit through digital literacy fostering.

2 Related Research In this section, we first discuss the current literature on learning and PD and more specifically, how it is related to the learning of older adults. Next, we provide a brief overview of the relevant literature pertaining to kits for older adults. Thirdly, we present our conceptual understanding of self-directed learning, both in learning sciences and in HCI. Finally, we point out a research gap. 2.1 Older Adults’ Learning and PD The basic assumption of geragogy and gerontology is that older adults’ learning is different from that of other age categories and hence need to be supported and fostered in a different way. The different learning needs connected to digital tools cannot be reduced to only physical changes connected to aging, such as shaky hands, problems with short term memory etc. [17]. Instead, older adults have experienced different learning trajectories and were socialized with technologies in a different way, making them used to different learning styles when it comes to new tools [26]. In addition, older learners often have heterogeneous learning goals; while some are interested in being able to do specific tasks, others want to stay updated or socialize through digitally interested communities [27]. One of the key aspects is motivation - research suggests that older adults are motivated to learn and use digital tools, if they make a connection to their individual interests [28, 29]. Further, a typical barrier is anxiety and fear connected to failing using digital tools [30]. For example, some older adults prefer sticking to doing the same task in the way they are familiar with, to avoid failing [31]. Another important aspect is the social context - as older adults learn using digital tools in informal ways, their social networks often become the main resource where they learn to use a range of digital devices [22]. This holds especially true when it comes to digital tools and online learning environments [32]. PD is an approach to technology development that prioritizes involvement of users into the design process [33, 34]. PD has a long tradition of highlighting autonomy - its origins are connected to bringing back control to the future users [34]. Traditionally, a key step in gaining autonomy is to learn about the new tools that are being designed

6

K. Cerna et al.

for one’s practice, and that one can “control” what tools the people will be using for their work. Hence, through the PD process one will get the opportunity to learn and as a consequence be able to decide better [35]. There are different types of learning processes that are taking place when people take part in PD [20]. One of the key types of learning takes place during the process of co-creating an artifact together, often referred to as “mutual learning” [36]. During this process, thanks to the “asymmetry of ignorance” [37] people with different types of experiences and expertise can come together and learn from each other. In the context of designing for and with older adults, this potentially becomes doubly beneficial because then not only can the researchers learn how to build a systems which might make more sense for users; but taking part in such a process can benefit older adults in various ways, for example, building up digital capacity [33]. Another key type of learning is appropriation of (digital) tools, which can take place during different phases of the PD process [20]. This, however, is not a straightforward matter. Digital tools themselves evolve very rapidly. Skills are skills only if they are relevant to the context in which they are to be used. How and what it is that older adults need to learn in order to start appropriating digital objects [38] requires careful examination. In PD projects focused on digital tools which take enabling participation seriously, older adults first need to be supported in the appropriation of the relevant digital tools [29]; and only after they have become more confident and digitally skilled and as a consequence able to envision new technologies, it is possible to proceed to co-design together [33]. We also need to pay attention to the different sociomaterial configurations of the PD space, as it will be creating and impacting the power dynamics between the researchers and older participants [21]. In other words, the current participatory and cocreation with and for older adults recognizes the need to address the building of digital capacity of older adults. Yet, so far, these different learning processes, we argue, have not been considered when building toolkits with and for older adults. 2.2 Demokits and Older Adults Demokits - a set of tools designed for a certain purpose - are becoming commonplace in HCI. Thus far, however, they are aimed for the most part at supporting a research project. They do not usually have an evolutionary character allowing for the possibility of developing a set of resources or tools that can themselves, while forming the basis for current use, be co-developed during the project with the aim of providing for flexible and sustainable use even after the end of the research project [39]. Put simply, co-creation has not been combined with the sustainability issue in the contexts we are concerned with. A common approach in participatory design with and for older adults is to develop a toolkit to support the co-creation process. For example, by making the co-designing process more accessible to older adults. Examples include, [40] designed infobricks, a set of tangible computer modules which could be combined in a flexible manner. Another toolkit focused on exploring possible future technologies through an inventor kit made of everyday objects and electronics [41]. Berger et al. [42] explored stories people tell through using an IoT-based toolkit to explore intimate relationships at home. Finally, some have also used crafts as a framing for exploring and making with an electronic toolkit [43] or set of sensors and other media [44]. These studies have hence focused on supporting the co-designing process over the period of a research project. However,

Designing for Self-directed Learning

7

fewer studies have considered the possibility of developing a set of resources or tools that might be co-created during the project as a means to sustain some of its results beyond the life of the research project. For example, [39] developed a set of tools together with teachers supporting neurodivergent children, so that the teachers can continue with codesign activities even after the end of the research project. We want to build on this body of work and develop a set of learning resources together with older adults as a way to support sustaining their ability to use digital tools. 2.3 Self-directed Learning in HCI Conceptually, we understand self-directed learning as a type of activity that older adults do when they are using digital tools which are to some extent new to them. Despite the term “self”, this activity is a combination of both autonomous as well as social (or collaborative) elements. These processes can happen in an informal situation (day-today living) or formal one (senior computer club). Self-directed learning is informed by broader constructivist philosophy, where instead of the teacher, it is the learner who directs their own explorations of the world based on their own interpretations [45]. Hence, there is no pre-defined curriculum created by a more knowledgeable other to follow. Brookfield defines it as “a process in which individuals take the initiative in designing learning experiences, diagnosing needs, locating resources and evaluating learning” [46], p. 40]. Brookfield further states that self-directed learning does not equal autonomy. Instead, Brookfield argues for “a more nuanced understanding of self-directed learning, one which promotes the awareness of new possibilities for adults and relates quite closely to the critical reflectivity and transformative learning”. In other words, developing motivation to learn, identifying one’s own learning goals, taking action to learn but also evaluating if one has learned in a successful way are the key aspects of self-directed learning. Despite the strong orientation of the learner, and the “self” in self-directed learning, learning does not happen in isolation or independently of others. On the contrary, even this type of learning is heavily grounded in social context [46]. Self-directed learning is also formed through collaboration and in interdependent relationships [47]. Despite the fact that the decisions within these learning processes have to be made by the individuals, the meaning through which they become able to make these decisions is not made independently of other people. It is an interconnected set of relationships in which “the individual takes responsibility for constructing meaning including the participation of others in confirming worthwhile knowledge” [48] p.19. However, we currently lack understanding how self-directed learning of older adults’ develops. Self-directed learning can differ in duration and can be also a continuous effort and a life-long effort; however, it holds especially potential for the older adults and their use of digital technologies [49]. A survey conducted by [50] suggested that self-directed learning holds the potential to foster higher quality of life. [51] presents self-directed learning of older adults as an approach that helps them to negotiate the changes in social life that are connected to aging. Through qualitative interviews he shows that older adults make their different learning projects work through approaching them as leisure activities. Self-directed learning can be triggered by either external (what others want the person to do) or internal motives (what the person wants to do) [49].

8

K. Cerna et al.

In the HCI literature, self-directed learning has not been conceptualized in any consistent way yet. In this type of literature, self-directed learning is an approach that is often used as a concept in different practice contexts for designing innovative learning experiences. Especially in formal, school-based learning contexts, the promotion of active and student-centered learning is an important goal that is pursued, among other things, with the design of combined analog and digital didactics and methodologies, e.g. in the form of “CrossActionSpaces" [52]. Novel student-centered learning spaces are also being pursued in formal school contexts through the use of game worlds such as Minecraft [53]. 2.4 Gap in the Current Research To sum up, the current literature on PD and co-creation with older adults puts an emphasis on enabling participation and learning during the design process in question [20, 29, 33]. Emerging research on kits for sustainability has so far included mainly professionals such as teachers [39]. In the field of HCI, this concept is addressed more in formal school-based learning contexts [52, 53]. A systematic examination for informal learning contexts of older and not technology-savvy persons is still missing. However, as older adults’ learning to use digital tools is (conceptually) different than that of for example professionals, we aim to contribute to this gap by co-design a demokit with and for older adults, targeting their self-directed learning.

3 Method In this study, we aim to both understand the studied phenomena of self-directed learning, as well as to foster it. In this section we describe the empirical setting where we studied self-directed learning of older adults and then we describe what data we used to illustrate our ideas. 3.1 Empirical Setting The background of this paper is an interdisciplinary and multinational project which aims to promote fostering digital literacy of older adults as a way for them to gain more autonomy. The project started in December 2019 and finished September 2021, taking place in a small town in Germany. Our goal was not to build a technology that would simply enable the autonomy of the older adults; rather, we wanted to create a learning environment, where through PD, the older adults would become capable of developing their self-directed learning connected to digital tools. The national subproject whose team’s work is reported in this paper is based in a small German town and focuses on the role of PD in this process and on exploring the measures for including older adults into participatory processes as a way to increase their digital literacy. The practical goal of the subproject was a demokit, a set of learning resources which were created for or together with the older participants. The purpose of this digital artifact was to foster learning of the older PD participants when creating the artifact, but also to foster the learning of people who will interact with the artifact at a later stage; these can be older

Designing for Self-directed Learning

9

adults or people working with this target population. The original plan was a physical object, i.e. a box with digital artifacts and didactical material, which would be possible to carry around and move from one context into another, hence making the findings of our project more sustainable (as in not limited to only the setting of its creation). However, due to the COVID-19 pandemic, we started to build a website first (more details below) as the first step of the demokit. This paper draws on the empirical setting where we presented the demokit in an interactive way. 3.2 Online Co-creation with Older Adults The original plan was not to conduct this empirical work online, however due to the empirical work taking part during the current COVID-19 pandemic, this was the only option how to go about it. As a consequence, we transformed the PD workshops into online format (more about the transformation itself we reported already here [25]. Altogether, we organized five in-person workshops and 27 online workshops (together 32 workshops). Each workshop took two hours and had on average eight participants. Workshops took place every two weeks first (Phase 1), and later once every month. Each workshop was focused on a different topic, which we have reported more in depth about here [23]. 17 older adults between 65 and 80 years old took part in the online activities (with an average number of eight at a time). The participants were not financially compensated for their time, they were provided coffee and snacks during the on-site workshops. All the online workshops were recorded, all the on-site activities were documented by writing field notes and taking videos and photos. The workshops recordings were watched multiple times, relevant sections were then transcribed and analyzed further on, through iterative coding within the research team. The analysis was taking place throughout the project and contributed to building of the demokit. 3.3 The Demokit The demokit is a website1 , where we gradually present the different digital artifacts, which we have to a varying extent co-created together with our older participants. We call these digital artifacts “didactic prototypes” and through their creation we are trying to address the learning processes involved in PD. The purpose of traditional prototypes is to demonstrate what we as designers or PD participants know about the given practice at that particular point of time [54]. They can for example represent understanding a workflow or knowledge needs so that it can be supported by a new tool. Didactic prototypes also capture what participants need to know, but in addition are oriented towards people who are outside of the PD process [24]. Didactic prototypes hence do not per se serve as a communication of design ideas; instead, when made available, they host the potential to serve as a learning resource for older adults who are outside of the project as well as other stakeholders relevant to older adults. As such, the didactic prototypes capture learning and as a consequence prompt learning in others. 1 The demo-kit was mainly co-created in German, however some pages are also in English:

https://acsdemo.wineme.fb5.uni-siegen.de/.

10

K. Cerna et al.

Currently, there are four different types of didactic prototypes included at the website. Each is illustrated by an example: • Created by us researchers based on the work with the older adults (An interactive instructions on how different parts of Zoom interface are called and how to use them) • Created by the older adults (a PDF with written by one of the participants on how to use online banking) • Created by students including different datasets with older adults (a video, where privacy concerns connected to Alexa are discussed in an animated scenario) • Co-created by us researchers and the older adults during our online workshops (Q/A on how to use robotic vacuum cleaner) The demokit website currently includes 33 webpages. We have discussed different versions of the demokit with the older participants multiple times during our online workshops and include their feedback into its further development. 3.4 Public Event Evaluation with Older Adults The next step in our activities involved understanding how the demokit might support self-directed learning of older adults “in the wild”, that is, outside of the workshop contexts. Because the goal of the demokit is to support learning of people interacting with it in different settings, we decided to take part in a local urban festival and present our demokit there (August 2021). The theme of the festival was playing, hence we framed our station as a test-it-out room, where participants could come and test out different digital tools but also learn about how to use them. Though we did not target specifically older adults, our assumption was that a station like this would attract mainly people who struggle with digital literacy rather than fully digital natives. The event took place over two days (four hours each day). The research team on the site included three researchers (first author, third and fourth author). First, we had to take certain measures to transform our demokit into a suitable form which could be presented during the event. Several types of translation work are necessary when creating a similar tool: when using the field work to create the demokit; when choosing the different aspects and how they can be formed through a hybrid of online and physical objects, and when on site, supporting the older adults in making sense of the demokit and its elements on the spot. We created 9 stations, each equipped with a table, a paper with brief instructions and a QR code. The QR codes served also as a learning moment, as we know from our workshops that it is not something that our older participants know how to use, however commonly encountered. Hence, the first thing we did with all our participants was to make sure they had a QR app on their phone and they knew how to use it. Only one of the participants had the app already installed, with the other five we had to install the app together.

Designing for Self-directed Learning

11

Fig. 1. Compilation of images representing the demokit demonstration during the public event (robotic cat, vacuum robot, memory game with technological terms).

Next, we included the following digital and material artifacts for people to explore: a robotic pet, smart speaker (Alexa), Telegram group (Telegram is a type of a messenger, which we used for coordination with the older adults during our project), vacuum robot, memory game with technological terms and online editor (see Fig. 1 for examples). The instructions were purposefully brief, because we wanted to motivate people to explore the digital artifacts by themselves and when in trouble, to interact with the demokit. These were also the instructions every participant got at the entrance. However we also re-assured the participants that we are there to help and they could turn to us at any moment. Third, to promote self-evaluation as an aspect of autonomous engagement with the demokit on the site, we designed a survey flier, which included spaces for the participants to map and reflect over their digital ecologies but also sections with statements starting “I am able to”, for example, “ask Alexa about her privacy settings” or “make the robotic cat roll over” connected to the tasks at the different stations. We have also invited our older participants through our Telegram group. Only six people came and all of them were from our PD workshops. However, that turned out to be a great opportunity to reconnect and even better understand the possibilities of how to understand and support their self-directed learning. More details about our participants can be found in Table 1. Table 1. An overview of the participants age, gender and digital literacy level. Participant #

Age

Gender

Digital literacy

Participant 1

82

Male

Worked with digital artifacts before (programming) interested in codes. Used to work with computers (came together with Participant 2)

Participant 1

82

Male

Worked with digital artifacts before (programming) interested in codes. Used to work with computers (came together with Participant 2)

Participant 3

74

Female

Uses Alexa at home, regular tablet and smartphone usage, uses Zoom on weekly bases (choir meetings)

Participant 4

81

Female

Very new to technology, beginner with tablet and smartphone

Participant 5

89

Male

Very firm with laptop, used to work with smartphone and smartwatch

Participant 6

78

Female

Owns a mower robot, sometimes troubles with laptop and smartphone handling

12

K. Cerna et al.

We have documented the event by writing down field notes, making videos and taking pictures. These materials were further analyzed together with the previously collected material. All the names of people in the transcripts and of the local computer club were changed to protect our participants’ privacy.

4 Self-directed Learning Through Demokit Exploration In the following section, we will show how self-directed learning was fostered through demokit exploration. Examples illustrate that by enabling different modes of collaboration, providing suitable support in guiding their attention and collaboratively creating connections between older adults’ lives and novel technologies. 4.1 Example 1: Contributing to Collaboration in Different Ways Self-directed learning takes place in relation to others and collaboration with them. Because older adults are a heterogenous group, the way they choose to contribute to such collaboration might differ. The first example comes from a station with the vacuum robot. The task here was to build hindrances for the vacuum robot and then start it with its app, which was downloaded on a provided smartphone. A QR code was again provided and an instruction to use our demokit with a Q/A about the vacuum (The Q/A were cocreated during the online workshops by the older participants). None of the participants was familiar with the app, hence it allowed us to well understand how such an app can be approached by the participants by themselves. Participant 3 and Participant 5 approached this station together (they knew each other from our PD workshops). Researcher 2 who was responsible for this station, provided the participants with brief instructions describing their task (placing hindrances on the floor and starting the vacuum through the app). Researcher 3 was observing and recording the situation. The participants placed the hindrances on the floor, and then they turned their attention to the app. Here, the researcher did not provide detailed instructions on how to navigate the app - instead, he prompted the participants to engage with in a way they thought was meaningful (to achieve the task of starting the robot). The participants started interacting with the smartphone placed on the table and the following situation took place. Participant 3 starts interacting with the app, seeking a way to activate the robot. She reads outloud what she sees on the screen, such as “Language support, debugging and errors, we don’t need all that (.) ne. That’s not the right thing to do.” The researcher confirms, and gives her small hints on what to do next “Yes, then you have to go back. Otherwise you can look again, if you find something else somewhere else.” Participant 3 continues exploring the previous vacuuming events, meanwhile Participant 5 continues standing next to her, observing her every step. After managing to get back to the main screen, the Participant 3 decides to explore robot settings, and the researcher prompts her by saying “You can just try it out.” and then the following conversation takes place: Participant 3: The thing has to run somehow. August (.) Researcher: August is the name of the robot.

Designing for Self-directed Learning

13

Participant 3: I thought you meant the month. [Presses on name August on the screen]. He didn’t make any noise. Participant 5: [Presses restart button, robot beeps] Participant 3: Ok. Researcher: Now you heard him once briefly. He restarted once. Participant 3: Yes, ok. Researcher 2: Now the app is rebooting once. So, it’s connecting. Otherwise, read again what it says “everywhere”. Participant 3: Vacuum everywhere. Researcher: You can click on that. Participant 3: [Presses the button and the robot starts vacuuming] (Transcript of a video, August 2021) The task to start the vacuum has been achieved through a collaboration between the older adults and the researcher. The participants approached the task in very different ways. Participant 3 kept on exploring the app, searching for the right button to start the robot. She was only very carefully tapping the screen, verbally describing what she sees in the interface, making available to the researcher not only what she sees but also how she understands the presented information on the screen. Participant 5 chose a very different approach - he let Participant 3 to explore the app, without any visible effort to change what she was doing (despite how correct or incorrect her efforts could have seemed in relation to the task). Once she reached the screen with the right button to start the vacuum, Participant 5 reached out and pressed the correct button, still without any verbal confirmation. At that point, the researcher provided an explanation of why the robot made a sound, interpreting the robot’s feedback to the participants so that they can progress with the activity. This example shows how self-directed learning is accomplished as a collaborative activity between the participants; and where the presence of the researcher helps the participants to make sense of the app, making sure both participants understand the different steps in the app (such as going back by swiping), vacuum feedback (sounds) or how to handle previous information displayed in the app. 4.2 Example 2: Guiding Attention to Self-correction Digital tools use might be challenging to those who are not familiar with them, because of seemingly similar interfaces or too small elements of the interfaces. However, they can still learn to navigate in them, especially when their attention is guided to self-correct. Another station focused on interacting with a Telegram group, which we created for the purpose of this event. Even though we used Telegram for our regular coordination with the older adults, they often forgot some of the key functions, hence part of the demokit was focused on practicing using Telegram functions as well. A common problem we have observed during both online workshops and on-site events is that the older participants cannot achieve the task they want to do with a digital tool, because they cannot find the right information in the interface. In contrast to the online space, in person it was possible to identify that this type of support might be taking place, and provide them with hints on how to solve it, as will be illustrated in the

14

K. Cerna et al.

following situation. Participant 4 and Researcher 3 came together to the Telegram station. The task here was to join the Telegram group through the provided QR code and then post an image in the group of the most favorite station. Participant 4 wrote her message in the group, however did not press the - quite small - send button and then jumped out of the conversation. Researcher 3 showed her on her own phone that the image did not arrive yet, prompting the Participant 4 to check again what happened in Telegram. Participant 4 opened her Telegram again, and clicked on the first chat that she sent the last message to (which was different from the group chat) and before the researcher could stop her, she sent the cat image to a person in her chat. The researcher pointed out again that the original message was still in draft on the phone of the participant, who wondered “Who did I send it to”. The researcher pointed to the name of the person stating that this is where you can see who you sent it to. Participant 4 explained that it was another participant from her computer club and only then understood that it is possible to read the name of the person one is sending a message to even when the chat is actually opened. She was then able to proceed to the original (right) chat and send off the drafted message. This example illustrates that the older adult was capable of doing the task (sending an image) and was able to correct it herself, when the researcher guided her attention as means to identify that a mistake was made. The key issue here was that the participant did not even notice that she had made a mistake. Learning from their own mistakes is a key element in learning. However, it is also important to help the older adults understand that it is fine making mistakes, as many of them are often afraid to “break” digital devices they are learning to use. This was the main reason why we included practicing Telegram into the demokit - because the older adults need support not only in the big things (sending images or making a telegram call) but also in these small details, such as pressing a small blue button or reading the right information at the right time. 4.3 Example 3: Waiting for Instructions vs Active Engagement In this example, we will again encounter Participant 3, this time at two different stations: the robotic cat and Alexa. First, the participant approaches the table with Alexa station on it. The Alexa task included instructions to ask her about privacy concerns and a QR code was provided to lead to the demokit website with a scenario about Alexa (as described above, section The demokit). From our workshops we knew that Participant 3 has Alexa at home and is a promoter of such technology for older adults. When she comes to the station, she ignores the instructions and instead, the following situation takes place: Participant 3: sat down at the Alexa table and told everybody around her that she also has an alexa hat home. She immediately shows off with her Alexa skills, that she knows how to start a radio station, stop the music, set a timer with a specific name, but when she tried to stop it, it didn’t work out at first try, because Alexa didn’t understand her properly and she proceeded to ask for the weather in [her town]. Participant 3: showed that she is signed up for the alexa newsletter and pulled up the latest email on her phone with the newest skills and she tried them out directly and also showed them to the person next to her, how it gets more features over the times with updates. (Field notes, August, 2021).

Designing for Self-directed Learning

15

These notes illustrate that Participant 3 approached the station with Alexa in a very active way, engaging with it without the instructions. She even applies Alexa’s new skills as well as shares them with a person sitting next to her. After she was done with the Alexa station, she continued exploring the room. We found her only passively sitting by the cat, with her arms folded. The task written on the paper close to the cat was to make the cat meow, which was possible by finding a button on the cat. The instructions pointed to a QR code, which led to instructions on our demokit website. When the participant was asked by the supporting researcher about how it is going for her, she exclaimed: “I am waiting for you to give me tasks!” (Field notes, August 2021). In turn, the researcher prompted her to actually read the instructions that were in front of her on the paper. She then started to engage with the written instructions and consequently also QR and was able to finish the task (and made the cat meow). She also expressed her concern that she did not see a clear purpose for the cat. Researcher explained that it is not clear yet for research either and hence why we are exploring it, for example to use it to accompany people with dementia. After the participant got home, she sent the following message to us: “ Dear Project Team! I enjoyed it with you - and I also learned something new. I was most interested in the [techno] memory, it would be nice if we could talk about it again in the group. The cat has triggered very mixed feelings in me! The thought of possibly becoming demented often occupies me and also worries me. You continue to have success!” The same person had a very different experience in each of the stations. The instructions for both stations could be considered similar (a paper with a written task on it and further support accessible through QR to the demokit website). However, the previous experience and familiarity or lack of it determined the whole experience of the participant. Meanwhile Alexa prompted playful demonstration of her own digital skills and even sharing extra information with us and others, the robotic cat did not prompt any of these feelings. This is not difficult to explain - the participant was used to Alex and hence even applying new skills was possible for her. On the other hand, handling a robotic cat is not a usual situation - it can in itself feel like an unfamiliar and strange practice for the participants. This example illustrates that in this context the older adult did not struggle with the technology because it was too complex to deal with, but because of potential lack of meaning for the older adult. However, as her final message suggested, the cat and conversation around its potential use has left an impact. There are several important points to make about self-directed learning here. Participant 3 maybe did not learn anything new from that moment, but by engaging with Alexa in an enthusiastic and confident manner, she could inspire and maybe also motivate others to engage with a tool that might be new to them. At the same time, when it comes to tools that are new to the participants, participants might need a stronger prompt to focus on the available instructions. The participant at the end could manage to accomplish a task as well as engage in making sense of the tool together with the researcher.

16

K. Cerna et al.

5 Discussion Autonomy in general has been substantially discussed in various fields, as we have pointed out. Though in HCI a frequently addressed topic, it is a concept that was not until recently very well conceptualized [24, 55]. In our paper, we focus on a specific aspect of autonomy connected to the learning of older adults, which we understand as self-directed learning. The basic assumption of self-directed learning is that it is the learner who organizes their own learning. However, when it comes to older adults’, learning to use digital tools entails an ambition to achieve a degree of autonomy but does not necessarily result, in the end, in a completely autonomous activity. In our study, we have focused on older adults’ self-directed learning connected to digital tools within the context of PD. To address the issue of the sustainability of these activities, we have co-created a demokit, a set of learning online resources, which we later transformed into a learning room during a public event. Based on our empirical examples from an ethnographically informed and action research oriented study we argue that it is necessary to foster the different facets of self-directed learning that can in turn enable older adults to learn by themselves. These facets include contributing to solving a problem in their own way (Example 1); supporting the ability to identify their own possible failures and supporting them in fixing them (Example 2); choosing to be active or passive in their own actions (Example 3). We argue that to achieve sustainable PD with older adults, we need to foster their self-directed learning by including and supporting their self-directed learning, orienting both towards other people within the design process and to the changing digital (and social) landscape. Using this approach the ongoing changes in the digital landscape are foregrounded. Our findings about self-directed learning of older adults are closely related to the way we decided to present the demokit. These findings led us to a much better understanding of the issues connected to translation of the demokit from its online version to the hybrid one. 5.1 Autonomous Aspects of Learning in PD In the context of PD, the connection between autonomy and learning to use digital tools is key. Taken seriously, PD puts emphasis on fostering the participants’ ability to formulate their own choices [32, 33]. The older participants’ learning autonomy should be enabled by allowing the people to make decisions regardless of the support they might need when taking action based on these decisions [24, 55]. Autonomy is gained through a different, and developing, configuration of actors (including carers and technology) [10, 56]. For this to happen, however, we need to recognise that autonomy is a process, not an outcome. And a very careful examination of that process, an examination that includes what the precise role of researchers is to be, and in what situations. Not least, this involves some consideration of the degree of control over design decisions with the older participants [3, 57]. From this landscape, a complex set of demands emerges for both the participants and the researchers. We contribute to this literature by providing examples which illustrate that selfdirected learning of older adults becomes possible through connecting to other people within the context of PD, which in turn makes the process and its outcomes sustainable.

Designing for Self-directed Learning

17

We argue that to learn in an autonomous way depends substantially on the situation at hand. Autonomy of the learners is hence something to be achieved over a period of time rather than received at one moment through one particular digital artifact. Our examples illustrate that older adults’ learning in a self-directed way might include contributing to a collaborative task in different ways (Example 1). Further, it may mean to make mistakes and choose to fix them with support when realizing the consequences (Example 2). Finally, it might also mean making decisions about the point at which one might engage and the support that one might be open to (Example 3). Self-directed learning is always relative - it will depend on the different configurations that form the learning situations. The older participants will become capable of doing different things by themselves at different times.. That might mean that older adults will be sometimes able to organize their learning in the way that makes sense to them more or less by themselves; and sometimes they will not. Different time scales will also play a role: sometimes their claiming autonomy short term might impact the PD and vice versa. If the cat got actually broken (as a consequence of handling the participant), we would not be able to use it further in the project. In the end, the situation turned out to be a compromise between what the researcher deemed as an acceptable approach to the artifact and what the older adults wanted to do with it. The fostering of autonomy of the older adults then also impacts on the activities in the PD process. In one of the above described co-design sessions we abandoned the topic we had planned for that particular session, because the older adults wanted to talk about different topics related to digital literacy than our co-creation. We chose to give up our plans to support the decisions and actions of the older adults. [58, 60] came to similar conclusions, but we emphasize rather more the idea that these changes of topic represent opportunities which only happen when the older adults are least beginning to appropriate their (autonomous) rights to determine aspects of their own learning. This speaks to the problem of ‘goals’ in a PD project (and beyond) and how they are to be decided. While our work certainly entailed some broad goals that we had determined, largely in accordance with the demands of funding, the smaller goals and the means to achieve them were always up for discussion. Part of the negotiation involved entailed giving up thinking about “the participants” and “our PD project”.The ownership of the project, we would argue, involves constant shifting of perspectives. At the risk of pretension, it entails a ‘continuous revolution’. 5.2 Collaboratively Directing Learning Through PD As with any other types of learning, self-directed learning is grounded in social interactions with others, either in situ or mediated through digital artifacts [23]. As the name suggests, who is responsible for driving the learning activities, and how, makes it different from other types of learning. However, just because the responsibility is on the learners that does not mean they have to make the decisions concerning their learning in isolation from others [55]. Like other types of learning, even here it is important to emphasize the collaborative character of learning [59]. Even including older adults into the co-creation process, it has to be said, does not lead automatically to them gaining the potential benefits that the researchers expect that they will gain [60].

18

K. Cerna et al.

We contribute to this literature by showing empirical examples of how older adults can learn when their activities can become co-directed instead of self-directed in a collaborative manner through the process of PD - either in collaboration with their peers (Example 1 and 3) or with the researcher (Example 2 and 3). An important lesson for us was that what is meant “to become digitally literate” within the context of the project was collaboratively achieved. That is to ‘learn’ under the umbrella of digital literacy is a broad area. Despite the fact that we had a broad understanding of it, what the content of the demokit became did not depend solely on us, but rather on the collaboration among us and the older participants. This approach was further facilitated by the scaffolding approach, where we tried building scaffolding frames which older adults could fill in and hence use them for developing their self-directed learning. We discuss this approach in more detail elsewhere [23].

6 Conclusion Self-directed learning of older adults is a complex problem - especially when it comes to reaching some level of autonomy by older adults in the context of digital technology, this is quite a complicated matter: for every older person it might mean something else, different priorities and abilities, will produce different configurations, which in turn change the living situation (impacted by for examples changes connected to aging). In light of this discussion, when this is brought into the PD, it produces further more complicated relationships which need to be addressed. Self-directed learning is one of the key learning offers suitable for older adults. However, we still lack understanding how to position these learning processes into sustainable PD. In our study, we have aimed to address the following research questions: How can we understand self-directed learning of older adults when older adults and younger researchers collaborate within participatory design? and How do older adults overcome the challenges connected to self-directed learning in a learning space? Our study illustrates the different facets of older adults’ self-directed learning which builds on the mutual interplay between the older participants’ self-directed learning and the (not only) researchers’ support to overcome the connected challenges with the help of the co-developed tool. More specifically, we can understand the self-directed learning and the challenges connected to it as contributing to solving a problem in their own way (Example 1); supporting the ability to identify their own possible failures and supporting them in fixing them (Example 2); choosing to be active or passive in their own actions (Example 3). The questions of “learnability” are an inseparable part of the design of digital tools. Therefore, it is important that HCI deals more with this in the context of artifact design and learning artifact design. Long-term PD projects offer very intensive opportunities for these processes if organized accordingly.

7 Limitations Though we believe our study has helped us understand self-directed learning of older adults within the context of participatory design, there are some limitations which need to be taken into account. First, the final evaluation took place only once during the writing

Designing for Self-directed Learning

19

of this report. Though we used an iterative approach and evaluated different steps with our older participants, we need to further understand how our demokit could be useful for self-directed learning of older adults. Second, it is the issue of sustainability of use of such a tool. Though we developed the demokit in a close collaboration with our older participants, it remains to be seen if they will keep on using it without the researchers’ initiative. Though we have seen potential of the tool to be able to support self-directed learning of the older participants during the evaluation, it is not clear if they will also choose to use it outside of the research context.

References 1. Nauta, L.W.: Historical roots of the concept of autonomy in western philosophy. Prax. Int. 4(4), 363–377 (1984) 2. Kisner, M.J.: Spinoza on Human Freedom: Reason, Autonomy and the Good Life. Cambridge University Press, Cambridge (2011) 3. Maslow, A.H.: A theory of human motivation. Psychol. Rev. 50(4), 370 (1943) 4. Vegliò, S.: The Urban Enigma: Time, Autonomy, and Postcolonial Transformations in Latin America. Rowman & Littlefield (2020) 5. Starr, A., Adams, J.: Anti-globalization: the global fight for local autonomy. New Polit. Sci. 25(1), 19–42 (2003) 6. Weisstub, D.N., Pintos, G.D.: Autonomy and Human Rights in Health Care: An International Perspective. Springer, Dordrecht (2007). https://doi.org/10.1007/978-1-4020-5841-7 7. Tranter, K.: Stories of human autonomy, law, and technology. Bull. Sci. Technol. Soc. 30(1), 18–21 (2010) 8. Sankaran, S., Zhang, C., Gutierrez Lopez, M., Väänänen, K.: Respecting human autonomy through human-centered AI. In: Proceedings of the 11th Nordic Conference on HumanComputer Interaction: Shaping Experiences, Shaping Society, Tallinn, Estonia, October 2020, pp. 1–3 (2020) 9. Nikolaidis, S., Hsu, D., Zhu, Y.X., Srinivasa, S.: Human-robot mutual adaptation in shared autonomy. Proc ACM SIGCHI 2017, 294–302 (2017) 10. Bratteteig, T., Saplacan, D., Soma, R., Oskarsen, J.S.: Strengthening human autonomy in the era of autonomous technology. In: Proceedings of the 11th Nordic Conference on HumanComputer Interaction: Shaping Experiences, Shaping Society (2020). https://doi.org/10.1145/ 3419249.3420097 11. Davy, L.: Philosophical inclusive design: intellectual disability and the limits of individual autonomy in moral and political theory. Hypatia 30(1), 132–148 (2015) 12. Ells, C.: Lessons about autonomy from the experience of disability. Soc. Theory Pract. 27(4), 599–615 (2001) 13. Ho, A.: The individualist model of autonomy and the challenge of disability. J. Bioeth. Inq. 5(2–3), 193–207 (2008) 14. Bonefeld, W., Psychopedis, K.: Human dignity: social autonomy and the critique of capitalism. Hum. Dignity. https://doi.org/10.4324/9781315253046-1/human-dignity-social-aut onomy-critique-capitalism-werner-bonefeld-kosmas-psychopedis 15. Trachtenberg, Z.: Human engineering and the value of autonomy. Ethics Policy Environ. 15(2), 244–247 (2012) 16. Coghlan, S., Waycott, J., Lazar, A., Neves, B.B.: Dignity, autonomy, and style of company: dimensions older adults consider for robot companions. Proc. ACM Hum. Comput. Interact. 5 (2021). https://doi.org/10.1145/3449178

20

K. Cerna et al.

17. Findsen, B., Formosa, M.: Lifelong Learning in Later Life A Handbook on Older Adult Learning. Sense Publishers (2011) 18. Caffarella, R.S.: Self-directed learning. New Directions for Adult and Continuing Education (1993). https://eric.ed.gov/?id=EJ458844. Accessed 07 Feb 2023 19. Pirhonen, J., Lolich, L., Tuominen, K., Jolanki, O., Timonen, V.: ‘These devices have not been made for older people’s needs’–older adults’ perceptions of digital technologies in Finland and Ireland. Technol. Soc. 62, 101287 (2020) 20. DiSalvo, B., Bonsignore, E., DiSalvo, C.: Participatory Design for Learning: Perspectives from Practice and Research. Routledge (2017) 21. Fischer, B., et al.: Co-design as learning: the differences of learning when involving older people in digitalization in four countries. Societies 11(2), 66 (2021) 22. Cerna, K., Müller, C.: Making online participatory design work: understanding the digital ecologies of older adults. In: Proceedings of 19th European Conference on ComputerSupported Cooperative Work (2021). https://dl.eusset.eu/handle/20.500.12015/4161 23. Cerna, K., Müller, C., Randall, D., Hunker, M.: Situated scaffolding for sustainable participatory design: learning online with older adults. Proc. ACM Hum.-Comput. Interact. 6(GROUP), 1–25 (2022) 24. Cerna, K., Müller, C.: Fostering digital literacy through a mobile demo-kit development: co-designing didactic prototypes with older adults. In: Adjunct Publication of the 23rd International Conference on Mobile Human-Computer Interaction, Toulouse & Virtual, France, September 2021, pp. 1–6 (2021) 25. Cerna, K., Paluch, R., Bäumer, F., Ertl, T., Müller, C.: Transformation of HCI co-research with older adults: researchers’ positionality in the COVID-19 pandemic. Interaction Design and Architecture(s) Journal - IxD&A 50, 27–47 (2021) 26. Sackmann, R., Winkler, O.: Technology generations revisited: the internet generation. Gerontechnology 11(4) (2013). https://doi.org/10.4017/gt.2013.11.4.002.00 27. Gregor, P., Newell, A.F., Zajicek, M.: Designing for dynamic diversity: interfaces for older people. In: Proceedings of the fifth international ACM conference on Assistive technologies, Edinburgh, Scotland, July 2002, pp. 151–156 (2002) 28. Loos, E.F.: Senior citizens: digital immigrants in their own country? Obs. (OBS*), vol. 6, pp. 1–23 (2012) 29. Meurer, J., Müller, C., Simone, C., Wagner, I., Wulf, V.: Designing for sustainability: key issues of ICT projects for ageing at home. Comput. Support. Cooperat. Work (CSCW) 27(3– 6), 495–537 (2018). https://doi.org/10.1007/s10606-018-9317-1 30. Hill, R., Betts, L.R., Gardner, S.E.: Older adults’ experiences and perceptions of digital technology: (Dis)empowerment, wellbeing, and inclusion. Comput. Hum. Behav. 48, 415–423 (2015) 31. Hawthorn, D.: Interface design and engagement with older people. Behav. Inf. Technol. 26(4), 333–341 (2007) 32. Deutsches Institut für Erwachsenenbildung. Selbstgesteuerte Lernprozesse älterer Erwachsener: im Kontext wissenschaftlicher Weiterbildung. Bertelsmann (2000) 33. Joshi, S.G., Bratteteig, T.: Designing for prolonged mastery. On involving old people in Participatory Design. Scandinavian J. Inf. Syst. (2016). https://www.semanticscholar.org/ paper/d5dd476c28cc8e2e06ed870d3d400c9e94bc23b0. Accessed 11 Oct 2019 34. Kensing, F., Blomberg, J.: Participatory design: issues and concerns. Comput. Support. Coop. Work 7(3), 167–185 (1998) 35. Bratteteig, T., Wagner, I.: Unpacking the notion of participation in participatory design. Comput. Support. Cooper. Work (CSCW) 25(6), 425–475 (2016). https://doi.org/10.1007/s10606016-9259-4 36. Simonsen, J., Robertson, T. (eds.): Routledge International Handbookof Participatory Design. Routledge (2013)

Designing for Self-directed Learning

21

37. Carmien, S.P., Fischer, G.: Design, adoption, and assessment of a socio-technical environment supporting independence for persons with cognitive disabilities. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Florence, Italy, 2008, pp. 597–606 (2008) 38. Cerna, K., Lundin, J., Islind, A.S., Steineck, G.: Supporting appropriation of self-monitoring tools in clinical settings: the case of pain in cancer rehabilitation. In: Proceedings of the 17th European Conference on Computer-Supported Cooperative Work-Demos and Posters (2019). https://doi.org/10.18420/ecscw2019_p01 39. Scheepmaker, L., Kender, K., Frauenberger, C., Fitzpatrick, G.: Leaving the field: designing a socio-material toolkit for teachers to continue to design technology with children. In: Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, Yokohama, Japan, May 2021, pp. 1–14 (2021) 40. Güldenpfennig, F.: Tailor-made accessible computers: an interactive toolkit for iterative codesign. In: Proceedings of the Twelfth International Conference on Tangible, Embedded, and Embodied Interaction, Stockholm, Sweden, March 2018, pp. 231–236 (2018) 41. Rogers, Y., Paay, J., Brereton, M., Vaisutis, K.L., Marsden, G., Vetere, F.: Never too old: engaging retired people inventing the future with MaKey MaKey. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Toronto, Ontario, Canada, April 2014, pp. 3913–3922 (2014) 42. Berger, A., Ambe, A.H., Soro, A., De Roeck, D., Brereton, M.: The stories people tell about the home through IoT toolkits. In: Proceedings of the 2019 on Designing Interactive Systems Conference, San Diego, CA, USA, June 2019, pp. 7–19 (2019) 43. Jelen, B., Freeman, A., Narayanan, M., Sanders, K.M., Clawson, J., Siek, K.A.: Craftec: engaging older adults in making through a craft-based toolkit system. In: Proceedings of the Thirteenth International Conference on Tangible, Embedded, and Embodied Interaction, Tempe, Arizona, USA, March 2019, pp. 577–587 (2019) 44. Ambe, A.H., Brereton, M., Soro, A., Chai, M.Z., Buys, L., Roe, P.: Older people inventing their personal internet of things with the IoT un-kit experience. In: Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems, New York, NY, USA, pp. 1–15. Association for Computing Machinery, (2019) 45. Cercone, K.: Characteristics of adult learners with implications for online learning design. AACE Rev. (formerly AACE J.) 16(2), 137–159 (2008) 46. Brookfield and Stephen. Understanding and Facilitating Adult Learning: A Comprehensive Analysis of Principles and Effective Practices. McGraw-Hill Education (UK) (1986) 47. Merriam, S.B.: Andragogy and self-directed learning: Pillars of adult learning theory. New Dir. Adult contin. Educ. 2001(89), 3 (2001) 48. Garrison, D.R.: Self-directed learning: toward a comprehensive model. Adult Educ. Q. 48(1), 18–33 (1997) 49. Roberson, D.N., Merriam, S.B.: The self-directed learning process of older, rural adults. Adult Educ. Q. 55(4), 269–287 (2005) 50. Morrison, D., McCutheon, J.: Empowering older adults’ informal, self-directed learning: harnessing the potential of online personal learning networks. Res. Pract. Technol. Enhanc. Learn. 14(1), 1–16 (2019). https://doi.org/10.1186/s41039-019-0104-5 51. Roberson, D.N.: Leisure and learning: an investigation of older adults and self-directed learning. Leisure/Loisir 29(2), 203–237 (2005) 52. Jahnke, I.: Digital didactical designs: Teaching and learning in CrossActionSpaces. Routledge 2015 53. Faas, T., Lin, C.: Self-directed learning in teacher-lead minecraft classrooms. In: Proceedings of the 2017 CHI Conference Extended Abstracts on Human Factors in Computing Systems, Denver, Colorado, USA, May 2017, pp. 2569–2575 (2017)

22

K. Cerna et al.

54. Redstrom, J.: Making Design Theory. MIT Press, Cambridge (2017) 55. Güldenpfennig, F., Mayer, P., Panek, P., Fitzpatrick, G.: An autonomy-perspective on the design of assistive technology: experiences of people with multiple sclerosis. https://doi.org/ 10.1145/3290607.3300357 56. Carmien, S., Dawe, M., Fischer, G., Gorman, A., Kintsch, A., Sullivan, J.F.: Socio-technical environments supporting people with cognitive disabilities using public transportation. ACM Trans. Comput.-Hum. Interact. 12(2), 233–262 (2005) 57. Bischof, A., Jarke, J.: Configuring the older adult: How age and ageing are re-configured in gerontechnology design. In: Socio-gerontechnology, pp. 197–212, Routledge (2021) 58. Waycott, J., Vines, J.: Designing technologies with older adults: ethical tensions and opportunities. In: Neves, B.B., Vetere, F. (eds.) Ageing and Digital Technology: Designing and Evaluating Emerging Technologies for Older Adults,. Singapore: Springer Singapore, 2019, pp. 173–187 59. Müller, C., Kasper, H., Pelzelmayer, K., van Holten, K., Struzek, D., Dickel, M.: Designing for sustainable caring communities – the CareComLabs framework (2019). https://doi.org/ 10.18420/ECSCW2019_P09 60. Knowles, B., Bull, C.N., Davies, N., Simm, W., Bates, O., Hayes, N.: Examining interdependencies and constraints in co-creation. In: Proceedings of the 2019 on Designing Interactive Systems Conference, San Diego, CA, USA, Jun. 2019, pp. 291–302 (2019)

Text Readability of Smartphone in Dark Mode: Effects of Font Type, Font Weight and Color Junxiang Chen1 , Pengyu Zhu2 , Jing Yun3 , Baixuan Tian4 , Yalan Yang3 , and Na Liu1(B) 1 School of Economics and Management, Beijing University of Posts and Telecommunications,

Beijing, China {chenjunxiang,liuna18}@bupt.edu.cn 2 School of Digital Media and Design Arts, Beijing University of Posts and Telecommunications, Beijing, China 3 School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China 4 International School, Beijing University of Posts and Telecommunications, Beijing, China

Abstract. This study investigates the effects of font type, font weight, background color and font color on the readability of Chinese characters in dark mode of smartphone. Forty subjects participated in the experiment and completed the reading comprehension task. Readability was measured by reading time, accuracy rate and subjective ratings. The results show that font type, background color and font color have significant effects on readability, while the effect of font weight is not significant. The combination of Hei font type, 1:12 font weight, dark gray (H: 0, S: 50, B: 12) background color, and light gray (H: 0, S: 0, B: 58) font color results in optimal readability. The research results provide useful information for font design of Chinese characters in dark mode for smartphone. Keywords: Dark Mode · Font Type · Font Weight · Background Color · Font Color

1 Introduction Recently the use of “dark mode” in intelligent terminals is increasingly common. The dark mode serves as an alternative to the default light mode and reduces the brightness emitted by the device’s screen. Adjusting the brightness according to the current lighting conditions can improve users’ experience in the dark environment, reduce visual fatigue, save battery power, and enhance the aesthetics of the interface [1]. There are controversial statements regarding the use of dark mode on smartphones. Some argue that the use of dark mode on smartphones is beneficial in low light environments as it “saves electricity” and “protects eyes”, as the lower brightness can make the eyes look at the screen more comfortable, reduce the irritation to the eyes and save electricity. Others believe that the dark screen negatively affects text reading and can exacerbate visual fatigue. This has led to questioning the effectiveness of dark mode in © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 23–33, 2023. https://doi.org/10.1007/978-3-031-34866-2_2

24

J. Chen et al.

“eye protection”. Despite these concerns, research has shown that there is no significant difference between dark and light modes in their impact on visual fatigue [1]. In fact, the dark mode does not only change the default mode of “black characters on a white background” into “white characters on a black background”, but also needs to further design the parameters such as color saturation and contrast according to the visual and cognitive characteristics of the background and text. At present, there are different dark modes in different applications, and inappropriate design might have a negative impact on users’ text reading and understanding. Thus, it is necessary to examine the readability of dark mode on smartphones. Previous studies suggest that people interpreted the reading material differently in dim and bright light. Lighting has a significant effect on the accuracy of participants’ answers to reading questions, and participants have a higher accuracy in understanding materials in an environment with strong lighting than in a dark environment [2]. For Chinese characters, Hei, Kai and Song are widely used to display text settings. Hei is the sans-serif body with the same stroke width and simple shape, while Kai and Song are the serif font with inconsistent stroke width. The Kai has more changes in stroke shape than the Song. Among the three types of fonts, the Kai is the most complex, followed by the Song, and the Hei is the most concise. Studies have shown that the simpler the structure of the Chinese font, the easier it is to read [3]. However, the actual sizes of the three fonts are different when the font size is the same. In the early Chinese books, em grid was used as the basic unit of Chinese font, which stipulated the design requirements of Chinese characters as blocks. The grid for designing Chinese characters was divided into “outer frame” (usually 1000 * 1000 upm) means character body frame and “inner frame” means literal frame, and the distance between the two was used to define the original spacing among Chinese characters. Due to different width and height of the text itself, the outer frame of the case of the same benchmark, the inner box will be different. It results in different characters of the same font or different fonts with different literal rate. The most intuitive difference is reflected in the visual size difference of different fonts under the same size. Research shows that the literal rate difference is larger in Hei than in Song, and larger in Song than in Kai, so the actual size of Hei is the largest and easier to read under the same font size [4]. Font weight is usually expressed as the ratio of stroke weight to character height. The smaller the font weight, the finer the letter. Some fonts have the same thickness, such as Hei, and some fonts have strokes of different thickness, such as Song. Common font weights are 1:8, 1:10, and 1:12. For English fonts, some researchers believe that under good lighting, the ratio of 1:6 to 1:8 is suitable for black letters on white background [4]. When illumination is reduced, thick letters are easier to read than thin letters, the clearest combination is stroke thickness to height ratio of 1:10[5]. In contrast to Latin letters, Chinese characters have more strokes and richer structure, prior research results may not applicable. Some researchers believe that the font weight of 1:18, 1:10 and 1:12 has no significant influence on article reading, while the interaction between font weight, font size and font complexity has a significant influence on readability [6]. This experiment was conducted in light mode with traditional white background and black background, while our experiment focused on dark mode. Suggestions on the optimal

Text Readability of Smartphone in Dark Mode

25

word weight for light mode cannot be directly applied to dark mode. Therefore, it is necessary to better understand the influence of word weight on readability in dark mode. To sum up, for Chinese text reading, without controlling the distance between the reader and the mobile phone screen, adjusting the viewing distance can offset the influence of font size on reading time, while reading text with large font requires more sliding and increases the reading time. It concludes that small font provides better reading performance than large font when reading text on a smartphone; fonts have no significant effect on reading time, but people’s preferences vary, with more popular fonts being Kaior Hei [7]. Therefore, the combination of 12pt font size and 1:10 stroke width can provide a better reading experience in the case of different complexity of Chinese characters, which is the recommended choice for VDT character design. Background color and font color are considered to be two important factors affecting font readability. In light mode, the background color is light and the font color is dark, that is, the text is dark under the light background. In the dark mode, the background color is dark and the font color is light, that is, the light text is presented on the dark background. Common background colors include black (H: 0, S: 0, B: 0) and dark gray (H: 0, S: 50, B:12), and common font colors include pure white (H: 0, S: 0, B: 100), light gray (H: 0, S: 0, B: 58). In night eye protection mode, both color pattern and brightness contrast (a measure of the brightness difference between the viewing object and its adjacent background in the display device’s field of view) had significant effects on visual fatigue and subjective preference. Dark mode (showing light text against a dark background) is better for reducing visual fatigue. The subjects showed a higher preference for light mode, in which dark text is presented on a light background, which may be related to their usage habits. Higher brightness contrast helps reduce visual fatigue, and people prefer higher brightness contrast [8]. High contrast has better visual recognition performance than low contrast. Visual performance improves with the increase of contrast. Contrast has more effect on the visual performance of TFT-LCD screens than text color. If the contrast is at an acceptable level, font color will not affect visual performance [9]. Based on aforementioned research results, we can basically understand the influences of font size, font weight, ambient illumination, contrast and color on font readability. At the same time, it is found that most studies on the readability of Chinese text on intelligent terminals focus on the effects of font type, font weight, line spacing and other factors in “light mode”, and the researches of the effects of dark mode on Chinese text readability were limited. Therefore, it is necessary to explore the readability of Chinese text in the “dark mode” of smartphones. In summary, this paper aims to explore the readability of Chinese characters in “dark mode” of smart phone and investigated the effects of font type, font weight, text color and background color on readability. The results will help provide suggestions and guidelines for the design and improvement of the information display and dark mode of smartphones.

26

J. Chen et al.

2 Method 2.1 Participants A total of 40 participants were recruited to participate in the experiment. All participants were undergraduate or graduate students and native Chinese speakers(20 males and 20 females, aged 18 to 25, the mean age = 20.93, SD = 1.38). At the same time, the visual acuity was reported to be normal with no clinically significant abnormal adjustment, and the corrected visual acuity could meet the standard of 5.0 (measured by logarithmic visual acuity chart). 2.2 Design Within-subject design was adopted in this experimental. The independent factors were font type, font weight, background color and font color. The font type had two levels namely Song and Hei, and the font weight had two levels namely 1:8 and 1:10. The background color had two levels, which are black (H: 0, S: 0, B: 0) and dark gray (H: 0, S: 50, B:12). Font color had two levels, which were pure white (H: 0, S: 0, B:100) and light gray (H: 0, S: 0, B: 58). All the independent variables produce a total of 16 combinations. The experimental dependent factors included task performance and subjective ratings. Task performance was measured by completion time and accuracy, and, subjective ratings included understandability, comfort and satisfaction, which were all measured by 7-point Likert scale. The experimental task is reading comprehension on smartphones. The experimental articles required for reading comprehension will be Hanyu Shuiping Kaoshi (HSK) test articles and questions. HSK is an international standardized Chinese proficiency test set up to test the Chinese proficiency of non-native Chinese speakers (including foreigners, overseas Chinese, Chinese Americans and Chinese minority examinees). In this experiment, a total of 16 HSK CET-6 articles were selected. The difficulty and topic of each article was similar, and the number of words in each article was controlled between 500–700. Articles were presented on the web page of smartphones. Each factor combination was corresponding to two articles. 2.3 Procedure Before the experiment began, the experimental software was installed on participants’ own smartphones. The participants were informed of the purpose of the experiment and were provided with instructions on how to complete one experimental trial. The participants signed informed consent forms. Then, the experiment started after participants tapped the “start” button on their smartphone. Once they started, participants were presented with an article page. Participants scrolled the page to read each article. Participants were asked to complete four multiple choice questions after finishing reading an article. When participants finished answering comprehension questions, they tapped the “Next” button at the bottom and filled in the subjective rating scale. The procedure was repeated after participants completed 16 reading comprehension tasks. To avoid visual fatigue, participants took a five-minute

Text Readability of Smartphone in Dark Mode

27

break after completing eight passages. After completing the experiment, the participants received rewards.

3 Results 3.1 Reading Time Table 1 shows the ANOVA results for reading time, and Table 2 shows the mean and standard deviation of reading time on all combinations of font type, font weight, background color, and font color. The results showed the interaction of font type, font weight, background color and font color had significant effects on reading time (F (1,624) = 4.038, p = 0.045). Among the 16 variable combinations, the group with font type in Hei, font weight of 1:12, background color of dark gray and font color of light gray had the shortest reading time (mean = 99.6, SD = 71.956). The second group is font type of Song, font weight of 1:10, background color of dark gray, font color of light gray (mean = 104.38, SD = 44.785). The longest reading time was in the group with font type of Song, font weight of 1:12, background color of black and font color of pure white (mean = 140.82, SD = 127.633). All other effects and interaction effects were not significant as shown in Table 1. Table 1. Analysis of variance for the reading time Source of Variation

SS

df

MS

F

P-Value

font type

17284.806

1

17284.806

2.931

0.087

font weight

6464.306

1

6464.306

1.096

0.296

background color

15840.4

1

15840.4

2.686

0.102

font color

726.756

1

726.756

0.123

0.726

font type * font weight

4895.156

1

4895.156

0.83

0.363

font type * background color

783.225

1

783.225

0.133

0.716

font type * font color

1995.156

1

1995.156

0.338

0.561

font weight * background color

756.9

1

756.9

0.128

0.72

font weight * font color

15.006

1

15.006

0.003

0.96

background color * font color

3404.025

1

3404.025

0.577

0.448

font type * font weight * background color

9211.225

1

9211.225

1.562

0.212

font type * font weight * font color

907.256

1

907.256

0.154

0.695

font type * background color * font color

1863.225

1

1863.225

0.316

0.574

font weight * background color * font color

1664.1

1

1664.1

0.282

0.595

font type * font weight * background color * font color

23814.4

1

23814.4

4.038

0.045

3.2 Subjective Rating As can be seen from Table 3, font type has a significant impact on subjective rating (F (1, 624) = 37.49, p < 0.05), and font color has a significant impact on subjective rating (F (1, 624) = 10.223, p = 0.001). The subjective rating of Hei (mean = 15.87, SD = 3.24)

28

J. Chen et al.

Table 2. Means and standard deviations of reading time for each level of independent variables font type

font weight

background color

Hei

1:10

black

font color

dark gray

1:12

black

dark gray

Song

1:10

black

dark gray

1:12

black

dark gray

Mean

SD

pure white

114.25

67.882

light gray

116.28

66.354

pure white

110.9

71.337

light gray

114.82

72.306

pure white

113.17

64.864

light gray

138.53

139.035

pure white

108.25

57.294

light gray

99.6

71.856

pure white

116.85

42.989

light gray

134.15

86.105

pure white

120.33

56.379

light gray

104.38

44.785

pure white

140.82

127.633

light gray

123.13

65.594

pure white

124.28

68.798

light gray

135.03

57.184

was higher than that of Song (mean = 14.4, SD = 2.883). In both font color levels, the subjective rating was higher for light gray (mean = 15.52, SD = 3.055) than for pure white (mean = 14.75, SD = 3.205). All other effects and interaction effects were not significant as shown in Table 3. Table 3. Analysis of variance for the subjective rating Source of Variation

SS

df

MS

F

font type

349.577

1

349.577

37.49

P-Value 0

font weight

0.189

1

0.189

0.02

0.887

background color

1.702

1

1.702

0.182

0.669

font color

95.327

1

95.327

10.223

0.001

font type * font weight

0.127

1

0.127

0.014

0.907

font type * background color

0.827

1

0.827

0.089

0.766

font type * font color

12.377

1

12.377

1.327

0.25

font weight * background color

2.889

1

2.889

0.31

0.578

font weight * font color

31.952

1

31.952

3.427

0.065

background color * font color

0.827

1

0.827

0.089

0.766

font type * font weight * background color

0.827

1

0.827

0.089

0.766

font type * font weight * font color

30.189

1

30.189

3.238

0.072

font type * background color * font color

1.502

1

1.502

0.161

0.688

font weight * background color * font color

1.702

1

1.702

0.182

0.669

font type * font weight * background color * font color

0.689

1

0.689

0.074

0.786

Text Readability of Smartphone in Dark Mode

29

Table 4. Means and standard deviations of subjective rating for each level of independent variables font type

font weight

background color

font color

Mean

SD

Hei

1:10

black

pure white

15.35

3.641

light gray

16.35

2.905

dark gray

1:12

black

dark gray

Song

1:10

black

dark gray

1:12

black

dark gray

pure white

15.4

3.47

light gray

16.53

3.047

pure white

15.2

3.625

light gray

16.25

3.053

pure white

15.45

3.396

light gray

16.48

2.552

pure white

13.8

2.766

light gray

15.18

2.836

pure white

13.63

2.686

light gray

15

2.612

pure white

14.3

3.04

light gray

14.25

3.528

pure white

14.88

2.441

light gray

14.15

2.878

3.3 Accuracy Rate Table 5 shows the ANOVA results for accuracy rate, and Table 6 shows the mean and standard deviation of accuracy rate on all combinations of font type, font weight, background color, and font color. The results showed that background color has significant effect on accuracy rate (F (1,624) = 3.89, p < .05). At both background color levels, the accuracy rate of dark gray group (mean = 90.78, SD = 16.353) was higher than that of black group (mean = 88.05, SD = 18.561). All other effects and interaction effects were not significant as shown in Table 5.

30

J. Chen et al. Table 5. Analysis of variance for the correct rate

Source of Variation

SS

df

MS

F

P-Value

font type

352.539

1

352.539

1.146

0.285

font weight

79.102

1

79.102

0.257

0.612

background color

1196.289

1

1196.289

3.89

0.049

font color

118.164

1

118.164

0.384

0.536

font type * font weight

282.227

1

282.227

0.918

0.338

font type * background color

118.164

1

118.164

0.384

0.536

font type * font color

219.727

1

219.727

0.714

0.398

font weight * background color

8.789

1

8.789

0.029

0.866

font weight * font color

165.039

1

165.039

0.537

0.464

background color * font color

610.352

1

610.352

1.984

0.159

font type * font weight * background color

0.977

1

0.977

0.003

0.955

font type * font weight * font color

118.164

1

118.164

0.384

0.536

font type * background color * font color

219.727

1

219.727

0.714

0.398

font weight * background color * font color

282.227

1

282.227

0.918

0.338

font type * font weight * background color * font color

711.914

1

711.914

2.315

0.129

Table 6. Means and standard deviations of correct rate for each level of independent variables font type

font weight

background color

font color

Mean

SD

Hei

1:10

dlack

pure white

88.75

20.373

light gray

86.87

14.965

pure white

92.5

15.191

light gray

90.63

15.697

dark gray

1:12

black

dark gray

Song

1:10

black

dark gray

1:12

black

dark gray

pure white

86.25

16.937

light gray

85.63

18.68

pure white

91.25

13.337

light gray

87.5

17.903

pure white

84.38

26.365

light gray

93.13

13.853

pure white

93.13

13.853

light gray

88.75

17.858

pure white

90.63

16.687

light gray

88.75

17.858

pure white

91.88

14.308

light gray

90.62

21.697

Text Readability of Smartphone in Dark Mode

31

4 Discussion Previous studies have shown that Song is more complex than Hei, and the simpler Chinese character structure, the better readability on desktop display [3]. Some scholars also said in their research that under the condition of light mode of intelligent terminal, font type has no significant impact on reading time. Compared with Song typeface in terms of subjective rating, the Hei has a higher preference for subjective rating, as the Hei is a sans serif font with the same width of solid strokes and shows minimalist modernism aesthetics. It is more concise than Song typeface and more popular with users [7]. Through this experiment, it can be found that this conclusion is still applicable to the dark mode reading situation in the dark environment. The results show that there is no significant difference in the accuracy and reading time between the Hei and the Song, while the average subjective rating of the Hei is higher, and the subjects give higher evaluation of comfort and satisfaction, which reflects that the subjects prefer the Hei. In previous studies, the recommended font weight of English text and digital text was studied in depth. Under good lighting, 1:6 to 1:8 is satisfactory for printed materials in English text with black letters on a white background [5]. When the brightness is reduced, thick letters are easier to read than thin ones. Digital text was tested and the results showed that the most obvious combinations for all numbers were the aspect ratio of 10:7.5 and the stroke thickness to height ratio of 1:10 [6]. However, in this experiment, the font weight of Chinese text was tested, and the research showed that the font weight had no significant effect on readability. This may be because Chinese characters have a more diverse number of strokes, types, and composition structures than alphanumeric characters. It is also possible that the reading was carried out in dark mode in this experiment environment, and the effect of font weight on the reading efficiency of the subjects was not significant. Previous studies of font weight have been in light-colored patterns, which make it easier to identify different levels of font weight. In the dark mode, different font weight levels were not easy to distinguish, and different font weight levels had no significant effect on the reading time, accuracy rate and subjective rating of the subjects. In order to match the dark mode, dark gray (H:0, S: 50, B:12) and black (H:0, S: 0, B:0) were selected for the level of background color variables. Previous studies of English text have shown that background color in dark mode has no significant effect on reading and subjective preference scores in dark mode under the same brightness contrast conditions [10]. But this research results showed that under the condition of Chinese text, the background color had a significant impact on the accuracy of answers, and the accuracy of dark gray was higher than that of black. It indicates that subjects under the dark gray background color achieve better reading comprehension and high accuracy. We suspect that the divergence of results may be due to the huge differences in complexity and writing style between Chinese and English texts. English text is divided by spaces, making it easier to distinguish the meaning of words. But Chinese text requires the reader to break the sentence himself. In terms of font color variables, font color has no obvious influence on accuracy and reading time, but has significant influence on subjective rating. The subjective rating of light gray (H:0, S:0, B:58) is higher than that of pure white (H:0, S:0, B:100), indicating that under the conditions of this experiment, subjects feel more comfortable with light

32

J. Chen et al.

gray font color than pure white font color, and the feeling of fatigue is lower. Through the communication with the subjects after the experiment, we learned that most of the participants thought that the pure white font was harsher in dark conditions than the light gray font, and the symptoms such as sore eyes when reading might be the reason for the lower subjective rating of the pure white font. Previous Studies have shown that high brightness contrast between font color and background color is preferred by users, higher brightness contrast ratio helps reduce visual fatigue, and people also prefer higher brightness contrast [11]. In dark mode, a brightness contrast of 0.969 resulted in the lowest visual fatigue, while subjects preferred a brightness contrast of 0.868, 0.855, 0.725 when using computer to read in a dark environment [9]. Our experimental results show that in the combination of variables in this experiment, when reading on the phone, the dark gray background color with the light gray font color has a higher subjective rating (brightness contrast is 0.764), which is close to the previous results.In the analysis of the interaction of variables, the interaction of four factors has a significant impact on the reading time, among which the variable combination of Hei font type, 1:12 font weight, dark gray background color and light gray font color has the shortest reading time. In terms of subjective scoring, the light gray font color is better than the white font color, and the Hei is better than the Song; In terms of accuracy, the dark gray background color is better than the black background color. It is concluded that among the 16 groups of variables designed in this experiment, this group of variable combination has the highest readability.

5 Conclusion In this study, the effects of font type, font weight, background color and font color on reading time, accuracy and subjective rating of smartphones have been investigated. The results showed that background color has significant effect on accuracy, dark gray background is better than black background in accuracy; The influence of font type on subjective rating is significant, and the Hei is preferred by people. Font color has a significant effect on subjective rating, and light gray is better than pure white in this experiment. This indicates that font color, font type and background color have obvious effects on the readability of text in dark mode under dark environment, while font weight has little effect. Under the experimental conditions of this study, the combination of Hei font type, 1:12 font weight, dark gray background color and light gray font color is the recommended font design combination. The research results provide suggestions for intelligent terminal design of user-friendly font type, font weight, background color, font color representation, and provide directions for further research. Acknowledgements. This work was supported by the grant from Natural Science Foundation of China (Project No. 71901033).

Text Readability of Smartphone in Dark Mode

33

References 1. Wang, A.H., Chen, M.T.: Effects of polarity and luminance contrast on visual performance and VDT display quality. Int. J. Ind. Ergon. 25(4), 415–421 (2000) 2. Zhou, X., Rau, P.-L.: Effect of illumination on reading performance and affect in a virtual environment. In: Rau, P.-L. (ed.) CCD 2018. LNCS, vol. 10911, pp. 460–471. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-92141-9_36 3. Cai, D., Chi, C.F., You, M.: The legibility threshold of Chinese characters in three-type styles. Int. J. Ind. Ergon. 27(1), 9–17 (2001) 4. Chi, C.F., Cai, D., You, M.: Applying image descriptors to the assessment of legibility in Chinese characters. Ergonomics 46(8), 825–841 (2003) 5. Navships, H.H.J.: Display Illumination Design Guide. Section 2: Human Factors. NAVAL ELECTRONICS LAB CENTER SAN DIEGO CA, 1973 6. Soar, R.S.: Height-width proportion and stroke width in numeral visibility. J. Appl. Psychol. 39(1), 43 (1955) 7. Liu, N., Yu, R., Zhang, Y.: Effects of font size, stroke width, and character complexity on the legibility of Chinese characters. Hum. Fact. Ergon. Manuf. Serv. Ind. 26(3), 381–392 (2016) 8. Huang, S.M.: Effects of font size and font style of Traditional Chinese characters on readability on smartphones. Int. J. Ind. Ergon. 69, 66–72 (2019) 9. Xie, X., Song, F., Liu, Y., et al.: Study on the effects of display color mode and luminance contrast on visual fatigue. IEEE Access 9, 35915–35923 (2021) 10. Hashim, A.A., Majid, M.A.: A study on colour combination guidelines for text and background in web legibility perspective 11. Li, Y., Huang, Y., Li, X., et al.: The influence of brightness combinations and background colour on legibility and subjective preference under negative polarity. Ergonomics 65(8), 1046–1056 (2022)

Human Factors Based New Media Design: Methodology and Assessment Shijun Ge1 , Yuanbo Sun1 , Yin Cui1(B) , and Hao Shen2 1 Beijing University of Technology, Beijing 100081, China

[email protected], [email protected] 2 High School Affiliated to Beijing Normal University, Beijing 100052, China

Abstract. Thanks to the advances in medium technology, recent media products are presented in a new form. The participation of audience becomes an important part of new media product. Therefore, the information input and output from audience is the key part of design. This research integrates theory and methodology of human factors and ergonomics into new media design. Taking the project ‘magic trip to Sanxingdui ruins’ as an example, we detail the human factors in each common step of new media design. This research integrates the system design of ‘man-machine-environment’ into new media design procedure, and proposes methodologies and assessment of human factor based new media design. Keywords: Human factors · New media design · Interaction

1 Introduction Based on the advances in media technologies, new media products become richer and more diverse in their expressions: from vision only to multi-modal, from single communication to interaction, from static to dynamic, from two dimension to multi-dimension, and from reality to virtual reality (Lu 2021). New media products have increased association to creator’s concept and style, and focus communication and value transfer. At the same time, the interaction of audience is also included into the products. Therefore, the boundary of a new media product is blurred, and many scholars see them as a part of art research. However, the major goal of the new media design is to transfer value, pass information, broadcast concept, output aesthetics and art, express emotion, etc. (Wei 2017). The interaction from audience becomes a crucial part of product. Thus, the design concept should be ‘audience centered’, which surpasses ‘artist centered’, so the paradigm of design research shall apply. New media products require audience participation and have a high level of interaction. Not matter how cool the media technology is and how diverse the expression, the essential design is the information input and output (Sun 2010). Both domestic and international scholars work on the interactivity in new media design. For example, new cognitive model for new media product (Wei 2017); research of interactive new media installations (Li 2011); design theory of new media product based on interactive narration (Sun We 2011), etc. Human factors are the foundations of interactive design, but there © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 34–44, 2023. https://doi.org/10.1007/978-3-031-34866-2_3

Human Factors Based New Media Design: Methodology and Assessment

35

is not yet any systematic research on human factors based new media design. Therefore, it is of necessity to study the new media design from the perspective of human factors, establish theory and methodology to guide design practice.

2 New Media Design and Human Factors Media is under constant changes (and has its own era and life cycle), but what is new media design? Current academics tend to understand the new media as ‘digital media centered’. That is to provide information and services via digital interaction using fixed or mobile multimedia terminals. All types of digital information device centered around computer are used to describe and realize design concept. Through digital information technology platform, digital technology based design products are finalized and published via different types of digital medium. This provides the audience a engageable and interactive experience (Liao 2008). Due the rapid development of media technology, new feature and new change from new media design lead to the changes in considering human factors in design process. The final product is no longer the focus, but the process that the audience and product interact with each other. This interaction can sometimes be just a moment. The audience is now a part of product. The designer captures, organizes, optimizes, and presents complex information. In this dynamic process, the designer is not only to keep the information accurate, efficient, and instant, but also to provide best interactive experience for the audience. This then allows a natural, efficient, and elegant interactive process. This is an important part of theoretical study of human factors. 2.1 Cognitive Model A good new media design must look at the fundamental interactions between human and design products from cognitive psychology perspective. From information perspective, the essential of new media design is the designer’s information cognition, process and feedback (Salvendy 2012). Therefore, it is important to at first understand cognitive model of the new media product (see Fig. 1). A New Media Design Work develops sensation via stimulating human’s information receiver (i.e., vision, hearing, smell, taste, and

Fig. 1. Cognitive model of the new media work

36

S. Ge et al.

touch), draws audience attention, mobilizes memory, forms decision, and reacts through body language, head movement, eye turns, blinks, languages, voices, face expressions, etc. The product reacts audience through light, audio, vibration, animation, etc. and produces emerging experience for the audience to perform the next interaction. The process repeats. The design practice shall pay special attention to the following aspects: (1) pay special attention to perception: perception includes sensory development, complex and dynamic interactions between action and environment information. The perception can affect human’s behavior during the process of human and media interaction. Under an environment, human action can also generate perception. Therefore, it is of necessity to understand biological and functional feature of human’s sensory (e.g., vision, hearing, taste, etc.) before design. (2) Attract attention: Human attention is selective, transferable, and sustainable. It is a prerequisite to understand characteristics of audience’s attention in order to design a form of expression that can keep high focus from audience. That is to consider the cognitive features of audience, to fulfill their requirements, and to spark interest. (3) Trigger memory: the signal in memory trigger and the flow between short- and long-term memory is fundamental to a cognitive process. When human receive new, strong, and rich multisensory stimulus, information transforms into a deeper level of memory. During design process, impressive elements can pass information to audience effectively, cognize, and transform into long-term memory. In addition, selecting familiar items, visual elements, sound, gestures, etc. can keep consistent approach for audience’s cognitive pattern, and assist memory weakness and other burden in the cognitive process. In this way, memory is minimized, and operation is more intuitive. In the end, the presence of audience is increased, and cognition efficiency is improved. (4) Guide a decision: In the process of human media interaction, the cognition can impact human decision. For example, people will sit and rest on structures like steps or rocks unconsciously. In cognition theory, this phenomenon shows special object can provide action potential and affect human decision. This feature provides a better interaction between human and environment but is easily transformable due to people’s mood. New media design often use this theory to guide the audience to make unexpected decision. (5) Reasonable feedback: Feedback is the process of return information to its source, so that audience know the result of own action. When audience perform certain operation, certain feedback can be provided by installation and environment. For example, through new picture, sound, etc. The feedback helps audience to understand if the operation is executed or not. Any feedback test should ensure sufficiently sensitive and prompt feedback. In addition, whether the form of feedback is pleased and actively engaged by the audience. In short, new media design already surpassed a single product design, and became Audience-Centered design. Because human factor consideration is throughout entire design process.

Human Factors Based New Media Design: Methodology and Assessment

37

2.2 Human Factors in the New Media Design Process For each new media design phase, questions and works related to human factor is illustrated in Table 1. It demonstrates important human factor work exist in the entire design process. Table 1. Human factors in new media design process Media Design Phases

Question related to Human factor

Theme Planning Phase

Who are the audience? Research, analyze and describe target audience What is the main theme?

Work related to Human factor

Design Phase of a Man-Installation System System Target and Work Requirement

Investigate the level of attention and interest to the theme from the target audience

Human characteristics? Study human factors and safety standard

Design Phase

Environment safety? How to utilize efficiently?

Analyze physical environment, eliminate potential safety issues, consider potential psychological effect from audience, and analyze pros and cons

Can current technology support design implementation and provide good experience?

Investigate application domain and concrete parameters of key technology

Do planned theme and its functionalities appeal to audience?

Propose theme and concrete functional design, perform audience survey, adjust when necessary, and decide the idea

Do the narratives around the theme appeal to audience?

Understand audience’s Initial Design perception, psychological and culture background, and follow their cognitive psychology and usual behavior

System Definition

(continued)

38

S. Ge et al. Table 1. (continued)

Media Design Phases

Question related to Human factor

Work related to Human factor

Design Phase of a Man-Installation System

How to select medium with its design?

Select suitable medium combination, including both physical and non-physical, to efficiently and promptly pass messages

How to perform interactive design?

Design human output and control, installation and environment input, and interactions among ‘human- installation -environment’

Output design

Design human interactions, for example body language, head movement, eye rotation, blink, language, voice, face expression, etc

Input design

Output feedback from design installation and environment, e.g., animation, light effect, sound, vibration, flavor, etc

Does comprehensive design (e.g., vision, hearing, smell, taste, etc.) fit with human factors?

Investigate physiological Ergonomics indicators and characteristics of vision, hearing, smell, taste and other biological indicators. The expression should conform to the aesthetics and cognition of the audience

Is time design reasonable?

The time design should fulfill the time requirements from cognition, action, and feedback (continued)

Human Factors Based New Media Design: Methodology and Assessment

39

Table 1. (continued) Media Design Phases

Test and Iteration

Question related to Human factor

Work related to Human factor

Is space design reasonable?

Interface space, physical space, virtual space are well coordinated. The physical logic and behavior logic should fit the cognition and behavior of the audience

What are testing dimensions?

Determine test requirements, e.g., communication (precise and efficient), interaction (nature and smooth), comfort, etc., and establish evaluation system

What are the testing methodologies and processes?

Based on scientific approach, recruit target audience to evaluate design, collect subjective feedback and objective biological data

Provide improvement ideas based on test results

Based on the experimental results from audience, perform improvement and iteration

Design Phase of a Man-Installation System

System Verification

3 Design Method and Practices China’s ‘Sanxingdui’ ruin is a culture heritage from 4800 to 3100 years ago. To demonstrate archaeological site, display precious culture relics, and promote traditional Chinese culture, China Media Group and Key laboratory of digital performance and simulation technology from Beijing Institute of Technology launched a large immersive digital interactive space “magic trip to Sanxingdui ruins” (see Fig. 2). Through instant cloud rendering technology, the project provides an innovative and immersive experience for scenes like archaeological excavation shed of ‘Sanxingdui’ ruin, ‘Sanxingdui’ digital museum, ancient Shu kingdom, etc.

40

S. Ge et al.

Figures 3 is the new media design procedure based on human factors and ergonomic system. Let us take “magic trip to Sanxingdui ruins” project as an example to elaborate further.

Fig. 2. Magic trip to Sanxingdui ruins

Fig. 3. Human factors based new media design process

3.1 Human Factors in Theme Planning Phase (1) Target audience is the object for receiving connotation and ideological value. It is also the most important factor in the interactive media design. The cognition and behavior from audience influence the expression process and effect. The target audience of “magic trip to Sanxingdui ruins” is the young group that are interested

Human Factors Based New Media Design: Methodology and Assessment

41

in Chinese history and culture relics. Through deep interview with target audience and observing focus group, audience’s perception and expectation are collected. Lifestyle, social form, environmental form of the audience is described, and precise persona can be depicted. Questionnaire-based quantitative approach can be used to collect user preferences, include visual style, presentation form, data content, etc. Moreover, based on the understanding to the audience, appropriate build, graphics, sound effect, and specific interactions are used to guide subsequent design. This is to improve user experience in media design. (2) Theme design can be determined by audience research and design requirement. For example, transfer knowledge, share information, spread idea, output aesthetic and artistic taste, express emotion, etc. They bring emotions, imagination, discovery, reflection, achievement, and other experiences to audience. The goal for “magic trip to Sanxingdui ruins” is to engage audience in archaeological site, enjoy cultural relics, present associated cultural relics through the themed cultural relics. The goal of spreading knowledge and promoting Chinese tradition is then achieved. In the theme design phase, questionnaire is used to understand audience’s interests, satisfaction of the form of interactions, satisfaction of interaction setup, whether the expression effect is acceptable, and what’s audience’s expectations. Through the research from audience, the design goals can be refined and detailed further. (3) Clarify design limitations. Human factors are mainly about the limitations from the target audience. For example, the limitations from human perception, cognitive ability, human mobility, biomechanical ability. In addition, there are environment limitations posed on humans. For example, the influence and limitations from environment, the restrictions on control and display design, etc. The project is mainly displayed through TV and digital screen. Therefore the main limitations are the feature of the transmission medium. (4) Technology confirmation. Good technologies can realize the design in a better way. Through real-time cloud rendering technology, the project provides a new immersive experience without local hardware limitations. This gives a low-cost immersive interaction, which communicates with ‘out of screen’. In addition, holographic display and light mapping technologies are used. 3.2 Human Factors Based Design Process After determining the theme with concrete design goals, it goes to scheme design stage. This design process prioritizes the narrative logic from beginning to end, the choice of appropriate media, and the construction of time and space scene. The goal is to design an integrated system with “Man-Machine-Environment”, in which the sense of audience is fully mobilized and attracted to the interactive ‘Audience- Installation -Environment’ system. Matterology. Even with brilliant interactive form seeing from outside, new media product cannot produce emotional resonance and form conceptual value with a lack of integrity, relevance, cultural resonance, etc. Therefore, during design the audience need to be guided in their cognitive process, like information selection. Integrative and expressive construction require certain design matterology (Guanzhong Liu, 2006). This

42

S. Ge et al.

is to connect individual design elements, arrange the way and order in which design elements are presented. ‘The magic trip to Sanxingdui ruins’ takes a story telling approach. The narratives are presented through text, language, images, etc., and the events within specific time and space are presented. This project allows audience to participant directly as a part of story. After some questions, digital persons are created and entered system, and three scenes are constructed: archaeological excavation shed of ‘Sanxingdui’ ruin, ‘Sanxingdui’ digital museum and ancient Shu kingdom. The tour scene is realized in a virtual word, in which ‘digital guide’ is created to improve experience and connect fragmented knowledge to cultural relics. Choice of Medium. Medium carries the content and determines how the content is presented. Medium itself is also an information. In the era of new media, every new media type shall trigger the evolution of design and art. All kinds of new medium enrich designer’s creative techniques and ways of expression. The project takes virtual interaction, creation of digital asset, real-time rendering by virtual engine, video communication technology, and other technologies. The audience experience through TV and digital screen served as medium. Interactive Design. As the audience and participants in media design, human behavior and language reflect the complete style of the product throughout the media product expression. The interactions in the media design product include designer with product, audience with product, audience with audience, and designer with audience. It also covers the interactions between product with space environment, product with product. It can simply be categorized as input problem and output problem. Under input problem, human receive information from external (machine, human or environment). Whereas under output problem, human produce information and control output. Humans obtain external information through their sensory system, and provide feedback as output through body language, head movement, eye turns, etc., It repeats this way to complete information interaction. ‘The magic trip to Sanxingdui ruins’ captures visual and auditory information from digital screens, provides feedback through actions, and links via multimodal interactions. The invisible audience in front of the TV becomes visible audience under live broadcast in digital space. At the same time, the live broadcast becomes bidirectional interactive communication, achieving ‘thousand people thousand faces’. Different network login provides different content experience, therefore enters the era of complete interactive communication. The interactive layer is the concrete representation of narrative layer. Based on the interactive approach defined by user cognitive tasks and goals, interactive action, element and feedback are clearly designed. When the audience interacts with the outside world, the design should ensure the feedback is timely and in line with user expectations. The design should have a certain degree of standardization and consistency. This is to ensure that the audience have a smooth experience in complex multi-sensory, immersive, multi-dimensional (time and space), and multi-experienced interactions. Overall Design. New media design includes multiple media applications with complex way of expression. It consists of six elements: people, time, space, installation, action, and information, and are associated through certain logic. People, the center element, is

Human Factors Based New Media Design: Methodology and Assessment

43

associated to object, time, space, and information and makes contact to all the elements. People controls electronic devices (Installation) and electronic devices provide feedback to human; People feels space, space puts people in a multidimensional environment; People performs linear activities along time, time put people under time-order. (1) Time experience. Time is the main thread throughout audience activities. It connects audience’s immersion, interaction and ideation. Time can be viewed as one dimension of space. Jointly with space, they present entire structure, logic and layers, and enriches the space form. The design of the project is according to time of simulated experience. (2) Space creation. The advance of digital technology provide additional support to space creation, combination of reality and virtual space, creation of richer form. Through digital technology, the project creates digital metaverse and belongs to cross-domain or virtual territory via network, interaction, dialogue. The cultural perception of material objects are explained by information simulation, which is a collection of technology and art in reality and fantasy (see Fig. 4). (3) Presentation of multi-modal information. Multi-modality is a fusion of multiple senses, which provide emerging experience to audience through technological interactions using light, sound, electricity, etc. In particular, vision is the main information channel, therefore the visual performance is the crucial part of media design. Human vision, hearing, touch, taste and smell have a certain integrity and connection. Design should not only consider their individual biological feature and limitation, but also their interrelations. When a stimulus passes into one sensory system, not only the corresponding sensory system experiences, but also produces conscious experience of another sensory system. That is synesthesia, a psychological phenomenon in cognition process. The project stimulates the participation of multiple senses and performs multi-modal interactive design in order to create a comprehensive feast of sight, hearing and touch. To summarize, design should consider the relationship between time, space, device, information and audience and systematically design ’time- space- human-device’. Information from multiple senses should be integrated so that audiences are interacted in the context of information space.

4 Human Factors in Test Phase The experience and assessment from audience is an important measurement to the value of media design. It is also important reference to iteration and revision. (1) build assessment score system. Establish a reasonable score system is the prerequisite and foundation for an accurate and objective assessment. Each score element should present certain feature of the assessment object. The score system should be comprehensive, representative, and operable. The first-level dimensions and second-level scores of the project ‘The magic trip to Sanxingdui ruins’ are filtered and matched through the indicators originated from user research. SPSS’s factor analysis is used to build score system. The interactivity,

44

S. Ge et al.

immersion, aesthetics, and communication are the first-level scores. For example, under the first-level interactivity, there are secondary scores like naturalness, feedback speed, and operator fatigue. Clarity first and second level score and score weights, calculate final result, and establish score system. (2) Assessment method. Appropriate score method should be selected before human factor tests. The assessment methods mainly include: expert assessment and audience interview. The assessment question list for expert assessment is determined qualitatively. On the other hand, the performance is measured from the operational data of predefined tasks, that are performed by an audience group. For example, the time spent and error rate. (3) System test. In the testing phase, qualified audiences are identified and recruited for certain mission experience. Through interview and questionnaire, encode the text using qualitative method and analyze common requirements, and statistical analysis is performed for multiple operational data. (4) Assessment and iteration. Data analysis is used to find improvement areas. Therefore design method can be iterated and corrected to improve user experience. After multiple tests and iterations, the project is finally widely recognized by audience.

5 Conclusions New media design is a deeply integrated product between technology and design. The applications are exploring new technology and use richer and more diverse expression to share knowledge, transfer information, disseminate ideas, output aesthetic and artistic taste, express emotions, etc. This research is based on human factor based new media design, and its feasibility is tested through ‘the magic trip to Sanxingdui ruins’ project. This is a response to the development of future new media design, in which interactions are multi-modal, and virtual and reality will co-exist.

References Liao, X.: What is new media?. Modern Commun. (J. Commun. Univ. Chin.) 5, 121–125(2008) Li, S.: Research on human-computer interactive styles in installation art. Art Des. 2(8), 146–148 (2011) Wei, Y.: Interactive Art of New Media. Beijing Chemical Press, Beijing (2017) Lu, X., Bu, Y.: Review of practice and development of information design. Packaging Eng. 42(20), 92–102+12 (2021) Liu, G.: Outline of Matterology. Central South University Press, Changsha (2006) Salvendy, G.: Handbook of Human Factors and Ergonomics, 4th edn. Wiley (2012) Sun, Y.: Human Factors Foundation and Design. BIT Press, Beijing (2010) Sun, W.: Interactive narration of digital media arts. Nanjing University of Arts. (2011)

Increasing Quality of Life by Playing Chess: A Blended Care Approach for Elderly People Matthias Haase1(B) and Jens-Martin Loebel2 1 Department of Engineering and Industrial Design, Magdeburg-Stendal University of Applied

Sciences, 39114 Magdeburg, Germany [email protected] 2 Department of Economics, Magdeburg-Stendal University of Applied Sciences, 39114 Magdeburg, Germany [email protected]

Abstract. As digitization in healthcare progresses, e-health technologies are becoming increasingly important and can complement traditional healthcare services. While e-health services are well accepted and used by younger people, usage barriers exist for people with disabilities and the elderly. Participatory Technology Design (PD) may be a promising approach to address these usage barriers, involving users in the development process to increase user acceptance and prevent potential reactance effects. Chess, one of the world’s oldest board games, has been shown to promote cognitive abilities in a similar way to computer-based training and increase the quality of life of elderly people. This study investigates the effect of a six-week blended care chess training course on elderly people in four retirement homes. The results show a tendentially significant increase in quality of life, indicating that blended care chess training of elderly people can lead to an increased quality of life. Keywords: e-health · elderly · cognitive training · participatory design · chess · quality of life · blended care · technology acceptance

1 Introduction 1.1 E-Mental Health Among Elderly As digitization progresses, e-health technologies become more important and can complement existing health care services. The importance of digital applications for the healthcare sector is illustrated by the fact that the German government established a regulatory framework for the healthcare system to pay for these applications. Financing covers, for example, telemedical consultations, applications for organic diseases such as Diabetes, Hypertension, as well as applications for the treatment of mental illnesses such as Depression, Anxiety Disorders or Eating Disorders. For example results of applications for mental illness have shown promising effects (e.g. Phillips et al. 2019, Stein et al. 2018). © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 45–55, 2023. https://doi.org/10.1007/978-3-031-34866-2_4

46

M. Haase and J.-M. Loebel

A systematic review of e-health among elderly people has identified a broad range of interventions that can promote increased physical activity and other health promoting practices. These interventions focused on improving physical, psychological, and social well-being through lifestyle changes. For example, e-health applications were able to raise awareness for physical activity (Buyl et al. 2020). But there remains a fundamental problem. While e-health services are well accepted and used especially by younger people, there are usage barriers for people with disabilities and the elderly. In fact, there have only been a few studies that have focused on e-health use for people with disabilities (Vázquez et al. 2018) or elderly people (Buyl et al. 2020). However, there is not only a lack of engagement from the research perspective. Even though the Covid pandemic led to a rapid increase in the use of e-health services among elderly people as well, adoption rates on average are still lower than among younger people. Studies dealing with the barriers in the use of digital technologies offer potential clues also for the field of e-health services. Reasons that have been identified are lower technology-related self-efficacy, a lack of ability and skills in using modern information technology, and limited access to modern technology. Furthermore, there could be concerns that social interaction is increasingly becoming less common (Vaportzis, et al. 2021; Wilson et al. 2021). Furthermore, there could be a fear that increasing digitization will lead to less social interaction. Even technologies like adaptive Smart Home systems and Ambient Assisted Living, that could play an innovating role in age-appropriate care and housing for the elderly, show low acceptance rates (Ehrenhard et al. 2014). Given this, the aim should be to involve people in developing healthcare applications. Several publications have shown that involving elderly in the entire developing process is promising (Kope´c et al. 2017). Given this, the aim should be to involve people in developing healthcare applications. Several publications have shown that involving elderly in the entire developing process is promising (Kope´c et al. 2017). One central field for e-health technologies among elderly is the preservation of their cognitive abilities, which is one of the fundamental requirements for a long-term independent life. Some commercially available computerized cognitive trainings, for instance, has been shown to support healthy cerebral aging and prevent possible agerelated cognitive deficits (Klimova 2016; Shah et al. 2017). Especially working memory and reasoning skills of the trainees can benefit from these interventions. However, it is doubtful whether such training methods will reach as many elderly as possible. Game-based intervention can increase accessibility to elderly people. In this regard the game of chess, which is one of the world’s oldest board games and played by millions of people around the world, deserves a special mention. Even if elderly people do not play this game themselves, they usually at least have some indirect relation via friends, family, or partners. According to the current state of research, chess is suitable for promoting cognitive abilities in a similar way as computer-based training (e.g. Vale et al. 2018). In addition to promoting cognitive abilities, playing chess also increases the quality of life of elderly people in many areas (e.g. Vale et al. 2018, Mikhaylova et al. 2021). In the area of early childhood development, it is considered that chess training affects creative thinking, logical thinking and other intellectual states (e.g. Sala and Gobet 2016; Sigirtmac 2016). It was even proven that children with learning disabilities could improve in various areas of learning with the aid of chess training (Scholz et al.2008).

Increasing Quality of Life by Playing Chess

47

Yet, while there is a wide supply of chess trainings for children, there are currently only few trainings offered for elderly people. Therefore, the authors and students of the Human-Technology Interaction degree program at the University of Applied Sciences Magdeburg-Stendal have dealt with this issue more intensively. Since October 2021, students have been offering six-week chess trainings in small groups of mostly three to five elderly individuals. First preliminary findings on possible changes in attention and quality of life as a result of the training courses will be presented here. Furthermore, broad steps for the realization of a so-called inclusive chess app will be presented. 1.2 Participatory Design and Other considerations In recent years, there has been a growing emphasis on participatory design as a methodology for developing software applications, particularly in e-health. E-health services have become an important part of modern healthcare, providing innovative solutions for managing health and wellness. However, while these services are well accepted and used especially by younger generations, there are still usage barriers for the elderly and people with disabilities. Among the reasons that have been identified are lower technology-related self-efficacy, a lack of ability and skills in using modern information technology, and limited access to modern technology (Wilson et al. 2021). Several publications have shown that involving elderly in the developing process is promising to help mitigate these issues (e.g., Kope´c et al. 2017). Participatory design may be a methodological approach to address users’ needs (e.g., preferences, attitudes) and requirements (e.g., potentially sensory limitations) during the development process as they are actively involved in the entire process from conceptualization to implementation (Duque et al. 2019). This could avoid or mitigate usage barriers, increase user acceptance, and prevent potentially reactance effects. Participatory design can increase acceptance rates, especially for older and technology inexperienced individuals, as well as give more consideration and encourage individual and creative approaches to solving problems when dealing with technological systems (Davidson and Jensen 2013). We designed a chess app utilizing the participatory design process in collaboration with the participants. Participants’ feedback directly went into the development process of the app, in an iterative process that involves continuous feedback and collaboration between developers and users from identifying the users’ needs and requirements, creating, and testing a prototype, to evaluating and iterating the design based on user feedback. In the context of e-health, there is evidence to suggest that increased user acceptance resulting from participatory design can also lead to increased adherence to health interventions. For example, Goyal et al. (2016) conducted a study to investigate the impact of user-centered design on user engagement and adherence to self-management tasks in a mobile health app for diabetes. The study found that involving patients in the design process led to higher levels of user engagement and improved adherence to selfmanagement tasks, as compared to a standard app design without user input. Similarly, a systematic review by Moore et al. (2019) concluded that participatory design in e-health applications can be a promising approach to improve adherence. Participatory design principles in the development of e-health apps for elderly people, such as the chess app

48

M. Haase and J.-M. Loebel

discussed in this paper, can therefore be an effective way to increase user acceptance and adherence, especially among technology-inexperienced individuals. By involving users in the design process, we can ensure that the app meets the unique needs and preferences of its target users, thereby increasing the likelihood of user engagement and adherence to the health intervention.

2 Methods 2.1 Inclusive Chess App The app was developed by students as part of a course in Human-Technology Interaction degree programme of Magdeburg-Stendal University of Applied Sciences to complement in person training regimes and to consolidate lessons learned, allowing for single player training sessions as well as free play. Memorizing the user’s advancements, the app recommends specific chess exercises and tracks skill progress. The procedure for developing the app is based on design thinking and participatory design methods. For us, the design thinking process started with the requirements profile of our participant group comprised of elderly people in retirement homes that still have age-related cognitive abilities. Based on the information about the user group defined there, requirements and obstacles in developing an app for elderly people were gathered using guidelines on age-appropriate app design (Darvishy et al. 2016). Furthermore, Next-JS framework in combination with TypeScript language (a superset of JavaScript) was used to provide a more structured and maintainable code base. NextJS is a popular React framework that offers several advantages, such as UI rendering, touch support, and easy deployment across a wide range of modern mobile devices independent of screen size and resolution. Typescript provides type annotations, making it easier to catch errors before they occur, and allows for better tooling and editor support, aiding the development process for students. React allows for building reusable components, which is a good practice that aids the iterative user-centric participatory design process. This feature allows for quick and robust redesigns and continuous deployment, making new features and designs available already for the next chess training session with retirement home participants. These technologies also enabled building a highly modular, reusable codebase, making it easier to add new features and maintain the app in the future. To implement the chess functionality of our app, several libraries, including Chess.js and react-chessboard were used. These libraries provided an easy-to-use interface for interacting with the game engine, as well as the ability to display the game board and pieces in a visually appealing and user-friendly manner. Additionally, MaterialUI (MUI) UI framework was used to create a consistent and accessible design language throughout the app. To provide single player training and move suggestions, the Stockfish open source chess engine and AI was utilized. Stockfish allowed for the implementation of a robust and intelligent chess engine, which can analyze game states and provide helpful feedback and training exercises to users. The Stockfish Chess Engine has many benefits, such as providing advanced chess analysis and improving the chess skills of users.

Increasing Quality of Life by Playing Chess

49

Android tablets were used as the deployment platform for the app. This decision was made based on the benefits of using tablets, such as their easy use in retirement homes, easy portability, battery power, and similar size to a book. The app was deployed as a web app, accessed via the tablets, which allows for offline-use and device independence. The app leverages web accessibility features and standards including those outlined by the Web Content Accessibility Guidelines (WCAG) and HTML5 for barrier-free design such as screen-reader support, visibility enhancing color schemes, and text alternatives. This makes the app accessible to a wide range of prospective users, including those with disabilities. The design thinking process was started with the requirements profile of the partner group/course, followed by research to understand the requirements and obstacles of developing an app for elderly people. Initial design prototypes were created using Figma. These prototypes served as a template for the actual working app. All source code for the chess app is available on GitHub (https://github.com/MTI-Schach-App/EinfachSchach). The app can be played via the corresponding website https://einfachschach.de. The developed chess app consists of several sections that cater to the needs of its target user group, currently elderly people residing in retirement homes. Upon first use, the app guides the user to set up an account using their first name and assesses their chess skills, providing options to choose between novice, experienced, or advanced/professional levels. After the initial setup, users can easily log in to the app using their chosen username. Alternatively, caretakers of the n facility can use the tablet camera to scan a QR code to access the app on behalf of the user. The app remembers login details, thereby eliminating the need for repetitive login procedures. Upon successful login, the app presents the user with two options: either to play a game of chess against the built-in AI, which adapts to the user’s skill level, or to select one of the pre-recorded chess exercises. The exercises are structured to train elderly users on specific chess situations and guide them through the solution process. The app animates moves and explains them, making it easier for users to understand and follow along (see Fig. 1). Additionally, users can play back moves at leisure, making the learning process more engaging and interactive. To assist users during play and exercises, the app employs help bubbles that provide quick in-UI feedback and guide users through the game. The app also highlights wrong moves and displays available moves for any selected chess piece, aiding users who may hesitate during play. The app facilitates easy touch movements and large UI elements, allowing users with motoric impairments to interact with the app by moving pieces through a simple touch interface. The app avoids drag-and-drop functionality, instead, users touch a piece and then touch the corresponding destination on the chessboard. In conclusion, the app design is focused on the ease of use, learning, and accessibility. The app is structured in a user-friendly manner, enabling elderly people with various skill levels to enjoy and improve their chess skills. The app’s various features, such as the adaptive AI, pre-recorded exercises, and touch-based movement, are all aimed at enhancing user experience and interaction with the app. The app’s design thinking process and use of well-established web-based technologies further bolster the app’s effectiveness and potential for use in nursing homes. Feedback from users was collected after each in person training session and incorporated into the next design iteration.

50

M. Haase and J.-M. Loebel

Fig. 1. The main game screen with movement hints and exercise/tutorial selection of the chess app.

2.2 Sample The following students listed in alphabetical order recruited the facilities, conceptualized the chess trainings, conducted the trainings, and surveyed the questionnaires at the beginning and at the end of the trainings: Moritz Bütow, Thomas Gäde, Maximilian Guenter, Vilen Rafayevych, Philipp Rambke, Alexander Schellbach, Loris Scholz, Fritz Süß, Kevin Timme and Eddi Wendel. Furthermore the following students in listed in alphabetical order took part in designing, developing, testing, and deploying the chess app as well as creating and maintaining the corresponding website: Pauline Bauer, Elias J. Bongartz, Ben Erbig, Benjamin Ernst, Florian Kümmel, Daniel Linke, Lea Pitzek, Margaretha F. Rimboeck. Trainings were organized at one senior living center, two nursing homes and one community center for elderly people. In total 20 participants joined the trainings. But only 14 gave their agreement to participate the Study. Those who declined participation mostly felt unable to fill out the questionnaires. The average age of the 14 respondents was 85.3 years. Eight of them were male and six female. Seven of them had a university degree, whereas the other had an vocational job (Gäde et al. 2022). 2.3 Chess Training During the six-week training, an interaction of knowledge acquisition, mobilization and activation took place (Gäde et al. 2022). At the beginning of the training sessions, participants and trainers exchange information about their training expectations and participants report if they already have

Increasing Quality of Life by Playing Chess

51

chess experience. Although these chess trainings were offered in facilities for the elderly and only residents of these facilities participated in the trainings, it is not certain that the participants already know each other. Still during the first session participants not just learned the rules of chess. They also have to familiarize with different targets and develop strategies to complete the tasks. Participants usually learn the basic rules in two to three sessions. However, the instructors ensured that the idea of playing the game was not neglected even while learning the basic moves, and so the participants already played against each other at the end of the first session. If participants had already played chess, the rules were refreshed, or tasks were given to them. For example, the exercises required a certain opponent’s figure be captured in a precisely defined number of moves. Especially the varying levels of participants’ skills required maximum flexibility from the instructors in designing the training sessions. Learning, repeating and consolidating new information and problem-solving strategies help to develop working memory, also called short-term memory. Each training was led by two instructors (Gäde et al. 2022). The inclusive chess app was used at this stage of that project sporadically, mostly for evaluation tasks and not in a systematic way. 2.4 Instruments Research on computer-based cognitive training to preserve health of elderly (Klimova 2016; Shah et al. 2017) as well as studies on chess playing of elderly (e.g. Vale et al. 2018, Mikhaylova et al. 2021) reported an improvement in quality of life as well as in cognitive functions. Measuring quality of life as well as cognitive functions is challenging for very elderly people. Taking only the test-psychological point of view, high test quality criteria should be taken into account. But reliable and valid instruments are usually quite extensive and could lead to an cognitive overload of the participants. Therefore, emphasis was given to a high economy of the tests. Therefore, the questionnaires for measuring quality of life (EUROHIS-QOL) (Power 2003) and well-being (WHO-5) (Bech 2004) as well as the instrument for self-assessment of concentration and performance (APSA) (Bankstahl and Görtelmeyer 2014) will be used. EUROHIS-QOL. The EUROHIS-QOL (Power 2003) assesses on first scale psychological, physical, social and on second scale Environmental Factors of quality of life using eight items. Additionally, a total score can be calculated using all items. The test subjects are asked questions about different aspects of the quality of life. Using a fivepoint Likert scale from “not at all” to “completely,” subjects have to rate the level of their agreement. The questions refer to the last two weeks. The internal consistency (α = .85) can be considered as good (Brähler et al. 2007). WHO-5. The WHO-5 (Bech 2004) is a questionnaire for assessing well-being. For this instrument, only an total score can be formed to assess well-being. The WHO-5 total score can also be used as a screening instrument. A total score below thirteen indicates a low or poor level of well-being. Statements are presented, which the subjects have to check for themselves with the help of a six-point Likert scale (“At no time” to “All the time”). For example, one statement says: “In the last two weeks, I was happy and had a good mood”. All questions refer to the last 2 weeks. The internal consistency (α = .93)

52

M. Haase and J.-M. Loebel

can be considered as very good (Brähler et al. 2007). A joint total score can be created based on the EUROHIS-QOL and WHO-5. APSA. The APSA (Bankstahl and Görtelmeyer 2014) uses 21 items to create two subscales, Prospective Memory Performance and Difficulties in Maintaining Focused Attention Performance. The subjects are confronted with a statement and have to rate how strongly they agree using a 5-point Likert scale (“never” to “always”). An example item for the scale Difficulty in Maintaining Focused Attention Performance: “Even a small noise from the environment could disturb me while I was reading”. An example item for the scale Prospective Memory Performance: “It has happened that I suddenly did not remember what I wanted to do at that moment. The internal consistencies for the total score (α = .90) can be considered as very good. Furthermore, the convergent validity was proven by means of correlations with appropriate procedures. It should be noticed that the authors specify that this test can be used up to the age of 65. But there are no appropriate self-assessment questionnaires for this age. Moreover, the primary aim was to make a pre-post comparison, so the instrument has been used.

3 Results These results have been presented in a similar form as part of a student project (Gäde et al. 2022). First, a test of normal distribution was performed. The KS Test determined a statistically significant result for the EUROHIS-QOL scale Environmental Factors only. This applied to the time point prior to the beginning of the chess training (t1) (p < .001) as well as after the end of the chess training (t2) (p < .008). This implies that for this scale only, there is no normal distribution and no ANOVA can be calculated. ANOVA revealed no significant change between training start (t1) and end of training (t2) for any scales. Only for the total score, which is calculated from the EUROHISQOL and the WHO-5, a tendentially significant improvement between the beginning and the end of the training could be determined (F = 3.86, p < .078). This can be recognized also descriptively on the basis of the relatively clear differences between t1 (M = 42.5) and t1 (M = 49.5) (Table 1). For ASPA, no differences for inferential statistical were identified. Neither for the Prospective Memory Performance scale (F = 0.02, p < .90) nor for Difficulties in Maintaining Focused (F = 0.12, p < .73). Table 1. EUROHIS-QOL and WHO-5 changes over the course of training (t1 to t2) t1 M

t2 SD

M

SD 4,5

EUROHIS-QOL Psychological, Physical and Social Environmental Factors

20,7

3,4

18,1

7,6

1,2

8

1,6 (continued)

Increasing Quality of Life by Playing Chess

53

Table 1. (continued) t1 M Total Score

t2 M

SD

25,6

SD 5,5

28,7

4

16,8

6,8

20,7

5,3

42,5

11,4

49,5

8,8

WHO-5 Total Score EUROHIS-QOL and WHO-5 Total Score

4 Conclusions Due to the quite small sample of 14 subjects, no reliable statements can be made about the effects of the chess training. However, it can be mentioned that the tendentially significant effect of the overall scale of the EUROHIS-QOL and the WHO-5 can be considered a success despite of the small sample. The fact that there were no results for the APSA could probably be attributed to the test procedure, which, with 21 items, seems too long for the elderly test subjects. Only a few subjects were able to complete this questionnaire without assistance. Maybe an econometric external rating should be used to measure the cognitive abilities. Participatory design is an important methodology for developing e-health apps for elderly people, particularly those with limited technology-related self-efficacy and skills in using modern information technology. By involving end-users in the design process, participatory design can increase acceptance and adherence rates, and improve the quality of life of elderly people. The described inclusive chess app designed according to the participatory design process is just one example of how this approach can be used to address the specific needs and requirements of elderly people when it comes to e-health applications.

References Bankstahl, U.S., Görtelmeyer, R.: APSA - Attention and Performance Self-Assessment - deutsche Fassung (2014) Bech, P.: Measuring the dimensions of psychological general well-being by the WHO-5. QoL Newslett. 2004(32), 15–16 (2004) Brähler, E., Mühlan, H., Albani, C., Schmidt, S.: Teststatistische Prüfung und Normierung der deutschen Versionen des EUROHIS-QOL Lebensqualität-Index und des WHO-5 Wohlbefindens-Index. Diagnostica 53(2), 83–96 (2007). https://doi.org/10.1026/0012-1924. 53.2.83 Buyl, R., et al.: e-Health interventions for healthy aging: a systematic review. Syst. Rev. 9(1), 128 (2020). https://doi.org/10.1186/s13643-020-01385-8 Darvishy, A., Hutter, H.-P., Seifert, A.: Altersgerechte mobile Applikationen, 2016, Winterthur: ZHAW - Zürcher Hochschule für angewandte Wissenschaften. https://ageweb.ch/fileadmin/ user_upload/documents/Altersgerechte_mobile_Applikationen_ZHAW.pdf. Accessed 24 Feb 2023

54

M. Haase and J.-M. Loebel

Duque, E., Fonseca, G., Vieira, H., Gontijo, G., Ishitani, L.: A systematic literature review on user centered design and participatory design with older people. In: Guimaraes, R.L., Saibel Santos, C.A., de Santana, V. F., Kronbauer, A. (eds.) Proceedings of the 18th Brazilian Symposium on Human Factors in Computing Systems, pp. 1–11. ACM (2019). https://doi.org/10.1145/335 7155.3358471 Davidson, J.L., Jensen, C.: Participatory design with older adults: an analysis of creativity in the design of mobile healthcare applications. In: Proceedings of the 9th ACM Conference on Creativity & Cognition, pp. 114–123 (2013). https://doi.org/10.1145/2466627.2466652 Ehrenhard, M., Kijl, B., Nieuwenhuis, L.: Market adoption barriers of multi-stakeholder technology: smart homes for the aging population. Technol. Forecast. Soc. Change 89, 306–15 (2014). https://doi.org/10.1016/j.techfore.2014.08.002 Gäde, T., Günther, M., Timme, K., Süß, F., Rambke, P., et al.: Gesunderhaltung im Alter durch Schach. Magdeburg: Hochschule Magdeburg-Stendal (2022) Goyal, S., Morita, P., Lewis, G., Yu, C., Seto, E., Cafazzo, J. A.: The systematic design of a behavioural mobile health application for the self-management of type 2 diabetes. Can. J. Diabetes 40(1), 95–104 (2016). https://doi.org/10.1016/j.jcjd.2015.06.007 Klimova, B.: Computer-based cognitive training in aging. Front. Aging Neurosci. 8, 313 (2016). https://doi.org/10.3389/fnagi.2016.00313 Kope´c, W., Skorupska, K., Jaskulska, A., Abramczuk, K., Nielek, R., Wierzbicki, A.: LivingLab PJAIT. In: Sheth, A. (eds.) Proceedings of the International Conference on Web Intelligence; 23 08 2017 26 08 2017: Leipzig Germany, New York, NY, USA: ACM (2017) Mikhaylova, I.V., Medvedev, I.N., Makurina, O.N., Bakulina, E.D., Ereshko, N.Y., Eremin, M.V.: The effect of playing chess on an aging or pathological organism. J. Biochem. Technol. 12(3), 47–52 (2021). https://doi.org/10.51847/CwcjG5IstX Moore, G., Wilding, H., Gray, K., Castle, D.: Participatory methods to engage health service users in the development of electronic health resources: systematic review. J. Particip. Med. 11(1) (2019). https://doi.org/10.2196/11474 Phillips, E.A., Gordeev, V.S., Schreyögg, J.: Effectiveness of occupational e-mental health interventions: a systematic review and meta-analysis of randomized controlled trials. Scandinavian J. Work, Environ. Health 45(6), 560–76 (2019). https://doi.org/10.5271/sjweh.3839 Power, M.: Development of a common instrument for quality of life. In: Nosikov, A., Gudex, C. (eds.) EUROHIS: Developing common instruments for health surveys, pp. 145–164. IOS Press, Amsterdam, Berlin (2003) Sala, G., Gobet, F.: Do the benefits of chess instruction transfer to academic and cognitive skills? A meta-analysis. Educ. Res. Rev. 18, 46–57 (2016). https://doi.org/10.1016/j.edurev.2016.02.002 Scholz, M., Niesch, H., Steffen, O., Ernst, B., Loeffler, M., Witruk, E., et al.: Impact of chess training on mathematics performance and concentration ability of children with learning disabilities. Int. J. Spec. Educ. 23(3), 138–148 (2008) Shah, T.M., Weinborn, M., Verdile, G., Sohrabi, H.R., Martins, R.N.: Enhancing cognitive functioning in healthly older adults: a systematic review of the clinical significance of commercially available computerized cognitive training in preventing cognitive decline. Neuropsychol. Rev. 27(1), 62–80 (2017). https://doi.org/10.1007/s11065-016-9338-9 Sigirtmac, A.D.: An investigation on the effectiveness of chess training on creativity and theory of mind development at early childhood. Educ. Res. Rev. 11(11), 1056–1063 (2016). https:// doi.org/10.5897/ERR2016.2676 Stein, J., Röhr, S., Luck, T., Löbner, M., Riedel-Heller, S.G.: Indikationen und Evidenz von international entwickelten Online-Coaches zur Intervention bei psychischen Erkrankungen – ein Meta-Review. Psychiatrische Praxis 45(1), 7–15 (2018). https://doi.org/10.1055/s-0043117050

Increasing Quality of Life by Playing Chess

55

Vale, R.G.D.S., et al.: Effects of resistance training and chess playing on the quality of life and cognitive performance of elderly women: a randomized controlled trial. J. Phys. Educ. Sport 18(3), 1469–77 (2018). https://doi.org/10.7752/jpes.2018.03217 Vaportzis, E., Clausen, M.G., Gow, A.J.: Older adults perceptions of technology and barriers to interacting with tablet computers: a focus group study. Front. Psychol. 8, 1687 (2017). https:// doi.org/10.3389/fpsyg.2017.01687 Vázquez, A., Jenaro, C., Flores, N., Bagnato, M.J., Pérez, M.C., Cruz, M.: E-health interventions for adult and aging population with intellectual disability: a review. Front. Psychol. 9:2323 (2018). https://doi.org/10.3389/fpsyg.2018.02323 Wilson, J., Heinsch, M., Betts, D., Booth, D., Kay-Lambkin, F.: Barriers and facilitators to the use of e-health by older adults: a scoping review. BMC public health 21(1), 1556 (2021). https:// doi.org/10.1186/s12889-021-11623-w

Lessons from a COVID-Era Controlled Trial of Online Learning and Socializing with Older Participants Benefits of Applying Consolidated Standards of Reporting Trials Najmeh Khalili-Mahani1,2,3,3(B)

, Berkley Petersen4 , and Shannon Hebblethwaite1,5

1 engAGE: Concordia Centre for Research on Aging, Montreal, Canada

[email protected]

2 Department of Design and Computation Arts, Concordia University, Montreal, QC, Canada 3 McGill Centre for Integrative Neuroscience, McGill University, Montreal, QC, Canada 4 Department of Psychology, Concordia University, Montreal, QC, Canada 5 Department of Applied Human Sciences, Concordia University, Montreal, QC, Canada

Abstract. Randomized Controlled Trials (RCTs) are the gold standard in health intervention research. However, in lifestyle intervention studies, this methodology is often challenged by the complexity of personal and contextual factors, which are hard to control. To address this challenge, researchers have adopted Consolidated Standards of Reporting Trials (CONSORT) as a framework to rank the quality of RCTs based on detailed documentation of the design, methodology, and results of the study. In this paper, we present a case study to demonstrate benefits of adopting CONSORT as a framework for designing and conducting a trial to investigate the affordances of information and communication technologies (ICT) in lifestyle interventions. Framed as a randomized controlled trial, our study targeted healthy community-dwelling older adults in Montreal. The study took place between Oct 2021-July 2022 and included qualitative and quantitative data collection, about the benefits of online-learning (digital photography), and on-Zoom group discussions. All pre-post psychometric assessments and interviews were conducted on zoom. We evaluate our study to illustrate: (i) To what extent such RCT design, methodology and reporting can satisfy CONSORT; (ii) Which factors contribute to trials deviations from the CONSORT framework; and (iii) What data acquisition strategies maximize the benefits of applying the CONSORT framework. Our analysis indicate that CONSORT provides a frame within which one can flexibly document and contextualize the dynamics of how ICT interventions could benefit their target population. Keywords: Randomized Controlled Trials (RCT) · Older Adults · Information and Communication Technologies (ICT) · Online Learning. Zoom · Recruitment · Study Design · CONSORT · User Studies · Mixed-Methods Research

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 56–74, 2023. https://doi.org/10.1007/978-3-031-34866-2_5

Lessons from a COVID-Era Controlled Trial of Online Learning

57

1 Introduction 1.1 Background Since the onset of the COVID-19 pandemic, interest in studying the benefits of information and communication technologies (ICTs) for older adults has grown significantly [1, 2]. As a result, interdisciplinary research studies that explore ICTs as mediators of social connectedness and activity are flourishing. This line of research intersects with a global interest in achieving healthy ageing, and a large body of multidisciplinary studies that stretch from molecular biology to lifestyle and public health. Accordingly, interventions range from pharmacological to performative therapies. However, it has proven difficult to validate the effectiveness of behavioral interventions quantitatively, thus researchers seek strategies for improving the data quality [3, 4]. Additionally, given that myriad cultural, psychosocial and medical factors underlie this line of research, there is debate about methodologies that are most suitable in this line of work [5]. Traditionally, a medicalized discourse of ageing has dominated research involving older adults. A search of the PubMed database with the keyword (Healthy Ageing) (in October 2022) returns 59,059 entries (academic peer reviewed articles), of which 4,363 are clinical or randomized controlled trials. Many of these studies are biomedical and focus on epigenetic or pharmacological strategies in search of quantifiable reversal of cellular ageing measured from direct or surrogate biological markers. A considerable number of these clinical trials (n = 2,345), which start appearing in 1990s are related to lifestyle interventions, diet, exercise, and activities. Whether using common methods such as randomized controlled trials (RCTs) are suitable for studying lifestyle or complex dynamic systems is debatable [6–8]. Conducting RCTs that involve ICT-based interventions for older adults is even more complex due to various factors. These include not only age-related digital divide [9–11], but also the heterogeneity of adoption and acceptance of ICTs [3, 12–14], e.g., related to personal choice [15, 16], technostress [17–21] or even personality [22]. In addition, the socioeconomic contexts and theoretical perspectives can bias data-collection about the affordances of ICTs for older populations [5, 23]. This raises the question whether standards for high-quality RCTs can be applied to user studies that involve ICTs or assistive technologies for health interventions. To answer this question, we critically review one of our own studies to underline the advantages of conducting research within a careful RCT framework, while also accommodating for flexibility in study design and interventions. 1.2 Study Aims and Objectives High-quality RCTs are usually ranked based on CONSORT (Consolidated Standards for Reporting Trials) criteria. CONSORT was first published in 1996 [24], revised in 2001 (to address the importance of reporting sources of funding, ethical review board approvals and trial registration in the study methods [25]), and further improved in 2010 (to increase the accuracy of recruitment, trial allocation and analysis [26, 27].)

58

N. Khalili-Mahani et al.

In this paper, we aim to examine the strengths and weaknesses of conducting our research (on the affordances of ICTs for the well-being of older adults) under CONSORT criteria. Specifically, we answer the following questions: (i) To what extent did our RCT design, methodology and reporting satisfy CONSORT; (ii) Which factors caused deviation from the CONSORT; and (iii) Which strategies helped us take advantage of framing our research within the CONSORT framework.

2 The Case-Study: Affordances of ICTs for the Well-Being of Older Populations in the Context of the COVID-19 Pandemic. 2.1 Theoretical Framework and Research Justifications Previous research indicates a positive association between ICT use, mental health, and psychosocial covariates in later life [28], as well as a positive association between ICT usage and increased agency, self-efficacy, and quality of life [29–33]. Thus, within the context of COVID-19 as a socially isolating and lonely experience, we sought to examine the benefits of introducing creative and educational ICT activities to older adults. The rationale for the research was based on data suggesting that adults older than age 64 would be at a higher risk of social isolation and loneliness compared to other demographics [34, 35] and that there would be health and quality of life (QoL) benefits to increasing social interactions [28, 36]. The objective of this RCT research study was to: (a) Determine whether socializing online can reduce loneliness and improve social connectedness, general self-efficacy, and mood in healthy participants aged 65+ (Quantitative); and (b) Identify facilitators and barriers to adoption of online social activities (Qualitative). Figure 1 provides a typical conceptual framework to support a study of the affordances of ICT in mitigating the adverse effects of COVID-19 social isolation.

Fig. 1. Conceptual framework that guided the RCT research design

Lessons from a COVID-Era Controlled Trial of Online Learning

59

2.2 Design Constraints Several constraints guided our study design. First, this study was to take place under the COVID-19 safety and security conditions and prudence for the health of participants, as well as researchers. The study recruitment began in October 2021, shortly after the vaccine passports were introduced. At this time, the restrictions for social gathering were eased, but a new wave of vaccine-resistant Omicron infection hit Quebec. As such, all elements of this study had to take place online and with minimal physical contact (except for receiving a mail package with instructions and consent forms). Second, this study was funded under the expectation that it would include both a qualitative and a quantitative component, and that it would benefit from an interdisciplinary collaboration between researchers focusing on sensory and cognitive well-being of older adults, as well as those interested in social connections and general well-being. The data gathered was intended to maximize the amount of knowledge that could be obtained from studying older adults’ experience with different types of ICTs. Specifically, the research team was interested in learning about the capacity of older adults to utilize ICTs in their homes and independently. Therefore, the study design needed to envision and test realistic scenarios of participants’ experiences within a complex ICT ecosystem and had to be designed to control for such complexity. Based on these requirements, we proposed an intervention that involved using ICTs for three different functions: • Conducting an online RCT involving multimodal data collection. • A four-week online learning intervention (on Coursera) about digital photography. • A condition for interaction and photo-sharing with other study participants on Zoom. Based on the rationale described above, we designed a randomized parallel design study under two conditions (social or solo) with a common intervention (online digital photography), and controlled for primary outcomes (pre/post psychometric evaluations), followed by a qualitative interview with participants to assess their overall experience (Fig. 2).

Fig. 2. Diagram of study design detailing the derailments (from target N = 128, to randomization and enrollment created by individuals expressing their needs and preferences).

60

N. Khalili-Mahani et al.

3 The Extent of Compliance with the CONSORT Framework Tables 1 and 2 provide a summary of which study elements checked against the CONSORT list for appropriate reporting of Methods and Results. Table 1. Study elements cross-checked against CONSORT list for reporting Methods METHODS

Item

Checklist item

Reportable

Trial design

3a

Description of trial design

YES

3b

Important changes to methods after trial YES commencement

4a

Eligibility criteria for participants

YES

4b

Settings and locations where the data were collected

NO

Interventions

5

The interventions for each group with sufficient details to allow replication, including how and when they were administered

NO

Outcomes

6a

Completely defined pre-specified YES primary and secondary outcome measures, including how and when they were assessed

6b

Any changes to trial outcomes after the trial commenced, with reasons

YES

7a

How sample size was determined

NO

7b

When applicable, explanation of any YES interim analyses and stopping guidelines

8a

Method used to generate the random allocation sequence

YES

8b

Type of randomisation; details of any restriction (such as blocking and block size)

NO

9

Mechanism used to implement the random allocation sequence (such as sequentially numbered containers), describing any steps taken to conceal the sequence until interventions were assigned

NO

Participants

Sample size

Randomisation: Sequence generation

Allocation concealment

(continued)

Lessons from a COVID-Era Controlled Trial of Online Learning

61

Table 1. (continued) METHODS

Item

Checklist item

Reportable

Implementation

10

Who generated the random allocation sequence, who enrolled participants, and who assigned participants to interventions

NO

Blinding

11a

If done, who was blinded after assignment to interventions (for example, participants, care providers, those assessing outcomes) and how

NO

11b

If relevant, description of the similarity of interventions

NA

12a

Statistical methods used to compare groups for primary and secondary outcomes

YES

12b

Methods for additional analyses, such as YES subgroup analyses and adjusted analyses

Statistical methods

Table 2. Study outcomes cross-checked against CONSORT list for reporting results RESULTS

Item

Checklist item

Reportable

Participant flow diagram

13a

The numbers of participants randomly assignment, received treatment, and were analysed for the primary outcome

YES

13b

For each group, losses and exclusions YES after randomisation, together with reasons

14a

Dates defining the periods of recruitment and follow-up

YES

14b

Why the trial ended or was stopped

YES

Baseline data

15

A table showing baseline demographic and clinical characteristics for each group

YES

Numbers analysed

16

For each group, number of participants (denominator) included in each analysis and whether the analysis was by original assigned groups

YES

Recruitment

(continued)

62

N. Khalili-Mahani et al. Table 2. (continued)

RESULTS

Item

Checklist item

Reportable

Outcomes and Estimations

17a

For each primary and secondary outcome, results for each group, and the estimated effect size and its precision (such as 95% confidence interval)

YES

17b

For binary outcomes, presentation of NA both absolute and relative effect sizes is recommended

Sample size Ancillary analyses

18

Results of any other analyses performed, including subgroup analyses and adjusted analyses, distinguishing pre-specified from exploratory

Harms

19

All the important harms or unintended YES effects in each group (for specific guidance see CONSORT for harms)

YES

3.1 Items that Satisfied CONSORT Checklist for Methods Description of Trial Design (Item 3a). The study was designed as a non-blinded, controlled, two-armed randomized trial. The intervention involved participating in an online digital photography course (via Coursera). One study arm (social) involved weekly Zoom video chat meetings between the study participants, a technical coach, and a researcher (the first author) to enable them to share their experiences of the photography course with each other. The other arm (solo) did not have this social component and participants engaged in the course individually without interacting with others in the course. The study aimed to investigate the impact of this Zoom socialization on wellbeing (see next sections). In both study arms, participants experienced the following: First, they completed a long pre-screening online survey (~40 min) to create data about each participant’s profile: demographics (employment, income, gender, age, marital status, education), health state (history of chronic illness, surgery), Social Participation Questionnaire (From Canadian Longitudinal Study of Ageing), Social Support Availability, as well Beck Depression scores, General Self-Efficacy, UCLA Loneliness, as to document participant characteristics and to gather baseline measurements of outcome scores on emotional and social well-being outcomes. Second, they were invited to a pre-study screening session on Zoom with a researcher (second author) where their functional states (cognition, vision, mobility) were measured using the Timed Up and Go test and 5 Times Sit to Stand for mobility, as well as MN Read Pocket Cards for visual acuity, Spot Checks for Contrast Sensitivity (vision) and the Montreal Cognitive Assessment (MoCA). Prior to this session, participants had received

Lessons from a COVID-Era Controlled Trial of Online Learning

63

a package by post, with materials and instructions to self-administer these tests while videoconferencing with the researcher. Third, they received instructions about the course and were assigned a group (solo or social). The social group agreed to meet at a certain time on a weekly basis (for 4 consecutive weeks). Forth, at the end of the course, cognition, and mobility we re-assessed via Zoom with the same researcher who administrated the pre-study measures. Finally, participants had an individual semi-structured interview on Zoom with one of the researchers. Questions focused on their general experience with online-learning and online-socialization, and their experience with the study. Important Changes to Methods After Trial Commencement (Item 3b). The trial was initially designed to include 128 participants to be randomized into two groups (n = 64) and the social group split into groups of 8 for weekly zoom interactions. However, the recruitment was not sufficient. Eligibility Criteria for Participants (Item 4a). Eligibility criteria were clearly set and announced. To participate in the study, individuals had to be aged 65+, English-speaking, without severe visual impairment, and with access to a computer, web camera, broadband internet, and a digital photography device (DSLR, phone, point-and-shoot camera, etc.) Completely Predefined Outcomes (Item 6). Predefined outcomes included General Self Efficacy, UCLA Loneliness Index, Geriatric Depression Scale. None of the outcomes were modified after the trial (item 6b).

3.2 Items that Satisfied CONSORT Checklist for Results Given that the study was designed and conducted with close attention to CONSORT standards, we had collected sufficient details to be able to satisfy the entire checklist for results. Participant Flow Diagram (Item 13). The participant flow into the study was carefully documented and represented in a diagram (Fig. 3). Dates of Recruitment were Recorded (Item 14). Recruitment took place between October 2021–July 2022, and through targeted advertisement to individuals who had subscribed to various research mailing lists to be called for volunteering in wellness studies. The advertisement was sent to at least 4400 individuals, and 37% of those opened email. Recruitment stopped due to not receiving any calls for two months after the last recruitment message was sent out. Baseline Data (Item 15). Was recorded for all explanatory variables listed in methods. Numbers Analyzed (Item 16). The existence of quantitative data allowed us to create noodle charts, to illustrates heterogeneity of change in outcome variables from the baseline (Fig. 4). Outcomes Estimations (Item 17). Despite the small sample size, the pre/post measurements provide pilot data to estimate an effect size for some of the variables included.

64

N. Khalili-Mahani et al.

Fig. 3. Detailed participant flow diagram documents attrition at every step.

Loneliness

Depression

Self-Efficacy

Fig. 4. Acquiring outcome-measures in two timepoint and noodle diagrams help illustrate withinand between subject variability in well-being outcomes. Discussion of the quantitative outcome is beyond the scope of this report.

Sample Size, Ancillary Analysis (Item 18). In principle, if we had recruited a largeenough sample it would have been possible to conduct such analysis. Harms (Item 19). We conducted exit interviews with each of the participants to register any possible harm or discomfort they could have experienced in the course of the study. This allowed us to identify a certain level of discomfort and ambivalence with online psychometric analyses.

Lessons from a COVID-Era Controlled Trial of Online Learning

65

4 What Factors Caused Deviation from the CONSORT? Several factors in the study Methods did not satisfy the CONSORT checklist (Table 1). These included: Setting of Participation (Item 4b). Setting and location of participation could not be controlled in terms of settings and locations where the data were collected. Participation was online via Zoom and the participants chose the location from which they could participate (i.e., where they had access to a device and internet connection). The location, therefore, was not common or standardized. The participants were not provided with digital devices., This led to not only exclusion of those who did not own them (digital photography, broadband internet, and camera-equipped devices for communication), but differences in the quality of the equipment owned by the participants. Intervention (Item 5). This study had two interventions, neither of which could be described or controlled in sufficient details to allow replication. While the online course (Coursera) was structured, and we assigned weekly assignments, we could not control when, how and to what extent the participants engaged with the course material. Indeed, several participants indicated that they did not comply with the suggested routine--due to finding the course material boring or exhausting. Sample Size (Item 7a). Given the diversity of outcome measures, we could not calculate the sample size based on a priori data, and as such, we chose a sample size to be maximally sensitive to detecting individual differences in visual acuity, as an explanatory variable for characterizing interindividual differences in engagement with on-screen activities. Randomization (Items 8b–10). Although we had a mechanism for randomisation in place (assigned to group A or B based on order of receiving call from volunteers), we could not stay within the protocol. We received explicit requests from 3 participants who wished to be included in the ‘socializing’ arm of the study, and given the small sample of interested parties, we honored their request. Also, because we had provisioned to hold 8-person socializing groups, we needed to put the first group of participants in the social group so we could start data collection and stay within the time constraint of research. We deviated from the protocol by forming 4-person groups (instead of 8) due to insufficient response rates to our recruitment efforts. Blinding (Item 11). In these types of interventions, blinding is impossible. Statistical Methods (Item 12). While we had initially aimed to conduct parametric analysis (mANOVA) to test the Group x Time interaction effects on primary outcome variables, we did not reach sufficient sample size to carry out such analysis.

5 Taking Advantage of the CONSORT to Increase Flexibility Provisioning for the possibility that adherence to strict RCT protocol would not be possible, we collected qualitative data to capture individual data about the experience of the trial (e.g., during the exit interview and personal communications). This data helped gain insight into the needs and preferences of study participants and sources of bias. As results in Sect. 4 show, almost all study factors related to participants’ choices, or their living circumstances of participants contributed to a deviation from our study protocol.

66

N. Khalili-Mahani et al.

5.1 Recruitment This study took place in Montreal (Oct 2021–Jul 2022), Canada, and targeted individuals who had enlisted in a pool of volunteers for lifestyle interventions for healthy ageing. In total, 78 individuals contacted us to inquire about the research and received a link to a screening survey. Of these, 33 opened the survey; and 27 completed it. Only 16 participants committed to the entire study. Our qualitative analysis of exchanges with participants revealed sources of bias in recruitment and retention. Self-Exclusion. Among those who received information about the study, 18 wrote back to state disinterest in the study, with 12 offering an explanation. The stated reasons for not participating were Zoom fatigue (n = 5), the amount of time commitment (n = 4), and general disinterest without specific reasons offered (n = 4). Two comments are noteworthy as they reflect the participants’ discomfort with the experimental paradigm itself. Two correspondences pointed out to the individual’s appraisal of the intervention itself: “This invitation caught my interest (and my anger). I know that I will not be considered a candidate for your study because I do not use Zoom. However, the exclusion caused by this fact over the past two years has been severe.” “The study seems highly invasive, requiring a lot of personal information and participation on Zoom where the subjects will be “studied”. I couldn’t connect the purpose of an online digital photography course with socializing on Zoom? Too much of this study is dedicated to being on a Zoom platform, just to collect data on an older demographic (65 years and older). It read more like a marketing activity rather than exploring “social isolation” of Seniors.” Two other correspondences illustrated discomfort with the experimental paradigm itself. One participant was not willing to be randomly selected to be in the social group: “The random selection of people for the discussion groups especially would be a problem. Is there a way not to be involved in that part of it?” Another participant objected to the cognitive evaluation test: “I would be interested in participating. As long as I do not have to do the Ladybug Elephant Zebra test, which always left me feeling like an idiot....” Reasons for Participation. The 16 individuals who committed to the study requirements were also ambivalent about the time commitment, and weary of ‘yet another Zoom’ activity. However, they were mainly drawn to it to learn or hone a skill (digital photography). Those who committed to the study had the following common characteristics: • They had already participated in other ageing studies conducted at our university. • They were all interested in the topic of the course (digital photography), had prior experience with it, and were interested in improving their photography skills. About

Lessons from a COVID-Era Controlled Trial of Online Learning

67

half of the participants who identified as professional and emerging artists remained involved and committed to taking the research one step further and chose to exhibit their photographs that were created through this initial encounter. • They tailored the degree of adherence to the course to their interests and deviated from it when necessary. • They enjoyed the social interactions with like-minded peers. • They enjoyed the creative process and exchanging ideas and feedback on their photography works. 5.2 Participants Adherence to the Study Elements Appraisal of the Study. Among the 16 individuals who committed and completed the study, only one person withdrew due to developing long COVID at the end of the group study. As many indicated in the exit interview, they had tentatively decided to join a study whose aims were unclear to them to see how it went; and ended up staying because they appreciated the group dynamics in the weekly meetings. All participants had a priori familiarity with research methods in studying healthy ageing. However, they could not make a direct connection between the extensive battery of health questionnaires. Both in the first group meetings, and later in the individual interviews, participants raised the question: ‘what do you think you can find?’ Almost all participants indicated that they could not make a connection between the interventions (learning digital photography in an online course) and the measurements (memory, movement, vision). While all participants expressed positive views of the dynamics of interactions among participants and the researcher and facilitator who participated in the online zoom discussions, three participants made comments about the implied ageism in the quantitative data collection. Adherence to Intervention. While participants had an interest in photography and enjoyment of group dynamics in common, they did not equally comply with the course instructions. Those who had prior photography skills enjoyed and followed the course closely. Those who did not found the online lecturer or the topics unstimulating or irrelevant to the reasons they had joined the study. Adherence to the online course was slightly better in those in the social group (as they discussed how they exercised lessons in their weekly photography). All three participants in the non-social group had prior experience with advanced digital photography, but they were delayed with the completion of the course and postponed the post-intervention interview. Other than weekly checkups, we did not know what other activities they did during the trial. Likewise, we could not ‘isolate’ social interactions outside of the weekly meetings to be able to attribute the effects to our intervention.

5.3 Allowing for Deviations from the Experimental Protocol In response to the reality that we could not recruit a large representative sample, we opted for flexibility to ensure we maximized the opportunity of retaining participants in the study. This meant adjusting the dynamics of weekly group meetings to ensure

68

N. Khalili-Mahani et al.

they were inclusive and equitable in satisfying the interests of participants with different degrees of familiarity with digital photography. In total, we held three socializing-group meetings (four meetings per group). Because the meetings were sequential (a reality that was imposed on the protocol due to slow recruitment), we took advantage of feedback from participants in the first group to improve the experience for the next group. This process allowed us to gather more qualitative data and to extend the study to enable deeper engagement of participants with the topic of affordances of ICTs for their social well-being. Figure 3 illustrates this learning process. The added flexibility resulted in an art book and exhibit where participants became full partners in knowledge-creation (Fig. 5).

Fig. 5. Diagram of adjustments made in the social interactions based on feedback from participants in weekly group meetings.

6 Discussion 6.1 Summary of Findings We have used the CONSORT framework to illustrate the strengths and weaknesses of an RCT regarding the benefits of online learning and socializing for the well-being of older adults within the context of the COVID-19 pandemic. Our analysis shows the limitations of RCT methods in standardization of location and experimental setting, keeping track of and controlling adherence to the intervention, calculation of the sample size, and randomization. It also shows the strength of framing such a study within CONSORT criteria, especially in reporting the results. The first significant challenge in this RCT framework was in recruitment, followed by compliance with the protocol. Recruitment influenced other critical requirements for a high-quality RCT: the sample-size, randomization, and reliability of outcome measures.

Lessons from a COVID-Era Controlled Trial of Online Learning

69

Provisioning for qualitative data collection, and documenting protocol modifications, allowed us to capture various sources of interindividual differences in adherence to the protocol, thus, informing future research. 6.2 Implications We have presented a novel approach to systematically evaluate the strengths and weaknesses of an RCT involving ICT-interventions during the COVID-19 pandemic. In ageing studies, these RCTs generally focus on outcomes measured in terms of changes in cognitive or functional status (e.g., activities of daily living) or health status over relatively short time periods [37, 38]. Remarkably, several systematic and metaanalytical reviews indicate that in studies that target health behavior and lifestyle in older adults, trial outcomes are not ubiquitously positive and often inconclusive [39–42], or ineffective [43]. Adding patient-reported outcome measures can increase the chances of measuring a positive effect but only if the outcome is predefined [42], or measured in short-term and with active support from researchers [44]. Ignoring complexity skews expectations from the reproducibility of science [45]. According to the CONSORT 2010 Statement, good RCT methodology is precise in describing its design, participant selection criteria, intervention details, pre-specified outcomes, sample size justification, randomization, double-blinding, and statistical modeling [27]. However, as our study reveals, introducing a complex ICT-based intervention with social, personal, technical, and practical components does not lend itself easily to a clinical RCT-design paradigm. In a mapping review of study-designs and methodologies used at the intersection of healthy ageing and assistive ICTs, Khalili-Mahani and Sawchuk showed that research framing and methodologies could bias sampling by alienating those who perceive implied ageism in research communication [8]. Our analysis corroborates previously reported challenges of recruitment and retention of participants in ageing studies, and the necessity of accounting for individual needs and interests [15, 16, 20, 46–48]. Our case-study corroborates that fun [15, 49–52] and meaningfulness [53–55] are more important factors for older adults to engage with ICTs, than cognitive and physical health factors. Therefore, collecting data about individual experiences within any RCT involving ICT interventions is important. For example, the timing of our study (taking place in summer) together with Zoom fatigue (due to the COVID-19 pandemic) were factors that contributed to failure of recruitment. However, maintaining an engaging and pleasant group atmosphere became an incentive that helped participants overcome their initial reservation about the study and engage with the study more deeply. 6.3 Guideline for Future Studies Adhere to CONSORT from the Beginning. The CONSORT aims to increase the precision of reporting, and to address challenges of reproducibility in science. The domains of information covered by CONSORT include not only the study protocol (study design, sampling, and reporting) but also deviations from the protocol, as well as the research context and conceptual framework that is influenced by who initiates the study. In that

70

N. Khalili-Mahani et al.

sense, it is not only useful for medical studies but also for studies about behavioral interventions in a population (Fig. 6).

Fig. 6. Summary of study elements to consider in the CONSORT framework

To have designed the study within the CONSORT framework helped us identify why our trial did not produce ‘high-quality’ effects (often judged by statistical power). The solid RCT protocol provided the initial frame whose limits could be clearly characterized and thoroughly reported (e.g., participant flowcharts, predefined outcomes (items 6a and 6b, Table 1), auxiliary analysis (item 8, Table 1) and baseline data (item 15, Table 2).) Introduce Flexibility to Minimize Attrition. As our case-study illustrated, human factors pressure the protocol frames. The CONSORT statement expects that all such deviations are recorded and reported (e.g., items 13–15, and 18–19, in Table 2). To adapt the protocol and methodology is provisioned in items 3b, 4b, 6b, 7b and 12b (Table 1, CONSORT Methods), and the decision to modify the protocol depends on research questions. In drug-studies, this flexibility does not exist (trials are go or no-go), but flexibility in ICT-related interventions provides insight into human factors. In our study, upon noticing the challenges of recruitment, we introduced several modifications. Namely, we deviated from the randomization protocol (if someone insisted to be in the group arm, we allowed them to.) Because we could not attract more than 33/78 participants to the study, and because only 16/33 agreed to be pre-tested, we showed flexibility to make it easier for them to complete the long trial (See Fig. 4). Therefore, we could gain better insight into older adults’ motivations for online engagement. While modification of the protocol reduced the quality of our RCT, it created an opportunity to retain an engaged group of participants who not only provided insightful knowledge about the needs of their age group but also extended the scope of the study to the next stage--community-based research and action. Conduct Mixed-Methods Research. Incorporating qualitative data acquisition methods in different stages of RCTs is essential for developing person-centred and tailored

Lessons from a COVID-Era Controlled Trial of Online Learning

71

interventions for the well-being of individuals and their communities. In this case-study, soliciting comments from those who refused to participate and seeking critical feedback at exit interviews confirmed ambivalence about cognitive and physical exams, and elicited concerns with ageism. This feedback helped us improve communication with participants and adjust the protocol to mitigate attrition.

7 Conclusions Conducting RCTs that involve ICT-based interventions are complex due to various human factors which are hard to control. This raises the question whether standards for high-quality RCTs can be applied to user studies that involve ICTs or assistive technologies for health interventions. To answer this question, we critically reviewed one of our own studies to underline the advantages of conducting research within a careful RCT framework, while also accommodating for flexibility in study design and interventions. In this case-study, we have illustrated the challenges of conducting an RCT involving ICTs, under complex and uncontrollable conditions (caused by a Pandemic forcing the research to take place outside a controlled environment, and human factors). Our casestudy adds to a large body of evidence aware of the difficulty of conducting reproducible and high-quality RCTs. Nevertheless, we illustrate that following the principles of RCT, and using the CONSORT framework, provide a template for rigorous data collection, enabling researcher to document and learn from transactional dynamics that demand flexibility. The limitations of our study are obvious: this is one case study, among the community-dwelling, digitally literate older adults, with access to costly equipment and data plans, enabling them to take online digital photography classes, and discuss them online. That said, the complexity of this interdisciplinary study design, and the multimodal mixed methodologies used for data collection and reporting present a ‘worse case’ scenario that further underlines the effectiveness of CONSORT in framing such flexible research. Against this frame, it become easier to document and accumulate new knowledge from tension points (in recruitment and randomization) and adaptive strategies that are necessary to conduct the study. While this study focused on older adults, the research strategy proposed here strives for person-centered and ecological ICT research and can be applied to all ages and across various user-studies. Interventions that target social and emotional well-being of individuals, need to be flexible but carefully framed.

References 1. Fiorini L., et al.: On the use of assistive technology during the COVID-19 outbreak: results and lessons learned from pilot studies. Sensors 22(17) (2022). https://doi.org/10.3390/s22 176631 2. Li Y., et al.: Cognitive decline and poor social relationship in older adults during COVID-19 pandemic: can information and communications technology (ICT) use helps? BMC Geriatrics 22(1). (2022). https://doi.org/10.1186/s12877-022-03061-z

72

N. Khalili-Mahani et al.

3. Wilson, J., Heinsch, M., Betts, D., Booth, D., Kay-Lambkin, F.: Barriers and facilitators to the use of e-health by older adults: a scoping review. BMC Publ. Health 21(1), 1556 (2021). https://doi.org/10.1186/s12889-021-11623-w 4. Buyl, R., Beogo, I., Fobelets, M., Deletroz, C., Van Landuyt, P., Dequanter, S., et al.: e-Health interventions for healthy aging: a systematic review. Syst. Rev. 9(1), 128 (2020). https://doi. org/10.1186/s13643-020-01385-8 5. Fernández-Ardèvol, M., Rosales, A., Morey, F.: Methods matter: assessment of the characteristics of a sample to analyze digital practices and social connectedness in later life. In: Gao, Q., Zhou, J. (eds.) HCII 2020. LNCS, vol. 12209, pp. 58–68. Springer, Cham (2020). https:// doi.org/10.1007/978-3-030-50232-4_5 6. Nichol, A.D., Bailey, M., Cooper, D.J.: Challenging issues in randomised controlled trials. Injury 41, S20–S23 (2010). https://doi.org/10.1016/j.injury.2010.03.033 7. Deaton, A., Cartwright, N.: Understanding and misunderstanding randomized controlled trials. Soc. Sci. Med. 210, 2–21 (2018). https://doi.org/10.1016/j.socscimed.2017.12.005 8. Khalili-Mahani, N., Sawchuk, K.: Double-bind of recruitment of older adults into studies of successful aging via assistive information and communication technologies: mapping review. JMIR Aging 5(4), e43564 (2022). https://doi.org/10.2196/43564 9. Mubarak, F., Suomi, R.: Elderly forgotten? Digital exclusion in the information age and the rising grey digital divide. INQUIRY: J. Health Care Org. Provision Financ. 59 (2022). https:// doi.org/10.1177/00469580221096272 10. Low, S.T.H., Sakhardande, P.G., Lai, Y.F., Long, A.D.S., Kaur-Gill, S.: Attitudes and perceptions toward healthcare technology adoption among older adults in Singapore: a qualitative study. Front. Publ. Health 9 (2021). https://doi.org/10.3389/fpubh.2021.588590 11. McDonough, C.C.: The effect of ageism on the digital divide among older adults. Gerontol. Geriatric Med. 2(1), 1–7 (2016). https://doi.org/10.24966/ggm-8662/100008 12. Md Fadzil, N.H., et al.: A scoping review for usage of telerehabilitation among older adults with mild cognitive impairment or cognitive frailty. Int. J. Environ. Res. Publ. Health 19(7) (2022). https://doi.org/10.3390/ijerph19074000 13. Kruse, C., Fohn, J., Wilson, N., Nunez Patlan, E., Zipp, S., Mileski, M.: utilization barriers and medical outcomes commensurate with the use of telehealth among older adults: systematic review. JMIR Med. Inform. 8(8) (2020). https://doi.org/10.2196/20359 14. Velayati, F., Ayatollahi, H., Hemmat, M.: A systematic review of the effectiveness of telerehabilitation interventions for therapeutic purposes in the elderly. Methods Inf. Med. 59(02/03), 104–109 (2020). https://doi.org/10.1055/s-0040-1713398 15. Hebblethwaite, S.: The (in)visibility of older adults in digital leisure cultures. In: Carnicelli, S., MacGillivray, D., McPherson, G. (eds.) Digital Leisure Cultures: Critical Perspectives, pp. 94–106. Routledge, London (2017) 16. Ivan, L., Hebblethwaite, S.: Grannies on the net: grandmothers’ experiences of Facebook in family communication. Roman. J. Commun. Publ. Relat. 18(1) (2016). https://doi.org/10. 21018/rjcpr.2016.1.199 17. Ivan, L., Cutler, S.J.: Ageism and technology: the role of internalized stereotypes. Univ. Tor. Q. 90(2), 127–139 (2021). https://doi.org/10.3138/utq.90.2.05 18. Marston, H.R., Genoe, R., Freeman, S., Kulczycki, C., Musselwhite, C.: Older adults’ perceptions of ICT: main findings from the technology in later life (TILL) study. Healthcare (Basel) 7(3) (2019). https://doi.org/10.3390/healthcare7030086 19. Nimrod, G.: Technostress: measuring a new threat to well-being in later life. Aging Ment. Health. 22(8), 1080–1087 (2018). https://doi.org/10.1080/13607863.2017.1334037 20. Khalili-Mahani, N., Sawchuk, K., Elbaz, S., Hebblethwaite, S., Timm-Bottos, J.: A socialmedia study of the older adults coping with the COVID-19 stress by information and communication technologies. In: Gao, Q., Zhou, J. (eds.) HCII 2022. LNCS, vol 13330, pp. 346–64. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-05581-2_26

Lessons from a COVID-Era Controlled Trial of Online Learning

73

21. Sawchuk, K.: Tactical mediatization and activist ageing: pressures, push-backs, and the story of RECAA. MediaKultur. 54, 47–64 (2013) 22. Birkland, J.L.H.: Understanding older adult technology use: an introduction to the ICT user typology. Gerontechnology, pp. 1–7 (2019) 23. Fernández-Ardèvol, M., et al.: Methodological strategies to understand smartphone practices for social connectedness in later life. In: Zhou, J., Salvendy, G. (eds.) HCII 2019. LNCS, vol. 11593, pp. 46–64. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-22015-0_4 24. Begg, C., Cho, M., Eastwood, S., Horton, R., Moher, D., Olkin, I., et al.: Improving the quality of reporting of randomized controlled trials. CONSORT Statement JAMA 276(8), 637–639 (1996). https://doi.org/10.1001/jama.276.8.637 25. Moher, D., Schulz, K.F., Altman, D.G.: The CONSORT statement: revised recommendations for improving the quality of reports of parallel group randomized trials. BMC Med. Res. Methodol. 1(1), 2 (2001). https://doi.org/10.1186/1471-2288-1-2 26. Moher, D., Hopewell, S., Schulz, K.F., Montori, V., Gøtzsche, P.C., Devereaux, P.J., et al.: CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. BMJ 340, c869 (2010). https://doi.org/10.1136/bmj.c869 27. Schulz, K.F., Altman, D.G., Moher, D.: CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ 340, c332 (2010). https://doi.org/10.1136/ bmj.c332 28. Erickson, J., Johnson, G.M.: Internet use and psychological wellness during late adulthood. Can J Aging. 30(2), 197–209 (2011). https://doi.org/10.1017/s0714980811000109 29. Cajita, M.I., Whitehouse, E., Budhathoki, C., Hodgson, N.: Association between Internet use and decision-making preference in older adults. Gerontechnology 14(2) (2016). https://doi. org/10.4017/gt.2016.14.2.008.00 30. Heo, J., Chun, S., Lee, S., Lee, K.H., Kim, J.: Internet use and well-being in older adults. Cyberpsychol. Behav. Soc. Netw. 18(5), 268–272 (2015). https://doi.org/10.1089/cyber.2014. 0549 31. Silva, P., Delerue Matos, A., Martinez-Pecino, R.: Confidant network and quality of life of individuals aged 50+: the positive role of internet use. Cyberpsychol. Behav. Soc. Netw. 21(11), 694–702 (2018). https://doi.org/10.1089/cyber.2018.0170 32. Khalaila, R., Vitman-Schorr, A.: Internet use, social networks, loneliness, and quality of life among adults aged 50 and older: mediating and moderating effects. Qual. Life Res. 27(2), 479–89 (2018). https://doi.org/10.1007/s11136-017-1749-4 33. James, B.D., Boyle, P.A., Yu, L., Bennett, D.A.: Internet use and decision making in community-based older adults. Front Psychol. 4, 605 (2013). https://doi.org/10.3389/fpsyg. 2013.00605 34. O’Rourke, H.M., Sidani, S.: Definition, determinants, and outcomes of social connectedness for older adults: a scoping review. J Gerontol. Nurs. 43(7), 43–52 (2017). https://doi.org/10. 3928/00989134-20170223-03 35. O’Rourke, H.M., Collins, L., Sidani, S.: Interventions to address social connectedness and loneliness for older adults: a scoping review. BMC Geriatr. 18(1), 214 (2018). https://doi.org/ 10.1186/s12877-018-0897-x 36. Myhre, J.W., Mehl, M.R., Glisky, E.L.: Cognitive benefits of online social networking for healthy older adults. J. Gerontol. B Psychol. Sci. Soc. Sci. 72(5), 752–760 (2017). https://doi. org/10.1093/geronb/gbw025 37. Ballesteros, S., Kraft, E., Santana, S., Tziraki, C.: Maintaining older brain functionality: a targeted review. Neurosci. Biobehav. Rev. 55, 453–477 (2015). https://doi.org/10.1016/j.neu biorev.2015.06.008

74

N. Khalili-Mahani et al.

38. Kastner, M., Cardoso, R., Lai, Y., Treister, V., Hamid, J.S., Hayden, L., et al.: Effectiveness of interventions for managing multiple high-burden chronic diseases in older adults: a systematic review and meta-analysis. Can. Med. Assoc. J. 190(34), E1004–E1012 (2018). https://doi. org/10.1503/cmaj.171391 39. Thomas, E., et al.: Physical activity programs for balance and fall prevention in elderly. Medicine 98(27) (2019). https://doi.org/10.1097/md.0000000000016218 40. Daskalopoulou, C., Stubbs, B., Kralj, C., Koukounari, A., Prince, M., Prina, A.M.: Physical activity and healthy ageing: a systematic review and meta-analysis of longitudinal cohort studies. Ageing Res. Rev. 38, 6–17 (2017). https://doi.org/10.1016/j.arr.2017.06.003 41. Beaudart, C., et al.: Nutrition and physical activity in the prevention and treatment of sarcopenia: systematic review. Osteoporos. Int. 28(6), 1817–1833 (2017). https://doi.org/10.1007/s00 198-017-3980-9 42. Smith, S.M., Wallace, E., Clyne, B., Boland, F., Fortin, M.: Interventions for improving outcomes in patients with multimorbidity in primary care and community setting: a systematic review. System. Rev. 10(1) (2021). https://doi.org/10.1186/s13643-021-01817-z 43. Lear, S.A., et al.: Assessment of an interactive digital health–based self-management program to reduce hospitalizations among patients with multiple chronic diseases. JAMA Netw. Open 4(12) (2021). https://doi.org/10.1001/jamanetworkopen.2021.40591 44. Choi, N.G., Pepin, R., Marti, C.N., Stevens, C.J., Bruce, M.L.: Improving social connectedness for homebound older adults: randomized controlled trial of tele-delivered behavioral activation versus tele-delivered friendly visits. Am. J. Geriatr. Psychiatry 28(7), 698–708 (2020). https://doi.org/10.1016/j.jagp.2020.02.008 45. Fink, D.S., Keyes, K.M.: Wrong Answers. Systems Science and Population Health, pp. 25–36 (2017) 46. Recruiting Older Adults into Research (ROAR) Toolkit: NIH National Institute on Aging (NIA) (2015). https://www.nia.nih.gov/health/recruiting-older-adults-research-roar-toolkit 47. McHenry, J.C., Insel, K.C., Einstein, G.O., Vidrine, A.N., Koerner, K.M., Morrow, D.G.: Recruitment of older adults: success may be in the details. Gerontologist 55(5), 845–853 (2015). https://doi.org/10.1093/geront/gns079 48. Schilling, I., Gerhardus, A.: Methods for involving older people in health research-a review of the literature. Int. J. Environ. Res. Publ. Health 14(12) (2017). https://doi.org/10.3390/ije rph14121476 49. Carnicelli, S., McGillivray, D., McPherson, G.: Digital leisure cultures (2016) 50. Khalili-Mahani, N., De Schutter, B.: Affective game planning for health applications: quantitative extension of gerontoludic design based on the appraisal theory of stress and coping. JMIR Serious Games 7(2), e13303 (2019). https://doi.org/10.2196/13303 51. Genoe, R., Kulczycki, C., Marston, H., Freeman, S., Musselwhite, C., Rutherford, H.: Eleisure and older adults: findings from an international exploratory study. Ther. Recreat. J. 52(1), 1–18 (2018). https://doi.org/10.18666/trj-2018-v52-i1-8417 52. Loos, E., Kaufman, D.: Positive impact of exergaming on older adults’ mental and social well-being: in search of evidence. In: Zhou, J., Salvendy, G. (eds.) ITAP 2018. LNCS, vol. 10927, pp. 101–112. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-92037-5_9 53. Baez, M., Nielek, R., Casati, F., Wierzbicki, A.: Technologies for promoting social participation in later life. Ageing Digit. Technol. 285–306 (2019) 54. Waycott, J., Vetere, F., Ozanne, E.: Building social connections: a framework for enriching older adults’ social connectedness through information and communication technologies. Ageing Digit. Technol. 65–82 (2019) 55. Chidiac, M., Ross, C., Marston, H.R., Freeman, S.: Age and gender perspectives on social media and technology practices during the COVID-19 pandemic. Int. J. Environ. Res. Publ. Health 19(21) (2022). https://doi.org/10.3390/ijerph192113969

Older Adults’ Perceptions and Use of Digital Engagement Channels: In the Case of Automobile Insurance Services Chaiwoo Lee(B) , Lisa D’Ambrosio, and Joseph F. Coughlin Massachusetts Institute of Technology AgeLab, Cambridge, MA 02139, USA [email protected]

Abstract. Online self-service applications are becoming increasingly available across different consumer-facing industries. While online channels offer potential benefits including cost savings, efficiency and convenience, their usage remains limited among older adults, as well as in service domains where complex decision making may be required to carry out transactions. In this study, a total of 120 adults 50 years of age and older participated in a set of research groups to explore attitudes toward, experiences with, and expectations regarding online self-service applications for automobile insurance. Study participants’ responses to a questionnaire, a use case walkthrough task, and group discussions indicated that phone/calling was the dominant channel for interacting with their automobile insurance company. People were mixed on their acceptance of online channels for insurance inquiries and transactions. Many believed that while online channels would offer convenience and speed, there were concerns about their use including tasks being complicated and the possibilities for making mistakes with great consequences. Results also suggested that prior experiences with or exposure to existing online channels, as well as general trust in technology, had a positive impact on future likelihood of using online channels for insurance. This further confirms the link between past experiences and future adoption intentions that was demonstrated by prior research in different technology domains, and suggests that providing and building experience may be key to increasing adoption and engagement. Implications regarding the interplay between technology-enabled online tools and human support are also discussed. Keywords: Online Interactions · Technology Adoption · Technology-Enabled Services · Risk Perception · User Experience

1 Introduction The growth of the internet and mobile technologies allow people to access a wide range of information and complete a variety of different tasks from anywhere at any time. People can shop online to have items delivered to their homes or to a locker or a local store, ready for pickup. Whereas once depositing a check into a bank account required a trip to the bank – or at least an ATM – people now can snap a photo with a mobile phone to deposit it. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 75–87, 2023. https://doi.org/10.1007/978-3-031-34866-2_6

76

C. Lee et al.

With developments in information technologies, as well as improvements in broadband access and availability, an increasing number of self-service applications are available in the digital, online environment through websites and mobile applications. Across industries, leveraging such technologies to offer consumers a self-service model is one means to tackle increases in labor and service delivery costs and to improve efficiency. For example, companies may require fewer employees to fulfill or complete certain kinds of tasks (e.g., handling customer inquiries or taking orders or payments). This has resulted in an increase in availability and usage of self-service technologies, which are defined as “technological interfaces that enable customers to produce a service independent of direct service employee involvement” [1]. Online self-service channels may also offer advantages beyond simply cost savings and efficiency. Such systems may help to increase customer satisfaction, as customers can do what they need to do on their own schedules. Once such assets are built, customers can transact their business directly through a website or app, ideally requiring little to no support from a sales associate or customer service representative. Customers can input their information directly into the system, reducing the need for an employee and reducing the chance that incorrect or inaccurate information might be entered. Customers can look for and get answers to questions they have at any point in time and connect more easily with companies when they do have issues. Thus, the development of digital and online channels can result in greater efficiencies for both companies and consumers. Customers too may realize savings through such systems such as reduced costs in transactions (e.g., no stamps, paper envelopes, etc.), incentives to conduct business on line from companies (e.g., a discount for going paperless), or through direct savings from a company (e.g., a company passing on some of savings from reduced human capital costs on to customers). Companies can design their digital channels with varying degrees of customer interaction with human service agents. Consumers’ experiences and engagements with companies can range from: 1) a full-service model, where service personnel deliver services for an individual’s needs; 2) an assisted model, in which a consumer explores given information and available options with some interactions with service personnel that assist the process; to 3) a completely self-service model, where the service experience is primarily delivered and produced by a consumer with little or no assistance from service personnel. While digital channels have become ubiquitous, not all consumers use them. The adoption of such channels is essentially a question of technology adoption: under what circumstances and in what cases do different people use digital channels to conduct their business? For example, different tasks that consumers are trying to accomplish might be better suited for degrees of automation; consumers might need little human assistance when paying a bill online, but they might seek to interact with service personnel before making a major purchase online, such as booking a vacation. Previous research on new technology adoption identified factors spanning demographic and socio-economic characteristics, past experiences, expected usefulness and ease of use, and perceived risk [2–4]. Age is another factor that may have an impact on people’s assessments of a technology’s ease of use and its usefulness, in turn affecting the likelihood of the technology being adopted.

Older Adults’ Perceptions and Use of Digital Engagement Channels

77

Positive drivers of digital self-service technologies include performance, feelings of control, speed, ease of use and perceived usefulness [5]. Similarly, customer satisfaction with self-service technologies is driven by cost and time savings, convenient access, compatibility, customization and ease of use [6, 7]. Another study has also found that a blend of trust in the channel and convenience and the efficacy of collecting information were important in consumers’ stated preferences for using an online versus traditional channel [8]. Similarly, while the specific types of transactions and interactions varied across studies, a significant effect of perceived usefulness on acceptance has been observed consistently in the context of digital self-service interactions [3, 9, 10]. However, frustration and uncertainty around problems that might occur – for example, difficulties with getting access to service personnel and lack of personalized assistance – while using selfservice technologies are often perceived as drawbacks of the convenience, accessibility and speed offered by them. Frustration, stress and even anxiety may result in avoidance and undermine consumer satisfaction [5]. The insurance industry is one example where companies place a high priority on managing and facilitating customer interactions and inquiries, and one where customers have frequent needs for information search and transaction requests. While many insurance companies have developed websites and apps to facilitate customers’ interactions with them, many people still do not make use of these channels. The purpose of this project is to understand people’s experiences with automation regarding their insurance business, primarily automotive, and how to foster the adoption and use of digital self-service platforms, particularly among older adults who are primary consumers of automotive insurance but may be less accepting of digital channels. More specifically, the work seeks to understand the different tasks for which adults ages 50 and older may be more or less comfortable using online self-service channels, what barriers they perceive to the use of these channels, and what kinds of incentives might motivate increased usage.

2 Data Collection To examine these questions about older consumers’ preferences for using digital channels to conduct insurance business, a mixed-methods study was conducted with adults ages 50 or older. All participants reported that they: 1) were the primary decision maker for automobile and home insurance in their household; 2) were licensed drivers who had an automobile insurance policy on the vehicle they drove most frequently; 3) had made a change to their automobile insurance in the 12 months prior to the study; 4) had internet access in their home; and 5) nor anyone else in their household currently or previously worked for a home and/or automobile insurance company, as an insurance sales agent, as an insurance or claims adjuster, or as an actuary. Eligibility was determined through telephone screening, during which participants also answered questions about their experience with technology, how easy they found it to learn to use new technologies, and the frequency with which they used the internet to pay bills and to make purchases. A total of 120 participants who met the screening criteria were recruited by Plaza Research. Based on responses to the telephone screening, six groups were created across three ages groups (ages 50–59, 60 to 69, and 70 +) and two levels of technology comfort

78

C. Lee et al.

(high and low). Each group included 20 people and was evenly balanced in terms of gender. A majority of the study participants were married, and a majority lived with at least one other person in the household. Most lived in the community and were relatively healthy by their own self-reports. They tended to have more than one car in the household, own their cars and own their homes. Just over half of the study participants (50.8%) reported that they were the sole decision maker when it came to automobile insurance for their vehicle; an additional 21.7% said they shared in the decision-making process for automobile insurance for their vehicle, but that their preferences were weighed more heavily. Just under a quarter of the participants (24.2%) said that they shared the decisionmaking equally with someone else. The research groups were held at Plaza Research’s facilities in Marlton, New Jersey, near Philadelphia, Pennsylvania, on August 15-17, 2018. Once participants arrived at the research facility, they completed human subjects consent forms approved by MIT’s Committee on the Use of Humans as Experimental Subjects (COUHES) and filled out a questionnaire about their attitudes and experiences using digital channels, particularly with respect to insurance. The questionnaire also included items about: general purchase preferences; communication preferences; technology attitudes and usage; household finance management behaviors; types and channels of contact with insurance company; perceptions regarding their insurance company’s online channels; thoughts and behaviors regarding online privacy and data security; and demographics. Following completion of the questionnaires, the groups took part in a researcher-led use case walkthrough task. In this task, each respondent read a scenario describing a reason for contacting their automobile insurer. Each participant was assigned one of four scenarios: having a billing question; having a policy question; adding or replacing a vehicle on the policy; or filing a claim. They then answered a series of questions as they described how they would approach and solve the given case, including their preferred channel to contact their insurer; the advantages and disadvantages of contacting by telephone or calling to handle this case; and the advantages and disadvantages of going online to handle this case. They were also asked to discuss the general advantages and disadvantages they saw for contacting their automobile insurer online to accomplish any of the kinds of tasks. Once the task was complete, the researchers facilitated a brief, general discussion with participants about their opinions and use of digital channels for insurance business, as well as other kinds of activities. Each research group session lasted about two hours and was led by two MIT AgeLab researchers. Upon completion of the group, participants were thanked for their participation and were given US$100 as compensation.

3 Results 3.1 Openness to Using Online Channels with Insurers Study participants contacted their automobile insurer for a variety of different reasons in the 12 months prior to the study. People reported an average of 2.8 reasons for why they contacted their automobile insurer (the range was from 1 to 6, with the median equal to 3 reasons and the mode equal to 2 reasons) (note that the measure is for reasons to contact the company, not actual number of contacts). Study participants were most

Older Adults’ Perceptions and Use of Digital Engagement Channels

79

likely to contact their insurer with a policy question, to initiate a new policy or to change over an existing one, or to make a change of some other sort to their policy. For each reason for contact with their automobile insurance company, a majority of participants reported using the phone. In the study sample, people who removed a driver from their policy did so universally by phone, followed by asking policy questions (86.4% by phone), making policy changes (83.3%), filing a claim (79.2%), getting a new policy or changing over an existing one (78.9%), or adding a driver (75%). Online contact was typically the second most frequently reported channel used to reach out to automobile insurers, followed by in person contacts. Participants were more likely to say they had used an online channel when they changed an address (40% used online), changed companies or providers (31.8%), or added a driver (25%); just over one in five (20.8%) of people who said they had filed a claim reported doing so online. Study participants were also asked about their prospective online contacts with the automobile insurance company. Specifically, people were asked to consider how likely they would be to contact their automobile insurance company online to make policy changes or to file a claim, and they were asked how easy they thought these tasks would be to do online. Figure 1 displays the likelihood that people thought they would contact their automobile insurance company online for these tasks.

Likelihood of doing automobile insurance tasks online over the next year Very unlikely

Somewhat unlikely

Making policy change (n=117)

Neither

37.6%

Filing claims (n=99)

17.9%

53.5%

0%

20%

Somewhat likely

22.2%

16.2%

40%

60%

Very likely

13.7% 8.5%

21.2%

80%

8.1% 1.0% 100%

Fig. 1. Likelihood of doing different automobile tasks online

Consistent with people’s reports of their past interactions with their automobile insurance companies, they were less likely to report that they would be likely to file a claim with their automobile insurance company online compared with making a policy change. A majority of people thought they would be very unlikely to file a claim online, and 69.7% said they would be very or somewhat unlikely to do so. In contrast, just over a third – 37.6% – of people said that they would be very unlikely to make a policy change online, and a little over half – 55.5% – said they would be very or somewhat unlikely to do so. People who reported having had at least one previous contact with their automobile insurance company online – for any reason – were somewhat more likely than people who had never had any online contacts with their automobile insurer

80

C. Lee et al.

to report that they would contact their automobile insurer online to make a policy change (F = 2.958, N = 82, p = .089), but they were no more or less likely to report a claim online. When it came to which tasks people thought would be easier to do online, people were somewhat more inclined to say that making a policy change would be easier to do online. As shown in Fig. 2, nearly two-fifths (39.5%) said they thought making a policy change online would be moderately or extremely easy, compared with 33.1% who thought filing a claim online would be moderately or extremely easy.

Perceived ease of doing automobile insurance tasks online Not at all easy

Slighly easy

Making policy change 8.8% (n=114)

Filing claims 10.7% (n=112) 0%

Somewhat easy

23.7%

28.1%

25.0%

20%

Moderately easy

22.8%

31.3%

40%

Extremely easy

60%

17.9%

16.7%

15.2%

80%

100%

Fig. 2. Perceived ease of doing different automobile insurance tasks online

3.2 Using Online Channels for Insurance – Advantages and Disadvantages Participants were asked to select potential advantages and disadvantages that they would see to going online to accomplish automobile insurance tasks. These questions were asked following the completion of the use case walkthrough tasks, after participants had thought through the decisions and actions that they would take in order to do given tasks online. When asked about the advantages, the top five selections included the ability to get the task done from anywhere (45.8% of all participants), getting the task done immediately (41.7%), being comfortable using the internet (35.0%), getting tasks done quickly/fast (21.7%), and ease of getting the task done (19.2%) as shown in Table 1. In short, people who were comfortable using the internet saw this as an advantage were they going to do other kinds of tasks online. Other highly selected reasons to go online included getting answers to questions quickly/right away (17.5%) and the lack of cost associated with doing tasks online (13.3%).

Older Adults’ Perceptions and Use of Digital Engagement Channels

81

Table 1. Advantages and disadvantages to using online channels to accomplish automobile insurance tasks (N = 120) Advantages

%

Disadvantages

%

I can get task done from anywhere

45.8

Hard to get answers to questions

39.2

Task gets done immediately

41.7

Worried I will make a mistake

39.2

I’m comfortable using the internet

35.0

Can’t remember my passwords/login information

30.8

Getting tasks done is fast/quick

21.7

Insurance is so complicated that in person or calling would be easiest

27.5

Easy to get task done

19.2

Looking for all of the information I need would take too long

15.8

Get answers to questions quickly/right away

17.5

Not how I usually do these kinds of things

15.8

Doesn’t cost anything

13.3

My data and personal information are not secure

14.2

Website is reliable

12.5

Website is confusing

13.3

This is how other people these days do these kinds of things

10.0

This task cannot be done online

11.7 10.0

Know that task gets done correctly

9.2

Insurance is too complicated to do online

Don’t need to have policy details handy

9.2

Delay in when task gets done

8.3

This is how I usually do these kinds of things

9.2

Would feel overwhelmed dealing with insurance issues so going in person or called would be easiest

8.3

Easy to get answers to questions

8.3

Don’t know how to do task online

6.7

I feel in control

8.3

Doing business on the web is not secure

5.8

Know my information is secure

2.5

Website is hard to access

5.0

Insurance is so complicated that going online would be easiest

1.7

Website is not personalized for my needs

5.0

I would feel overwhelmed dealing with insurance issues so going online would be easiest

0.8

I’m not comfortable using the internet

5.0

Would be feeling upset or emotional about with insurance issues so going online would be easiest

0.0

I don’t understand how insurance works

3.3

Website is not reliable

2.5

Would be feeling upset or emotional dealing with insurance issues so in person or calling would be easiest

2.5

A comparison across subgroups (age, gender and technology comfort) revealed a general consistency, with only few statistically significant group differences that emerged among people’s choices of the reasons they saw as advantages to going online. People ages 60 to 69 were more likely to select getting tasks done is fast/quick as an advantage relative (35.0% of subgroup) to people ages 50 to 59 (12.5%) and those ages 70 and older (17.5%). People ages 50 to 59 were more likely to select getting answers to questions

82

C. Lee et al.

quickly/right away as an advantage (35.0% of subgroup) relative to the older groups (5% of people ages 60–69 and 12.5% of people ages 70+). People in the lower technology comfort groups were more likely to select that the website is reliable and knowing that the task got done correctly compared to people in the higher comfort groups; people in the higher technology comfort groups were more likely to see not needing to have policy details (e.g., policy number) handy as an advantage to going online, compared with the people in the lower comfort groups. Prior experience dealing with one’s automobile insurer online also had some impact on people’s selections; people who had had prior experiences were more likely to cite their comfort with the internet as an advantage and more likely to select that it was easy to get the task done online, compared with people who had not previously used an online channel to address an automobile insurance question or issue. Similarly, when asked to select the top disadvantages to using online channels to accomplish automobile insurance tasks, there was a great deal of consistency in the top reasons people cited not to go online, as also shown in Table 1. The key reasons people identified for not wanting to go online to accomplish automobile insurance business included that it was hard to get answers to questions (39.2%), worry about making a mistake (39.2%), not remembering passwords or login information (30.8%), and that insurance was so complicated that it would be easier to call or to go in person (27.5%). Other disadvantages selected included thinking that looking for all of the information needed would take too long (15.8%), not usually having done these kinds of things (15.8%), concerns over the security of personal information and data (14.2%), and confusing websites (13.3%). Similar to the advantages, only few significant differences in responses among subgroups emerged for disadvantages selected. Younger people were more likely than older people to cite not remembering passwords or login information as a disadvantage (50% of people ages 50–59, 25% of people ages 60–69, and 17.5% of people ages 70 +); people in the higher technology comfort groups were also more likely to select this reason as a disadvantage (35% of subgroup) compared with people in the lower technology comfort groups (26.7%). Compared with people in the higher technology comfort, people in the lower technology comfort group were more likely to be worried about making a mistake or to select that they would feel so overwhelmed dealing with insurance issues that it would be easier to go in person or call. 3.3 Perceived Risks In line with only a small subset of participants choosing data security concerns as a potential disadvantage to doing automobile insurance tasks online (see Sect. 3.2), participants generally indicated a very low level of worry regarding online interactions with automobile insurance companies. When asked about how risky they thought online interactions with their insurance company would be, 63.9% said they were not worried at all (see Table 2). Men were somewhat more likely to report an average rating of more worry than women were (F = 3.676, N = 119, p = .058). There were no differences in average worry score by age category, technology comfort level, or past experience with conducting an automobile insurance transaction online.

Older Adults’ Perceptions and Use of Digital Engagement Channels

83

Table 2. Concerns about risks in different online interactions Question

Answers

n

Worry about risks of online interactions with automobile insurance company

Not at all worried: 63.9% Slightly worried: 21.0% Somewhat worried: 10.1% Moderately worried: 4.2% Extremely worried: 0.8%

119

Concerned about privacy while using the internet in general

Not concerned at all: 1.7% A little concerned: 17.8% Somewhat concerned: 13.6% Moderately concerned: 27.1% Extremely concerned: 39.8%

118

Worried about my personal information being hacked or stolen

Strongly disagree: 1.7% Somewhat disagree: 4.2% Neither agree nor disagree: 20.2% Somewhat agree: 42.9% Strongly agree: 31.1%

119

People’s confidence around online interactions with their automobile insurer did not necessarily reflect or translate to confidence in the safety of the online activities as a whole. As shown in Table 2, about two-thirds (66.9%) were moderately or extremely concerned about privacy when they used the internet, and 74.0% agreed that they were worried about their personal information being hacked or stolen through online interactions. Demographic differences in terms of concerns about online interactions generally were minimal. Regression analysis further confirmed the lack of demographic impact on people’s risk perceptions around online interactions, both generally and with their insurance company. Several independent variables were used to fit regression models describing outcome variables regarding people’s concerns with online activities as described in the questions shown in Table 2. The independent variables included whether people had past online interactions with their automobile insurance company, general technology comfort and trust, household income, level of education, racial identity, gender, and age. Previous online interactions with one’s automobile insurer was a significant predictor only of people’s responses to concerns about their privacy while using the internet; it was not related to people’s concern level about having their information hacked or stolen online or, surprisingly, people’s perceptions of risk about online contacts with their insurer. The only consistently significant predictor of people’s attitudes, however, was their self-reported trust levels in technology in general. People with higher levels of trust were less likely to be concerned about the risks of online interactions with their insurer, less likely to be worried about having their information stolen or hacked online, and less likely to be worried about their privacy while using the internet.

84

C. Lee et al.

4 Discussions and Conclusion With the evolution of the internet and the adoption of mobile technologies, notably smartphones, people today can accomplish a wide range of tasks from wherever they are. These online self-service technologies have revolutionized how people communicate, how they work, how they shop, and how they manage their personal and financial matters. For an aging population, such technologies offer opportunities to connect and the potential for greater convenience around aging in place. Yet online channels have risks or costs associated with them as well; people perceive risks around data privacy and online security, and in some cases, such as insurance, they worry that poor online decisions could have long term repercussions. Online channels offer great convenience, but in domains such as insurance, many people may also seek comfort and confidence in knowing the choices they have made were the right ones for them. Encouraging people to move to a higher degree of digital engagement with their insurance companies may mean helping people to try to interact through an online channel for a lower stakes issue, as well as building in supports that provide people with the comfort and confidence in their online insurance activities. In this study that involved 120 adults 50 years of age or older, people were mixed on how easy they thought the process of accomplishing insurance tasks online would be, but, consistent with technology adoption models, prior experience or exposure to online channels had a positive impact: people who had previously had one or more online contacts with their automobile insurer were on average more likely to think that going online to make policy changes or to file a claim would be easier than people who had not had such experiences. Similarly, people in the higher technology comfort groups were generally more likely to believe that doing such tasks online would be easier, compared with people in the lower technology comfort groups. While there was little variation in the perceptions regarding the ease of doing insurance tasks online, these older study participants generally favored using the phone and calling to accomplish tasks over using an online channel. There are likely several reasons for this high degree of consensus. First, convenience and comfort underlie many of the key advantages people viewed of the different channels. Study participants perceived that both channels (telephone/calling and online) offered a relatively high degree of convenience. Online convenience revolved around: speed (tasks done quickly – not taking up too much of the participant’s time; tasks/changes in place immediately – speed of implementation); ease (easy to get tasks done); and access (task could be done anywhere). One of the key advantages of phone convenience revolved around ease (easy to get tasks done) and speed as well, but this speed had to do with insurer implementation (task done immediately), not the amount of time an individual might need to accomplish the task. The key distinction between the two channels, in terms of the advantages they offer to people based on the use case walkthrough task data, however, is that the phone also offers people comfort: feeling better if you talked to someone was the primary advantage of phone, and the ease of getting questions answered – meaning less uncertainty for people – is another dimension of comfort. The importance of comfort was apparent too in the boost that people in the file claim use case gave to “would be feeling upset or emotional about the task so calling would be easiest” relative to the selection of

Older Adults’ Perceptions and Use of Digital Engagement Channels

85

this reason in the other use cases. Similarly, comments in the discussion reinforced the comfort that some people experienced when they dealt with their issue via phone. Second, the key disadvantages of using the phone as the communications channel with the insurer were related to the frustrations of: having to deal with automated systems that took up more time; being put on hold or being hung up on; and generally taking too much time to accomplish given tasks. In short, the primary cost to consumers of using the phone channel is their own time. To the extent that older consumers – particularly ones who may be retired – have more available time, they may be more willing to use the telephone channel. In contrast to the phone channel, however, the key disadvantages to going online that people identified were more varied. They include not being able to get answers to questions (in turn perhaps necessitating a phone call), concern about making a mistake (which people did recognize could have significant implications for their lives), and, related to that concern, the belief that insurance is so complicated that it would be easier to use a different channel, and, more prosaically, having to remember yet another set of passwords and/or login information. The results also shed light on how perceptions and attitudes regarding acceptance of technology-enabled services may differ greatly by industry domains or task types. One key reason that people favored using the phone over online channels for automobile insurance was the nature of insurance itself. In group discussions, people noted that insurance mistakes could have far-reaching and catastrophic effects. As a result, a number of people expressed that insurance was too important not to have had some conversation with a person about it. Further, some people felt that in such conversations service representatives or agents were able to help them make the right choices, saving money and/or ensuring that they had the right kind of coverage for what they needed. The particularity of insurance is also suggested by the study participants’ responses to questions on risks and concerns regarding online activities more broadly. While participants generally showed a high level of concern regarding online privacy and identity theft, the level of worry regarding risks of online interactions with their automobile insurance company was generally very low. However, in both cases – trust in online interactions with insurers and trust in online interactions generally – people’s self-rated general level of trust in technology was found to be a significant determinant of worry/concerns, suggesting a need for a more detailed investigation into the role of past experiences and general technology attitudes on different types of online activities. The study results offer some direction for thinking about what may affect acceptance and behavior change among older consumers to realize the benefits that online self-service channels promise. First, prior online experiences tended to be positive for people, and, consistent with findings in [5], people who had had such experiences were more open to using online channels in the future. This tie between exposure to relevant past experience and future acceptance of technology has been demonstrated in other domains as well, including smart home systems (where short-term use of a smart home technology prototype was found to be significantly associated with an increase in willingness to incorporate technologies into daily tasks [11]) and automated vehicles (where drivers currently using driver assistance features were more accepting of higher-level

86

C. Lee et al.

automation such as full self-driving [12]). Insurers, as well as service providers and product developers in other consumer-facing industries, can think about ways to encourage customers to go online for any reason to interact with them and try related features as a means of initiating them into using their online channel. Further, supplementing the online channel with options to provide consumers with greater feedback around comfort making decisions may be useful. In this study, participants were more open to carrying out simpler transactions or information search using online channels, but hesitant to use them to do more complex tasks (e.g., choosing a particular policy or coverage) where human knowledge could be important. Similar findings were observed in a recent report on consumer adoption of artificial intelligence (AI), where consumers indicated having greater confidence in AI to perform rote tasks and less confidence in scenarios where AI would play key decision making roles, and where it was suggested that maintaining a more traditional or human interface may help to drive successful adoption of AI systems [13]. While study participants did not focus directly on this notion of providing feedback to consumers who use online platforms, tools like scheduling a call from a person to review changes or to be available to answer any questions or online chat with service representatives may be a means to provide people with the confidence that they have made the right decisions and not mistakes around their actions when they engage via online communications platforms. Finally, considerations and complexity regarding the interplay between technology capabilities and human support need to be better understood to describe consumers’ ideal interactions and engagements with different online systems. Future research could more generally explore the topic of incentives to better understand what kinds of inducements (e.g., financial compensation, human follow up, etc.) yield greater comfort around doing various tasks online, and pilot different options and interventions to determine which are more effective to support online engagement around different kinds of tasks, domains and consumers. Furthermore, as technologies and related services continue to change and advance, it would be important to note that consumers’ criteria for what constitutes a more or less complex issue or question, and what roles may be expected of technologies and humans, may change over time as well. Consumers themselves may also vary over time in their definitions for what constitutes a complex question; for example, an older consumer who has many years of experience as an automobile insurance customer may regard insurance policy decisions as much less complex than a younger consumer who is making an automobile insurance purchase for the first time might. As such, developing a more complete picture of consumers’ comfort and confidence in online communications platforms, and in their willingness to use these, will require a more comprehensive exploration including consumers of different ages and life stages. Acknowledgements. The authors gratefully acknowledge the support of The Hartford Financial Services Group, Inc. For this research.

Older Adults’ Perceptions and Use of Digital Engagement Channels

87

References 1. Meuter, M.L., Ostrom, A.L., Roundtree, R.I., Bitner, M.J.: Self-service technologies: Understanding customer satisfaction with technology-based service encounters. J. Mark. 64(3), 50–64 (2000) 2. Venkatesh, V., Morris, M.G., Davis, G.B., Davis, F.D.: User acceptance of information technology: toward a unified view. MIS Q. 27(3), 425–478 (2003) 3. Featherman, M.S., Miyazaki, A.D., Sprott, D.E.: Reducing online privacy risk to facilitate e-service adoption: the influence of perceived ease of use and corporate credibility. J. Serv. Mark. 24(3), 219–229 (2010) 4. Lee, C.: User-centered system design in an aging society: An integrated study on technology adoption. Doctoral thesis, Massachusetts Institute of Technology, Cambridge, MA (2014) 5. Johnson, D.S., Bardhi, F., Dunn, D.T.: Understanding how technology paradoxes affect customer satisfaction with self-service technology: the role of performance ambiguity and trust in technology. Psychol. Mark. 25(5), 416–443 (2008) 6. Meuter, M.L., Bitner, M.J., Ostrom, A.L., Brown, S.W.: Choosing among alternative service delivery modes: An investigation of customer trial of self-service technologies. J. Mark. 69, 61–83 (2005) 7. Yen, H.R.: An attribute-based model of quality satisfaction for internet self-service technology. Serv. Ind. J. 25, 641–656 (2005) 8. Choudhury, V., Karahanna, E.: The relative advantage of electronic channels: A multidimensional view. MIS Q. 32(1), 179–200 (2008) 9. Hoehle, H., Huff, S., Goode, S.: The role of continuous trust in information systems continuance. J. Comput. Inf. Syst. 52(4), 1–9 (2010) 10. Shamdasani, P., Mukherjee, A., Malhotra, N.: Antecedents and consequences of service quality in consumer evaluation of self-service internet technologies. Serv. Ind. J. 28(1), 117–138 (2008) 11. Cerino, L., FakhrHosseini, S., Lee, C., Lee, S.-H., Son, H., Coughlin, J.: Towards a more connected home: user attitudes and perceptions after an integrated home technology exposure. Proc. Hum. Fact. Ergon. Soc. Ann. Meet. 66(1), 1917–1921 (2022) 12. Lee, C., Seppelt, B., Reimer, B., Mehler, B., Coughlin, J.F.: Acceptance of vehicle automation: effects of demographic traits, technology experience and media exposure. Proc. Hum. Fact. Ergon. Soc. Ann. Meet. 63(1), 2066–2070 (2019) 13. MIT AgeLab: AI and Longevity – Consumer and Expert Attitudes toward the Adoption and Use of Artificial Intelligence, https://agelab.mit.edu/static/uploads/mit-agelab-ai-longevity_ wp-04-21-1357_ada.pdf (2021). Accessed 6 Jan 2023

Promoting Positive Emotions in Older Adults: A Self-help Relational Savoring e-Intervention Elisa Pancini1(B) , Francesca Pesce2 , Lucia Scuzzarella2 , and Daniela Villani1,3 1 Research Center in Communication Psychology, Department of Psychology,

Università Cattolica del Sacro Cuore, Milan, Italy [email protected] 2 Department of Psychology, Università Cattolica del Sacro Cuore, Milan, Italy 3 Research Unit in Digital Media, Psychology and Well-Being, Department of Psychology, Università Cattolica del Sacro Cuore, Milan, Italy

Abstract. Promoting positive emotions in older adults appears to be a path to support healthy aging and prevent mental disorders such as depression. To reach this goal, relational savoring (RS) appears to be a promising approach. RS is a specific type of savoring that emphasizes relational contents and consists in stimulating people to focus on positive relational experiences. This study aims to investigate the feasibility of a self-help relational savoring e-intervention on savoring beliefs and strategies and on positive and negative emotions in older adults. This study also evaluates the user experience and the perceived usefulness of the e-intervention. Eleven participants were involved and completed the e-intervention, consisting of two online RS exercises per week for three weeks, for a total of six exercises. To investigate participants’ changes in psychological dimensions (e.g., savoring beliefs and strategies and positive and negative emotions), online questionnaires were proposed before and after the intervention; to evaluate participants’ experiences, a qualitative questionnaire was proposed at the end of each exercise. Paired t-tests revealed significant improvements in savoring strategies and positive emotions. The online platform was positively evaluated, and most of the participants considered the training useful. The results of this feasibility study are promising and suggest that the intervention has the potential to promote savoring and positive emotions in older adults. Future research could involve a larger sample and include randomized controlled trials to evaluate its efficacy. Keywords: Older Adults · Relational Savoring · Internet Intervention · Positive Emotions

1 Introduction 1.1 Savoring Positive Emotions and Experiences Over the years, research on older people has increasingly focused on methods and strategies aimed at promoting positive aging. In particular, research has analyzed the shortand long-term outcomes of positive emotions. They have both preventive and therapeutic effects, strengthen the immune system, and motivate to implement flexible, adaptive © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 88–101, 2023. https://doi.org/10.1007/978-3-031-34866-2_7

Promoting Positive Emotions in Older Adults

89

behaviors and a healthier lifestyle [1]. Specifically, they act as a protective factor against adversities such as stress and depression [2]. Furthermore, they improve physical health, are linked to increased longevity, and may protect against age-related decline [3–8]. A promising approach aimed at promoting positive emotions is represented by savoring. Savoring can be considered a tool through which people can capitalize on positive emotions and enhance their well-being [9]. It is the ability to live, appreciate, and increase positive experiences in our lives [10–13]. Savoring can also be defined as a type of perceived control and regulation of positive emotions [14]. it has been defined as a mechanism strictly connected to the two theories proposed by Fredrickson, i.e., the broaden and build theory of positive emotions and the undoing hypothesis, as it allows to amplify and extend the benefits deriving from positive experiences and emotions [15–17]. Savoring can be divided into three temporal orientations: the first is to savor through positive reminiscence. In this case, the memory of positive events allows one to relive in the present the same positive emotions experienced in the past and to experience new ones. For example, thinking about a nice past holiday can elicit in the present the joy and enthusiasm experienced then, and there is the possibility to experience a new positive emotion such as satisfaction. The second is to savor the present moment. For instance, during a walk, it is possible to listen to the sounds, pay attention to the scents, feel the warmth of the sun on the skin, and appreciate the beauty of the sky and nature. Finally, the third temporal orientation involves savoring through the anticipation of an imminent positive event [11]. Indeed, it is possible to appreciate in the present the positive emotions elicited by the anticipation of a future positive experience; for example, the experience could be meeting a close friend or relative or eating tasty food [11]. Empirical evidence suggests the presence of individual differences in assessing one’s ability to savor positive affect (savoring beliefs): generally, in fact, people differ in the type and quantity of emotions they want to experience [11, 18, 19]. Savoring can be implemented through different strategies used to amplify and intensify the positive experience, such as sharing positive feelings with others, building positive memories, self-congratulation, getting absorbed in the moment, or counting gifts [11]. Numerous empirical studies suggests that savoring predicts better well-being, a greater number of positive emotions, positive reflections on life, and a lower intensity of negative emotions and depressive symptoms [14]. Furthermore, Smith et al. [20] confirmed that savoring interventions have a positive impact on quality of life, life satisfaction, psychological and physical well-being, interpersonal relationships, and a sense of meaning in life [21]. 1.2 Personal and Relational Savoring in Older Adults Savoring interventions can be particularly suitable for improving positive emotions in the older population. Thus, positive emotions and savoring play a fundamental role in old age. For example, research conducted by Smith and Bryant [22], which involved 266 adults aged between 55 and 94 years, showed how participants with a greater savoring ability can maintain high levels of life satisfaction even in poor health conditions. These data suggest that savoring can protect older people from age-related physical decline, poor health, and its harmful psychological effects [16, 22]. Furthermore, in

90

E. Pancini et al.

the older population, a greater ability to savor is connected to better perceived control and greater involvement in the activities performed [16, 17, 22, 23]. Thus, people who have good savoring skills can experience a greater number of positive emotions, which consequently have the power to motivate and involve them in more activities, reduce depressive symptoms, and increase longevity [7, 10, 16, 23–25]. In this way, savoring can therefore improve both the quality of life and the lifespan, and it may be particularly important for older people who are isolated or have limited mobility as they have fewer opportunities to participate in external activities [16]. Some recent studies have examined the efficacy of savoring interventions in the older population. For example, in a study carried out by Salces-Cubero et al. [26], 124 older adults aged between 60 and 89 were involved, and the results showed that savoring, compared to other approaches such as optimism, produces the strongest effects in increasing positive affect, happiness, and life satisfaction, as well as reducing negative affect. Moreover, the researchers observed that in older adults, a greater capacity for savoring is related to more happiness. For instance, the ability to savor positive experiences can protect older people from some harmful psychological effects caused by health problems. Thus, high levels of savoring may help older adults better manage stress caused by health problems by increasing the frequency or duration of positive emotions [11, 27–29]. In recent years, in addition to personal savoring, so-called relational savoring has also been developed. Relational savoring (RS) is a specific type of savoring that consists of stimulating people to focus on positive relational experiences and in remembering the moments in which they respond sensitively to others’ needs, thus emphasizing relational contents [30]. Thus, while personal savoring primarily aims at empowering personal positive emotions, relational savoring also seeks to improve interpersonal processes [31]. One of the main premises of relational savoring is that in everyday life, many positive moments shared with loved ones happen, but they are often not appreciated. Through RS, older people can value positive moments that are frequently overshadowed by everyday worries or negative experiences. Therefore, one purpose of relational savoring is to create opportunities to intentionally focus on the meaning of these shared positive experiences and appreciate them. A small number of RS randomized trials have provided evidence on the effectiveness of this technique in improving well-being in the short term. When compared to a personal savoring condition and a control condition, relational savoring significantly improves positive emotional state and relational satisfaction [30, 31]. Furthermore, in a study of 63 adults aged 60 to 90 years, the relational savoring group showed lower cardiovascular reactivity than the personal savoring group [32]. 1.3 Online Savoring for Older Adults Thanks to advances in technology and to the numerous trials that have been recently conducted, it is known that technology-based positive interventions are promising for assisting individuals in many aspects of well-being [33–35]. Positive technology (PT), that is the scientific use of technology to enhance the quality of the individual experience, appears to be effective in enhancing positive emotions and reaching socially isolated older adults or those with limited mobility [36, 37]. It is therefore possible to distinguish between three different categories of PT [36]: hedonic technologies, which aim

Promoting Positive Emotions in Older Adults

91

to induce pleasant experiences, eudaimonic technologies, which support self-realizing experiences, and social technologies, which aim to enhance social integration and connections between people and groups. Within the PT framework, internet-based interventions allow the possibility to sustain people’s well-being remotely both with and without the presence of a professional. Thus, for the older population with enough digital literacy about the use of mobile technology (smartphones and tablets) and internet access, the self-help approach represents an opportunity. According to Riva et al. [38], positive technology can therefore allow older people to take care of their health and well-being actively and to generate involvement in the activities carried out. As for personal online savoring interventions, only in one study involving 155 participants aged between 25 and 75 years, online savoring was used in integration with other positive psychology strategies to increase subjective well-being [39]. The online savoring group showed higher levels of positive emotions and life satisfaction and lower levels of negative affect, depressive symptoms, and perceived stress compared to the control group. Moreover, up to now, studies of relational savoring offered online to the older have not yet been conducted. 1.4 The Present Study Starting from these premises, we investigated the feasibility of a self-help relational savoring e-intervention with a small group of older adults. We assumed that (1) participants would increase their perceptions of their ability to enjoy positive events (savoring beliefs) and (2) would increase behaviors and thoughts that can amplify positive emotions (savoring strategies). We also hypothesized that (3) the intervention would increase positive affect and reduce negative affect. Moreover, we described the intensity of the emotions aroused by each exercise, and we qualitatively analyzed the participants’ comments made following each exercise. Finally, we evaluated the user experience and the perceived usefulness of the intervention.

2 Methods 2.1 Participants The sample was recruited through e-mails sent to facilities dedicated to the older adults, and also through word of mouth. In the present study, 11 Italian older adults aged between 65 and 95 years old were involved. The inclusion criteria were age over 65, fluency in the Italian language, and having a technological tool for internet access, while the exclusion criteria were the presence of neurological pathologies or dementia proven by a selfdeclaration. Demographic characteristics, including age, gender, geographical origin, educational level, employment status, marital status, and residence, were collected at baseline and reported in Table 1. Participants come from two regions of Northern Italy: eight from Lombardy and three from Emilia-Romagna. The continuous variable (age) is reported as mean and SD, whereas categorical variables are reported as frequencies and percentages.

92

E. Pancini et al. Table 1. Participants’ characteristics. N = 11

Age, Mean (SD) Gender, N (%) Education level (%)

73.09 (10.47) Male

5 (45.5)

Female

6 (54.5)

Elementary school

1 (9.1)

Middle school

3 (27.3)

Senior high school

3 (27.3)

Bachelor’s degree

3 (27.3)

Master’s degree or higher

1 (9.1)

Marital status (%)

Married

8 (72.7)

Widowed

3 (27.3)

Residence (%)

Private home

9 (81.8)

Nursing home

2 (18.2)

Retired

11 (100)

Employment status (%)

2.2 Measures Both before and after the intervention, savoring beliefs and strategies were assessed. The Savoring Beliefs Inventory (SBI) was used to investigate savoring beliefs [10]. Two independent translators translated items in the original version of the SBI into Italian and back translated them. The outcome was discussed until all the translators agreed on reaching a consensus on cross-language equivalence. The SBI consists of 24 items that can be answered using a 7-point Likert scale, with 1 = strongly disagree and 7 = strongly agree. The SBI is made up of three subscales: anticipation (α = .63), present (α = .80) and reminiscence (α = .80). Adding together these subscales, the total score can be obtained (α = .85). High scores on the SBI indicate a greater perceived ability to savor positive events. An adapted, short version of the Ways of Savoring Checklist (WOSC) was used to measure savoring strategies [11]. Two independent translators translated items from the original version of the WOSC into Italian and back translated them. The outcome was discussed until all the translators reached a consensus on cross-language equivalence. The original version consists of ten sub-scales and a total of 60 items that can be answered on a 7-point Likert scale from 1 = strongly disagree to 7 = strongly agree. In this study, we used an adapted short version of 10 items focusing only on a few amplifying strategies (WOSC Amplifying, α = .65) including sharing with others, memory building, self-congratulation, absorption, and counting blessings. Positive and negative affect were assessed before and after the relational savoring e-intervention. The Italian version of the Scale of Positive and Negative Experiences (SPANE) was used to assess positive and negative affect [40–42]. This 12-item self-report scale is divided into two subscales: the SPANE-P (α = .93) which measures positive

Promoting Positive Emotions in Older Adults

93

emotions (6 items: positive, good, pleasant, happy, joyful, contented) and the SPANE-N (α = .89) which measures negative emotions (6 items: negative, bad, unpleasant, sad, afraid, angry). A 5-point Likert scale can be used to answer (1 = very rarely or never and 5 = very often or always). Scoring can be obtained through the sum of positive emotions and the sum of negative ones. Moreover, at the end of the first five activities, participants were asked to write down their reflections about the exercise and to rate the intensity of the positive and negative emotions experienced during the exercise (joy, satisfaction, enthusiasm, hope, gratitude, fear, anger, sadness) through a 7-point Likert scale (1 = never and 7 = very much). Finally, at the end of the intervention, the User Experience Questionnaire (UEQ) (Italian version: Mauro Zenoni, available online on the website www.ueq-online.org) was used to investigate the participants’ experiences in using the online platform and doing the relational savoring exercises [43]. It contains 6 scales and is composed of 26 total items. For this study, we used four scales: attractiveness (α = .86), which refers to the overall impression of the product; dependability (α = .73) which refers to how secure and predictable the interaction with the site is); stimulation (α = .68) which is the interest and enthusiasm generated by its use; and novelty (α = .77) which focuses on its innovation and creativity. Because the internal consistency of the items appeared unacceptable (α < .5), we did not include perspicuity (the perceived ease of use) and efficiency (the organization and practicality of the website). The UEQ can be answered through semantic differentials, and each item can be assigned a score between -3 and + 3, where -3 represents the most negative answer, 0 a neutral answer, and + 3 the most positive answer. For each scale, values between -0.8 and 0.8 indicate a neutral evaluation, scores greater than 0.8 represent a positive evaluation, and lower than -0.8 indicate a negative evaluation [44]. A final item on a 5-step Likert scale (1 = very little and 5 = very much) was used to assess the perceived usefulness of the relational savoring e-intervention (Referring to the entire intervention, how much do you think it was useful for you?) together with the opportunity of write reflections about the intervention as a whole. 2.3 The Relational Savoring e-intervention The intervention was designed and developed exclusively online. The WordPress platform was used to create an ad hoc, intuitive website with simple graphics (at the domain www.progettosavor.it). It was accessible from the principal operating systems (Android, iOS, Windows, macOS, etc.) and from technological devices such as smartphones, tablets, and computers. Participants could access it with a personal username and password. Once they have entered the website, they could complete the online questionnaires through a link to Qualtrics and the intervention exercises (see Fig. 1). Participants could contact the researchers at any time if they had difficulties accessing or using the online platform. This 3-week intervention included two exercises a week, for a total of six exercises. Older adults, at the time of publication on the website of each exercise (on Monday and Thursday), received a reminder by email. Each exercise lasted about 10 min and was proposed through recorded-audio and a written text to give participants the opportunity to choose their favorite fruition mode.

94

E. Pancini et al.

The first week of the intervention focused on the past. • In the first exercise, older adults were asked to draw up a list of 3 positive moments and experiences they had with their loved ones (e.g., their partner, their son or daughter, a friend, etc.) throughout their lives. The aim was to promote awareness of positive events shared with close people. • In the second exercise, participants were guided to recall and amplify a positive past experience shared with a loved one. The goal was to help participants immerse themselves in the chosen event and to relive the positive emotions they experienced then. The second week focused on the present. • In the third exercise, participants were asked to think about two or three people in their life whom they were happy and grateful for, with the aim of promoting reflection and awareness of important people in older adults’ lives.

Fig. 1. The savoring activities.

Promoting Positive Emotions in Older Adults

95

• In the fourth activity, participants were invited to select one of the people identified in the previous activity and to write and deliver a gratitude letter to him or her. The third week focused on the future. • In the fifth exercise, participants were asked to perform acts of kindness towards one or more people close to them and reflect on their feelings during, before, and after these acts. Specifically, participants were asked to focus more on the future positive emotions elicited by the exercise before carrying it out. • During the last exercise, participants watched a 5-min relaxation video in which music and natural scenarios were integrated. A guided narrative retraced the various strategies learned or strengthened throughout the intervention. Finally, it was anticipated that, after completing the post-intervention questionnaires, participants would receive a certificate of participation and access to the website for a full year as a gift for their participation.

2.4 Procedure An email was sent to each older adult interested in participating with information about the intervention. Then, each participant who enrolled received personal credentials for accessing the website. The feasibility study included two assessment moments. In T0, participants were invited to fill out informed consent forms and online baseline questionnaires (SBI, WOSC Amplifying, SPANE). The relational savoring e-intervention took place over the next 3 weeks. At the end of the intervention (T1), participants completed the same questionnaires (SBI, WOSC Amplifying, SPANE), the UEQ, and a perceived usefulness item.

3 Results 3.1 Feasibility of the Relational Savoring e-intervention The data were analyzed using IBM SPSS 27.0. Paired t-tests revealed a significant increase in savoring strategies (WOSC Amplifying) and in positive emotions (SPANEP) in post-intervention compared to pre-intervention. Descriptive data are shown as mean (M) and standard deviation (SD) in Table 2. Table 2. Psychological dimensions pre- and post-intervention. Pre-intervention M (SD)

Post-intervention M (SD)

Paired t-tests t

p

SBI Anticipation

38.09 (6.56)

39.00 (6.32)

−.897

.391

SBI Present

37.73 (7.55)

39.55 (7.96)

−1.070

.310

(continued)

96

E. Pancini et al. Table 2. (continued) Pre-intervention M (SD)

Post-intervention M (SD)

Paired t-tests t

p

SBI Reminiscence

42.55 (11.06)

45.09 (5.12)

−.732

.481

SBI Total

118.36 (21.85)

123.64 (15.50)

−1.164

.272

WOSC Amplifying

4.73 (0.66)

5.20 (0.67)

−2.486

.032a

SPANE-P

19.55 (5.82)

22.64 (4.69)

−2.211

.051a

SPANE-N

12.73 (5.56)

11.64 (4.71)

.719

.448

a Significant (p < 0.05)

3.2 Descriptive Analysis of Emotions Elicited by the Exercises From the participants’ comments, in the first exercise (listing 3 positive events with a loved one), the most frequently remembered event (by 5 participants) was the birth of their children, and the emotions experienced with more intensity were joy, satisfaction, and gratitude (see Table 3). As for the second exercise (recall and amplify a positive past experience shared with a loved one), older adults mainly referred to trips shared with their partner and experienced high levels of joy and satisfaction. In the third exercise (think about two or three people they are grateful for having in their life), participants thought of people belonging to their family (e.g., partner, children, grandfather, father, siblings), but also of people outside the family circle (e.g., nun, employee at work). The emotions felt with greater intensity were gratitude and satisfaction. In the fourth exercise (write and deliver a gratitude letter), some older adults wrote or considered writing a gratitude letter to loved ones who are alive, while others chose to address it to people who are no longer alive (e.g., friends, siblings, partners). Most participants experienced high levels of gratitude, satisfaction, and joy for writing the letter, even without delivering it. In the Table 3. Intensity of the emotions elicited by the exercises. First exercise M (SD)

Second exercise M (SD)

Third exercise M (SD)

Fourth exercise M (SD)

Fifth exercise M (SD)

Joy

6.09 (1.81)

6.27 (1.27)

5.64 (1.03)

5.64 (1.21)

5.60 (1.65)

Satisfaction

6.09 (1.81)

5.45 (1.92)

6.09 (0.50)

5.91 (1.22)

5.70 (1.50)

Enthusiasm

5.73 (1.74)

5.18 (1.25)

5.09 (0.70)

5.00 (1.73)

5.00 (1.82)

Gratitude

6.36 (0.92)

5.18 (2.32)

6.64 (0.50)

6.18 (0.87)

4.90 (1.52)

Hope

4.82 (2.04)

3.82 (2.18)

4.82 (1.40)

4.45 (2.07)

3.90 (1.73)

Fear

2.45 (1.92)

1.91 (1.58)

1.36 (0.50)

1.45 (0.82)

1.10 (0.32)

Anger

1.18 (0.40)

1.36 (0.67)

1.36 (0.50)

1.91 (1.92)

1.10 (0.32)

Sadness

2.18 (1.83)

3.18 (2.53)

3.09 (2.21)

2.82 (2.04)

1.10 (0.32)

Promoting Positive Emotions in Older Adults

97

fifth exercise (perform acts of kindness), participants struggled to identify specific acts of kindness. That is because, as older adults reported in the comments, they perceived kindness more as a way of being than as a specific action. Thus, they stated that kindness is a natural act towards one’s family and loved ones. Only three older adults specified the acts of kindness they performed. The emotions felt most intensely were satisfaction and joy. 3.3 User Experience and Perceived Usefulness

Table 4. User experience and perceived usefulness of the e-intervention. M

SD

Positive cut-off

Range

Attractiveness

1.33

1.11

>0.8

−3/ + 3

Dependability

1.18

0.63

>0.8

−3/ + 3

Stimulation

1.04

0.71

>0.8

−3/ + 3

Novelty

1.52

1.16

>0.8

−3/ + 3

Perceived usefulness

3.54

0.68

n.a.

1–5

Furthermore, at the end of the relational savoring e-intervention, participants filled out the User Experience Questionnaire. Participants reported a positive evaluation of the online platform in terms of novelty, attractiveness, dependability, and stimulation. Furthermore, older adults evaluated the intervention as useful (the average rating is between “quite useful” and “very useful”). Details are reported in Table 4. Concerning perceived usefulness, the relational savoring e-intervention was appreciated and fostered openness to a different way of looking. For example, a 70-year-old man stated, “It was a beautiful, engaging, constructive experience; it stimulated me to do and write things that I never thought I would do or write. Thank you for this wonderful experience.” Furthermore, a 68-year-old woman affirmed, “I was not used to reflecting on some aspects raised during the different activities. I think this was important.” Thus, the relational savoring exercises led participants to adopt a different perspective; for instance, a 66-year-old woman stated, “Thinking of something positive is not as easy as it might seem. It would be an exercise to do even for small everyday experiences, such as seeing a slice of blue in a cloudy sky and think that the sun will come.” Another 69-year-old man declared, “The intervention interested me, helped me focus more on the positive aspects of my life and not take them for granted.” A 71-year-old woman also stated, “It was very nice to participate, you managed to bring out a part of me that I didn’t think was there anymore.”

98

E. Pancini et al.

4 Discussion This research aimed to investigate the feasibility of a self-help relational savoring eintervention. The results partially supported our hypotheses. Regarding the effect of the intervention on savoring beliefs (Hp 1), participants reported a not significant improvement in their perceived ability to savor future, present, and past positive experiences shared with their loved ones. One possible explanation is that the SBI was conceived to investigate personal savoring and thus may be less sensitive to capturing actual changes in relational savoring. Another possible explanation is that physical and cognitive abilities gradually decline as people get older and, as a result, particular challenges may arise regarding savoring. For instance, they may no longer be motivated to savor the same positive experiences in the same way as when they were younger [12]. Alternatively, they may take longer to increase their perceived savoring abilities, considering that this intervention is relatively brief. Concerning the effect of the intervention on savoring strategies (Hp 2), participants reported a significant improvement. This means that older adults, after the intervention, felt more aware and more able to use actions and thoughts to amplify their positive emotions. Indeed, the relational savoring exercises may have helped participants experience and practice different strategies for savoring positive events. Furthermore, participants can also better adapt to aging by having a wider variety of savoring strategies at their disposal [12, 23]. Concerning the effect of the intervention on positive and negative emotions (Hp 3), participants reported a significant increase in positive emotions. Indeed, older adults had the opportunity in each exercise to reflect on positive events they had personally experienced and to think about important people in their lives. Such reflections were also written down, and this may have fostered awareness of the positive emotions experienced. This result is in line with other relational savoring interventions [30, 31]. We found only a slight decrease, but not a significant one, in negative emotions. This result may relate to the propensity to experience mixed emotions that increases with age. Indeed, in older adults, thinking back to positive events experienced with loved ones may also trigger negative feelings, especially if the event refers, for example, to a loss suffered [18, 45, 46]. In these cases, older adults may also feel sadness, resulting in a dual experience of positive and negative affect [18, 47]. However, descriptive analysis of the emotions elicited by the exercises showed that participants experienced high levels of positive emotions like satisfaction, joy, gratitude, and enthusiasm and low levels of negative emotions like sadness, fear, and anger. In the written comments for the first two RS activities, participants referred mainly to family members such as their children and their partner. In the following activities, they also referred to people outside the family, such as friends or employees at work. Interestingly, participants also thought of people they are grateful to have had in their lives who are no longer here. Concerning user experience, older adults perceived the ad hoc website as attractive, secure, stimulating, and innovative. Furthermore, most participants evaluated the intervention as useful. In conclusion, this feasibility study revealed that the relational savoring eintervention promoted a positive increase in savoring strategies and positive emotions and was appreciated by the older participants. The online self-help delivery effectively

Promoting Positive Emotions in Older Adults

99

supported the progress of activities that actively engaged participants through simple, short, and well-focused actions and reflections about their positive experiences and emotions shared with significant others. The results are therefore promising and represent a starting point for future studies that are called to investigate the effectiveness of savoring protocols aimed at enhancing positive emotions and well-being of older adults with controlled trials and to explore eventual differences based on participants’ digital skills, also considering the suggestions of Smith and Hollinger-Smith [23]. Indeed, they propose to include in these interventions activities that promote positive reminiscence and appreciation of the present and, to a lesser extent, activities with a focus on the future [23]. Such interventions can bring many benefits, including allowing older people to become aware of and enhance the positive experiences that are already happening in their lives. These reflections therefore suggest that practicing savoring can help people adapt effectively to aging. Funding. Funded by the European Union – Next Generation EU (Project: age-it – 8. Consequences and challenges of ageing). The views and opinions expressed are those of the authors and do not necessarily reflect those of the European Union or the European Commission. Neither the European Union nor the European Commission can be held responsible for them.

References 1. Salovey, P., Rothman, A.J., Detweiler, J.B., Steward, W.T.: Emotional states and physical health. Am. Psychol. 55(1), 110–125 (2000) 2. Santos, V., et al.: The role of positive emotion and contributions of positive psychology in depression treatment: systematic review. Clinical practice and epidemiology in mental health (2013) 3. Fredrickson, B.L., Cohn, M.A., Coffey, K.A., Pek, J., Finkel, S.M.: Open hearts build lives: positive emotions, induced through loving-kindness meditation, build consequential personal resources. J. Pers. Soc. Psychol. 95(5), 1045 (2008) 4. Pettit, J.W., Kline, J.P., Gencoz, T., Gencoz, F., Joiner, T.E., Jr.: Are happy people healthier? the specific role of positive affect in predicting selfreported health symptoms. J. Res. Pers. 35(4), 521–536 (2001) 5. Veenhoven, R.: Healthy happiness: effects of happiness on physical health and the consequences for preventive health care. J. Happiness Stud. 9(3), 449–469 (2008) 6. Carstensen, L.L., et al.: Emotional experience improves with age: evidence based on over 10 years of experience sampling. Psychol. Aging 26(1), 21 (2011) 7. Danner, D.D., Snowdon, D.A., Friesen, W.V.: Positive emotions in early life and longevity: findings from the nun study. J. Pers. Soc. Psychol. 80(5), 804 (2001) 8. Ostir, G.V., Markides, K.S., Black, S.A., Goodwin, J.S.: Emotional well-being predicts subsequent functional independence and survival. J. Am. Geriatr. Soc. 48(5), 473–478 (2000) 9. Jiao, J., Kim, S., Pitts, M.J.: Promoting subjective well-being through communication savoring. Communication Quarterly, 1–20 (2021) 10. Bryant, F.B.: Savoring Beliefs Inventory (SBI): a scale for measuring beliefs about savouring. J. Ment. Health 12(2), 175–196 (2003) 11. Bryant, F.B., Veroff, J.: Savoring: A new model of positive experience. Erlbaum Associates, Mahwah (2007) 12. Bryant, F.B., Chadwick, E.D., Kluwe, K.: Understanding the processes that regulate positive emotional experience: unsolved problems and future directions for theory and research on savoring. Int. J. Wellbeing 1(1) (2011)

100

E. Pancini et al.

13. Croft, A., Dunn, E.W., Quoidbach, J.: From tribulations to appreciation: Experiencing adversity in the past predicts greater savoring in the present. Soc. Psychol. Personality Sci. 5(5), 511–516 (2014) 14. Hurley, D.B., Kwon, P.: Results of a study to increase savoring the moment: differential impact on positive and negative outcomes. J. Happiness Stud. 13(4), 579–588 (2012) 15. Quoidbach, J., Berry, E.V., Hansenne, M., Mikolajczak, M.: Positive emotion regulation and well-being: comparing the impact of eight savoring and dampening strategies. Personality Individ. Differ. 49(5), 368–373 (2010) 16. Smith, J.L., Bihary, J.G., O’Connor, D., Basic, A., O’Brien, C.J.: Impact of savoring ability on the relationship between older adults’ activity engagement and well-being. J. Appl. Gerontol. 39(3), 323–331 (2020) 17. Smith, J.L., Bryant, F.B.: Enhancing positive perceptions of aging by savoring life lessons. Aging Ment. Health 23(6), 762–770 (2019) 18. Palmer, C.A.: Experimentally induced savoring and its impact on positive emotions, cognitions, and behaviors: investigating individual differences in effectiveness (2014) 19. Bryant, F.B.: Current progress and future directions for theory and research on savoring. Frontiers in Psychology, 12 (2021) 20. Smith, J.L., Harrison, P.R., Kurtz, J.L., Bryant, F.B.: Nurturing the capacity to savor: Interventions to enhance the enjoyment of positive experiences. In: Parks, A.C. (ed.) The Wiley-Blackwell Handbook of Positive Psychological Interventions, pp. 42–65. Oxford: Wiley-Blackwell (2014) 21. Bryant, F.B., Smith, J.L.: Appreciating life in the midst of adversity: savoring in relation to mindfulness, reappraisal, and meaning. Psychol. Inq. 26(4), 315–321 (2015) 22. Smith, J.L., Bryant, F.B.: The benefits of savoring life: savoring as a moderator of the relationship between health and life satisfaction in older adults. Int. J. Aging Hum. Dev. 84(1), 3–23 (2016) 23. Smith, J.L., Hollinger-Smith, L.: Savoring, resilience, and psychological well-being in older adults. Aging Ment. Health 19(3), 192–200 (2015) 24. Fredrickson, B.L.: Positive emotions broaden and build. In: Advances in experimental social psychology, vol. 47, pp. 1–53. Academic Press (2013) 25. Ramsey, M.A., Gentzler, A.L.: Age differences in subjective well-being across adulthood: The roles of savoring and future time perspective. Int. J. Aging Hum. Dev. 78(1), 3–22 (2014) 26. Salces-Cubero, I.M., Ramírez-Fernández, E., Ortega-Martínez, A.R.: Strengths in older adults: differential effect of savoring, gratitude and optimism on well-being. Aging Mental Health 23(8), 1017–1024 (2019) 27. Zautra, A., Smith, B., Affleck, G., Tennen, H.: Examinations of chronic pain and affect relationships: applications of a dynamic model of affect. J. Consult. Clin. Psychol. 69(5), 786 (2001) 28. Carver, C.S., Scheier, M.F., Segerstrom, S.C.: Optimism. Clin. Psychol. Rev. 30(7), 879–889 (2010) 29. Smith, J.L., Hanni, A.A.: Effects of a savoring intervention on resilience and well-being of older adults. J. Appl. Gerontol. 38(1), 137–152 (2019) 30. Borelli, J.L., et al.: Relational savoring: an attachment-based approach to promoting interpersonal flourishing. Psychotherapy 57(3), 340 (2020) 31. Borelli, J.L., et al.: With or without you: Preliminary evidence that attachment avoidance predicts nondeployed spouses’ reactions to relationship challenges during deployment. Professional Psychol. Res. Practice 45(6), 478 (2014) 32. Borelli, J.L., Bond, D.K., Fox, M., Horn Mallers, M.: Relational savoring reduces physiological reactivity and enhances psychological agency of older adults. J. Appl. Gerontol. (2019)

Promoting Positive Emotions in Older Adults

101

33. Gaggioli, A., Villani, D., Serino, S., Banos, R., Botella, C.: Editorial: positive technology: designing E-experiences for positive change. Front. Psychol. 10, 1571 (2019) 34. Carr, A., et al.: Effectiveness of positive psychology interventions: a systematic review and meta-analysis. J. Posit. Psychol. 16(6), 749–769 (2021) 35. Abbaspur-Behbahani, S., Monaghesh, E., Hajizadeh, A., & Fehresti, S.: Application of mobile health to support the elderly during the COVID-19 outbreak: A systematic review. Health Policy and Technology, 100595 (2022) 36. Riva, G., et al.: Positive and transformative technologies for active ageing. Med. Meets Virtual Real. 22, 308–315 (2016) 37. Grossi, G., Lanzarotti, R., Napoletano, P., Noceti, N., Odone, F.: Positive technology for elderly well-being: a review. Pattern Recogn. Lett. 137, 61–70 (2020) 38. Riva, G., Villani, D., Cipresso, P., Gaggioli, A.: Positive technology: the use of technology for improving and sustaining personal change. In Integrating technology in positive psychology practice, pp. 1–37. IGI Global (2016) 39. Heintzelman, S.J., e al.: ENHANCE: evidence for the efficacy of a comprehensive intervention program to promote subjective well-being. J. Exp. Psychol.: Appl. 26(2), 360 (2020) 40. Diener, E., et al.: New measures of well-being. In: Assessing Well-Being, pp. 247– 266. Springer, Dordrecht (2009) 41. Diener, E., et al.: New well-being measures: Short scales to assess flourishing and positive and negative feelings. Soc. Indic. Res. 97(2), 143–156 (2010) 42. Corno, G., Molinari, G., Baños, R.: Assessing positive and negative experiences: validation of a new measure of well-being in an Italian population (2016) 43. Laugwitz, B., Held, T., Schrepp, M.: Construction and evaluation of a user experience questionnaire. In: Symposium of the Austrian HCI and Usability Engineering Group, pp. 63–76. Springer, Heidelberg (2008) 44. Rauschenberger, M., Schrepp, M., Pérez Cota, M., Olschner, S., Thomaschewski, J.: Efficient measurement of the user experience of interactive products. How to use the user experience questionnaire (UEQ) (2013) 45. Carstensen, L.L., Pasupathi, M., Mayr, U., Nesselroade, J.R.: Emotional experience in everyday life across the adult life span. J. Pers. Soc. Psychol. 79(4), 644 (2000) 46. Carstensen, L.L., et al.: Emotional experience improves with age: evidence based on over 10 years of experience sampling. Psychol. Aging 26(1), 21 (2011) 47. Isaacowitz, D.M., Blanchard-Fields, F.: Linking process and outcome in the study of emotion and aging. Perspect. Psychol. Sci. 7(1), 3–17 (2012)

The Relationship Between Older Drivers’ Cognitive Ability and Takeover Performance in Conditionally Automated Driving Qijia Peng1 , Yanbin Wu2 , Toshihisa Sato2 , and Sunao Iwaki2(B) 1 University of Tsukuba, 1 Chome-1–1 Tennodai, Tsukuba, Ibaraki 305-8577, Japan 2 National Institute of Advanced Industrial Science and Technology, 1-1-1 Umezono, Tsukuba,

Ibaraki 305-8560, Japan [email protected] Abstract. In takeover process of conditionally automated driving, cognitive abilities, especially the executive function abilities, are found to play a significant role in driver’s performance. During the automated driving period, engaging in nondriving related tasks (NDRTs) also significantly affects takeover performance; moreover, different attributes of NDRT such as different task modalities were found to have different influences on takeover performance. This study aims to explore the relationship between the influence of different modalities of NDRT and the corresponding executive function abilities in older drivers during conditionally automated driving. We designed computerized cognitive experiments to evaluate older drivers’ executive function abilities in different modalities, a simulated driving experiment to evaluate older drivers’ takeover performance, and then investigated their relationship by a correlation study. Twenty-four participants were recruited in this experiment. The results showed that instant lateral stability of takeover performance tends to be better when engaged in auditory n-back task than visual SuRT, and longer continuous lateral control was most unstable without any NDRT. The results from correlation analysis indicated that older drivers with worse executive function abilities in either auditory or visual modality would perform less stable takeover behavior when engaged in NDRTs requiring the same cognitive modality. Overall, these findings verified the correlation between takeover performance and executive function abilities, and provide understanding of the correlation between modalities of NDRTs and corresponding cognitive abilities of older drivers. Keywords: Takeover performance · Older drivers · Executive functions

1 Introduction 1.1 Background The automated vehicles are expected to benefit older drivers by enhancing their mobility [1, 2]. During level 3 conditionally automated driving [3], the system is expected to take control of the vehicle, and thus drivers are allowed to engage in non-driving related tasks (NDRTs); while in certain situations the system will prompt a takeover request (TOR) to the drivers, and they must be ready to regain control of the vehicle. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 102–115, 2023. https://doi.org/10.1007/978-3-031-34866-2_8

The Relationship Between Older Drivers’ Cognitive Ability

103

The implementation of automated vehicles can benefit older drivers in many ways, including the engagement of NDRT during automated driving. However, the effect of NDRT is crucial especially for older drivers, considering the decline in their driving related abilities. Older drivers’ takeover performance would be more cautious or conservative, such as more brakes and longer safe distance [4, 5], or slower in reactions and decision making [6]. For older drivers, cognitive abilities decreasing with age were demonstrated to be correlated with dangerous driving behaviors and more crashes [7, 8]. Higher order functional abilities, especially executive function abilities are found to be more dominant in older drivers’ performance [9–12]. To be specific, working memory is related abilities were found to play an important role in decision making [13], and performance in inhibition related tests is associated to on-road driving performance [14] and observation errors [8]. Peng et al. also suggested that executive functions (shifting and working memory updating components) are vital factors in older drivers’ interaction with automated system [15]. In addition, older drivers were found to be willing to engage in various NDRTs [4, 16]. Previous researches have made efforts on the different influence of NDRTs with various modalities (commonly in visual, auditory, cognitive or physical) on takeover performance. For instance, cognitive load of NDRTs significantly reduced both longitudinal and lateral control, but the effects of physical and visual modality was not significant [17]. Research about takeover time also found that physical attributes of NDRT significantly affect motor reaction of takeover, while gaze and cognitive attributes influence mental reaction of takeover [18]. Although previous studies showed that NDRTs in various modalities significantly influence driving behaviors, knowledge about the different influences of NDRT modalities and the comparison between them is still limited. Moreover, the underlying correlation with influence of NDRT modalities and the cognitive abilities involved correspondingly is still unclear. 1.2 Study Objectives In this study, we plan to fill these gaps by exploring the relationship between the influence of different modalities of NDRT and the corresponding executive function abilities, especially for the older drivers. The purpose of this study should be two-fold: first, to confirm the effects of NDRTs in different modalities on takeover behaviors of older drivers in automated driving, and second, to investigate the correlation between modalities of executive function and takeover performance under the effect of corresponding NDRT modalities. We designed a set of computerized cognitive tasks to evaluate older drivers’ executive function abilities in different modalities, a simulated driving tasks to evaluate older drivers’ takeover performance, and then investigated their relationship by a correlation analysis.

104

Q. Peng et al.

2 Methods 2.1 Participants Twenty-four participants (12 males and 12 females) were recruited from advertisements in local community in this experiment. All of the participants were active driver in daily life, and were all above 60 years (M = 67.6 years, SD = 4.5 years). All participants provided informed consent complying with the research protocol approved by the Institutional Review Board of the AIST. 2.2 Experiment Task Design Takeover Task Design in Simulated Driving For takeover performance evaluation, a 3 (NDRT engagement: no NDRT, SuRT and N-back) × 2 (TTC lead time: 4 s and 8 s ahead) within-subjects factor experiment was designed. A typical takeover task was designed for this simulation study: participants encountered a variant version of the Lane Changing Task after TOR during a simulated level 3 automated vehicle in highway (see Fig. 1). Similar scenarios are commonly used in previous researches to evaluate drivers’ takeover performance when engaging with NDRT [4, 19].

Fig. 1. Scenario of takeover request and driving tasks

During the experiment, participants (Vehicle 1) were in auto-driving mode most of the time. An audio message in Japanese (TOR, takeover request) warned them to switch to manual driving in certain driving scenarios: participants were required to take control of both steering wheel and pedals to avoid collision by changing lanes to pass a stationary truck (Vehicle 2), revealed by a leading car (Vehicle 3) ahead of the participants. After passing the truck, they changed lanes back again to return to the original lane. Participants also had to pay attention to extra moving cars (Vehicle 4) in the right lane while changing lanes, and they had to make proper decisions for lane changing. The speed for automated driving and all moving vehicles were set at 82 km/h (based on the environment database of the simulation). The TOR occurred at 4 s or 8 s (correspondingly in each condition) in front of the stationary vehicle. Extra moving cars were running in the right lane next to the self-vehicle’ current lane at the same speed as the automated driving. The scenario ended when the participant drove 200 m (approximately

The Relationship Between Older Drivers’ Cognitive Ability

105

8–10 s) after the stationary vehicle, also noted by an audio notice of “Automated driving starts.” There were 3 simulated driving trial in total, including the 3 conditions of NDRT engagement (no NDRT, SuRT and N-back). All driving trial contained conditions with 4 s or 8 s TTC lead time. The sequence was randomized and balanced to avoid learning effect. NDRT Design During Simulated Driving In this study, we implemented standardized NDRTs during simulated driving according to previous similar research [20, 21]: a visual surrogate reference task (SuRT) [22] involving visual workload, and an auditory n-back task involving auditory and cognitive workload. These two NDRTs were also commonly used in previous researches focusing on the effects of NDRT on takeover performance as standardized secondary tasks [23]. • Surrogate Reference Task: The SuRT designed in this experiment was a slight variant version on a tablet computer fixed in the central console of the car. It required the participants to visually search for a target circle surrounded by a set of distractor circles on a screen. The target and distractor circles were all white circles presented in front of a black background (shown in Fig. 2). The target was approximately 1/6 larger than the distractors, and they were all randomly distributed over the screen. The participants were asked to respond to the target by locating the zone where the target was, and pressing the approximate location of the target circle. Figure 3 shows the way of interaction when participants were engaged in the SuRT in this experiment. The SuRT was implemented in a Windows Surface fixed in the center console of the cabin. • N-back task: For every 2 s, voice of a letter would be randomly played by the speaker, and participants were required to report orally if they heard the target letter. In this experiment, we implement a 2-back task, in which the target was any letter identical to the letter appearing two trials before. The auditory n-back task continued during the automated driving period, until the TOR occurred and stopped right after the audio of TOR was played. Cognitive Task Design In order to investigate the relationship between takeover performance influenced by NDRT in different modalities and the executive function abilities concerning corresponding modalities, we have to evaluate drivers’ executive function abilities corresponding to the NDRT modalities (auditory and visual). The tasks in the cognitive tests must be in the similar modalities in the NDRT. We chose the following paradigm with different modalities involved in the stimuli of the tasks: a visual Simon Task for inhibition control in visual modality, an auditory Simon Task for inhibition control in auditory modality, and a Visuospatial Working Memory Task for working memory updating in visual modality. Then we designed three computerized cognitive tasks based on these paradigms for the evaluation of executive functions concerning different modalities and abilities.

106

Q. Peng et al.

Fig. 2. Example screen of Surrogate Reference Task (SuRT)

We designed detailed instructions and training process for the old participants. Illustrations of each task are shown in Fig. 4. The Simon task were designed to evaluate cognitive abilities in inhibition control of executive functions of older drivers. • Visual Simon task: For Visual Simon task, participants were asked to respond to visual stimuli in different colors presented on either side of the screen. In each trial, a red or green square appeared on the right or left side of the screen and remained visible for 500 ms. Participants were instructed to press the left button (F button in keyboard) when they saw a green square, and the right button (J button in keyboard) when they saw a red square. The squares would appear congruently or incongruently on the left or right randomly in corresponding to the requested button. There were in total 96 (24 * 2 * 2) trials in the experiment, with 24 replications, 2 sides (stimuli left or right) and 2 conditions (congruent or incongruent). All trials were presented in randomized order. • Auditory Simon task: For Auditory Simon task, participants were required to wear headphones and respond to auditory stimuli in high/low pitches. In each trial, a “bee” sound effect was played in the right or left earphone. The sound effect had two types of tones: it would be either in “high pitch” (1000 Hz) or “low pitch”

The Relationship Between Older Drivers’ Cognitive Ability

107

Fig. 3. Illustration of the interaction with the touch screen for the SuRT

Fig. 4. Illustration of the cognitive tests (for some conditions)

(500 Hz). During the experiment, the sound would be only played in either right or left earphone. Similar to the previous task, participants were instructed to press the left button (F button in keyboard) when they heard a lower sound, and the right button (J button in keyboard) when they heard a higher sound. The sounds would appear congruently or incongruently on the left or right randomly in corresponding to the requested button.

108

Q. Peng et al.

There were in total 96 (24 * 2 * 2) trials in the experiment, with 24 replications, 2 sides (stimuli left or right) and 2 conditions (congruent or incongruent). All trials were presented in randomized order. • Visuospatial Working Memory task: For Visuospatial Working Memory task, participants were required to memorize and select target from a matrix presented on the screen. This task was designed to evaluate participants’ visuospatial working memory capacity [24]. This ability could be related with visual search speed [25], and could play a role in supporting drivers’ situation awareness for traffic location [26]. In each trial, the task began with 9 blue squares shown on the screen in 3 by 3 matrix. There would be multiple target squares (from 2 to 9, according to different conditions), and they were shown by the changing of the color (from blue to yellow). Each target square would keep “shown” in yellow color for 500 ms, and then turn blue again. The target squares were revealed one by one (500 ms for each target, and 500 ms for the interval) in a random sequence. The participants were required to memorize both location and sequence of the targets. After all target squares in this trial was shown, the participants were asked to select the target squares shown before with click and confirm their choice. There were in total 30 trials in the experiment. Target number started with 2. If one participant was correct about the locations and sequence in the last trial, the target number would increase by 1 in the next trial; if the participant’s response was incorrect, the target number would decrease by 1 in the next trial. Subjective Questionnaire About Driving Experience In order to investigate participants’ subjective evaluation about their perceived difficulty and fatigue during different NDRT, we introduced a simple questionnaire including 5 items: effort, fatigue, boredom, difficulty and sleepiness. After each condition of simulated driving, the participants were required to report their perceived feelings on the 5 items by rating at a degree from 1 to 9 (from “not … at all” to “extremely feeling …”). The higher the ratings, the more effort the participant perceived to pay, and the more fatigue, boredom, difficulty and sleepiness the participant felt during the driving. 2.3 Apparatus The cognitive tasks were conducted on a PC and programmed in the PsychoPy® platform. The driving tasks was conducted on a high-fidelity driving simulator (Mitsubishi Precision Co.), including a real car cabin and 6-axis motion platform, 8 projectors with an approximate 300-degree field of view for visual presentation, and 8-speaker sound system for audio display. The simple illustration of the system structure is shown in Fig. 5.

The Relationship Between Older Drivers’ Cognitive Ability

109

Fig. 5. Simple illustration of the driving simulator structure and a real view of the cabin Note: Pictures are from a previous research using the same driving simulator [19]

2.4 Measurements • Simulated driving task There are two main measures for lateral stability: sdSteer: Standard deviation of the steering wheel angle, calculated by the standard deviation of the steering wheel angle after TOR. In this study, the sdSteer was calculated within 1 s after TOR, indicating an “instant” lateral stability of the drivers. SDLP: Standard deviation of lateral position, calculated by the standard deviation of the lateral distance between vehicle and lane center after TOR and before the end of manual driving. In this study, the SDLP was calculated on the right lane after the first lane changing, indicating a relatively longer “continuous” lateral stability of the drivers. • Cognitive tests The measures for inhibition control were the Simon Effect (calculated based on the difference in reaction time between congruent and incongruent trials in mixed blocks) in both auditory and visual Simon tasks. The measure for visuospatial working memory capacity was the average number of the target among the correct responses. Outliers were identified as observation points outside 1.5*Inter Quartile Range, which was the difference between the 75th and 25th quartiles of the dataset. The observations higher and lower than the boundary were replaced by the higher and lower limits of the values of the 95th and 5th percentiles of the dataset.

110

Q. Peng et al.

3 Results and Discussion 3.1 Subjective Measures One-way ANOVA was performed individually to all measures to compare the effect of the three different conditions of NDRT engagement. Benjamini & Hochberg adjustment method was used in all pairwise comparison between group levels. Descriptive data are shown in Table 1. Table 1. Descriptive statistics for subjective measures in different NDRT conditions NDRT

Effort

Fatigue

Boredom

Difficulty

Sleepiness

No task

3.67 ± 0.26

3.17 ± 0.27

3.04 ± 0.23

3.92 ± 0.33

3.08 ± 0.31

SuRT

4.88 ± 0.27

4.38 ± 0.27

1.87 ± 0.17

4.58 ± 0.33

1.79 ± 0.21

Nback

5.71 ± 0.30

4.92 ± 0.34

1.63 ± 0.17

5.46 ± 0.37

1.96 ± 0.24

From the results, we can find that the engagement of cognitive n-back task significantly increase effort (F(2,46) = 12.02, p < 0.001) and fatigue (F(2,46) = 9.35, p < 0.001), especially compared with no task conditions. On the other hand, participants showed significantly more bored (F(2,46) = 11.71, p < 0.001) and sleepy (F(2,46) = 8.56, p < 0.001) when not engaged in any NDRT. 3.2 Driving Performance After TOR The differences of the two measures in simulated driving performance in different conditions of NDRT and TTC lead time in error-bar charts are presented in Fig. 6. Two-way ANOVA was conducted separately on each dependent variable (sdSteer and SDLP, shown in Table 2).

Fig. 6. Differences of driving performance (sdSteer and SDLP) under different conditions of NDRT engagement and TTC lead time

The Relationship Between Older Drivers’ Cognitive Ability

111

Table 2. ANOVA Table for measures of driving performance after TOR

sdSteer

SDLP

Effect

DFn, DFd

F

p

Significance Level

NDRT

2,46

3.86

0.028

TTC

1,23

27.72

< 0.001

NDRT*TTC

2,46

0.57

0.567

NDRT

2,46

1.42

0.251

TTC

1,23

7.72

0.011

*

NDRT*TTC

2,46

4.29

0.020

*

* ***

Note: Significant level:. < 0.1; * < 0.05; ** < 0.01; *** < 0.001.

For standard deviation of steering wheel angle, the results from a two-way ANOVA revealed that there was significant main effect of NDRT engagement on sdSteer between at least two groups (F(2,46) = 3.86, p = 0.028). Post-hoc analysis for multiple comparisons found that sdSteer in conditions engaging in N-back tasks was relatively lower than the other two conditions, but the results were not statistically significant. Significant main effect was also found in conditions with different TTC lead time, where sdSteer were significantly larger in 4 s than 8 s (F(1,23) = 27.72, p < 0.001). No significant interaction effects were found for sdSteer. For standard deviation of lateral positions, we also found significant main effect in conditions with different TTC lead time, where SDLP were significantly larger in 8 s than 4 s (F(1,23) = 7.72, p = 0.011). A significant interaction between TTC lead time and NDRT type was found (F(2,46) = 4.29, p = 0.020). Simple effect test revealed that in conditions without NDRT engagement, SDLP was significantly higher in 8 s TTC lead time than 4 s lead time (t = -3.42, df = 45.27, p = 0.001). The results showed different patterns between instant and longer lateral stability, indicating that older drivers’ instant lateral stability (sdSTeer) tended to be worse when engaged in visual SuRT than auditory n-back task, and longer continuous lateral control (SDLP) was most unstable without any NDRT. The post hoc analysis of sdSteer showed that the instant lateral stability tends to be better when engaged in auditory n-back task than visual SuRT. The results could be explained as participants had to share their visual attention with the visual NDRT [27]. Therefore, it is likely that it took those participants longer to develop situation awareness [28]. Similar conclusion could be found in the research on NRDTs by Zeeb et al., suggesting that lateral vehicle control deteriorated when drivers are visually distracted, such as reading the news or watching videos [29]. The results from SDLP showed that the longer continuous lateral control was most unstable without any NDRT. This could be explained by our findings previously in subjective measures: when not engaged in any NDRT, the boredom rating of participants was significantly higher than engaged in NDRTs; and similar results also appeared with Sleepiness rating. In the easiest condition (without NDRT, with longer lead TTC), it is possible that the participants had lower arousal when they encountered a TOR, and thus

112

Q. Peng et al.

they needed more time to regain control. So the difference may appear in the measure of SDLP, for it reflects a longer period of lateral maneuvering. 3.3 Correlations of Executive Function Abilities and Driving Performance Correlations between the three executive function abilities (visual and auditory Simon task, and Visuospatial Working Memory task) and driving performance (sdSteer and SDLP) were analyzed in each condition (different engagement in NDRT, different TTC lead time) using Pearson’s correlation coefficients. In condition with 4-s TTC lead time, when engaged in SuRT as a NDRT, significant correlations were found between sdSteer and Visuospatial Working Memory Capacity (t = −2.18, df = 22, p = 0.039, r = −0.42). In condition with 8-s TTC lead time, when engaged in N-back task as a NDRT, significant correlations were found between sdSteer and Simon Effect in auditory Simon task (t = 2.17, df = 22, p = 0.041, r = 0.42). We did not detect significant correlation between executive function measures and driving performance in other conditions. The results from correlation analysis showed consistency with previous findings on the influence of executive functions [15]. The capacity of working memory will affect drivers’ takeover process by affecting the updating of memory about driving task, including remembering vehicle status, traffic conditions or relative operations. Older drivers with lower working memory capacity may have difficulties in those tasks, and thus lead to a weaker lateral control of the vehicle. For inhibition related abilities, driving task could be correlated with a relevant process called Prepotent Response Inhibition referring to a deliberate suppression of dominant or prepotent responses [12]. In the takeover process in our experiment, the participants have to suppress their intention for the NDRT, and the requirement was specifically stronger when the NDRT requires more attention and produced more arousal (in the condition when n-back as NDRT). Moreover, in this study we particularly found the correlation in corresponding modalities: when visually engaged with SuRT, the significant correlation appeared with cognitive measures from visual working memory capacity; when auditory engaged with n-back task, the significant correlation appeared with cognitive measures from auditory Simon task. The results indicated that older drivers with worse executive function abilities in either auditory or visual modality would perform less stability in takeover behavior when engaged in NDRTs requiring corresponding cognitive modalities.

4 Conclusions and Future Improvements In this study, we investigated the takeover performance under interaction with different NDRTs, and explored the correlation between the influence of different modalities of NDRT and the corresponding executive function abilities of older drivers. We designed a set of computerized cognitive tasks to evaluate older drivers’ executive function abilities in different modalities, and a simulated driving tasks to evaluate older drivers’ takeover performance, and then investigated their relationship by a correlation analysis. Twenty-four participants were recruited in this experiment.

The Relationship Between Older Drivers’ Cognitive Ability

113

We can conclude from the results that both instant and continuous lateral control would be affected by engagement of NDRTs: instant lateral stability tends to be better when engaged in auditory n-back task than visual SuRT, and longer continuous lateral control was most unstable without any NDRT. The results from correlation analysis of cognitive abilities proved that older drivers with worse executive function abilities in either auditory or visual modality would perform less stability in takeover behavior when engaged in NDRTs requiring corresponding cognitive modalities. Two limitations of the present study should be mentioned. First, we did not measure takeover time in response to TOR due to the design of the NDRT. Takeover performance could be measured more comprehensively by both takeover quality and takeover time. Since executive functions may be highly correlated with reaction time, further investigation on the correlation between reaction time related measures and executive function ability will be important. Second, the NDRT attribute considered in future researches could be more comprehensive. In our study, NDRTs only have different modalities (auditory-cognitive and visual). According to other research, the attributes of NDRT in real world are always very complex [30]. Future research about the influence of NDRTs should be more cautious about the interference from other attributes to draw more comprehensive conclusions. The results of the study verified the correlation between takeover performance and executive function abilities shown in previous researches, and confirm the corresponding correlation between modalities of NDRTs and cognitive abilities of older drivers. The findings could provide insights to the methodology of takeover related interaction design to improve driving experience of older drivers. Acknowledgments. This work was supported in part by the JSPS Grant #17H01758.

References 1. Fagnant, D.J., Kockelman, K.: Preparing a nation for autonomous vehicles: opportunities, barriers and policy recommendations. Transp. Res. Part A: Policy Practice 77, 167–181 (2015). https://doi.org/10.1016/j.tra.2015.04.003 2. Bellet, T., Paris, J.-C., Marin-Lamellet, C.: Difficulties experienced by older drivers during their regular driving and their expectations towards Advanced Driving Aid Systems and vehicle automation. Transp. Res. Part F: Traffic Psychol. Behav. 52, 138–163 (2018). doi.https:// doi.org/10.1016/j.trf.2017.11.014 3. SAE International: Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles 4. Clark, H., Feng, J.: Age differences in the takeover of vehicle control and engagement in non-driving-related activities in simulated driving with conditional automation. Accid. Anal. Prev. 106, 468–479 (2017). https://doi.org/10.1016/j.aap.2016.08.027 5. Körber, M., Gold, C., Lechner, D., Bengler, K.: The influence of age on the take-over of vehicle control in highly automated driving. Transport. Res. F: Traffic Psychol. Behav. 39, 19–32 (2016). https://doi.org/10.1016/j.trf.2016.03.002 6. Li, S., Blythe, P., Guo, W., Namdeo, A.: Investigating the effects of age and disengagement in driving on driver’s takeover control performance in highly automated vehicles. Transp. Plan. Technol. 42, 470–497 (2019). https://doi.org/10.1080/03081060.2019.1609221

114

Q. Peng et al.

7. Stutts, J.C., Stewart, J.R., Martell, C.: Cognitive test performance and crash risk in an older driver population. Accid. Anal. Prev. 30, 337–346 (1998). https://doi.org/10.1016/S0001-457 5(97)00108-5 8. Anstey, K.J., Wood, J.: Chronological age and age-related cognitive deficits are associated with an increase in multiple types of driving errors in late life. Neuropsychology 25, 613–621 (2011). https://doi.org/10.1037/a0023835 9. Daigneault, G., Joly, P., Frigon, J.-Y.: Executive functions in the evaluation of accident risk of older drivers. J. Clin. Exp. Neuropsychol. 24, 221–238 (2002). https://doi.org/10.1076/jcen. 24.2.221.993 10. Adrian, J., Postal, V., Moessinger, M., Rascle, N., Charles, A.: Personality traits and executive functions related to on-road driving performance among older drivers. Accid. Anal. Prev. 43, 1652–1659 (2011). https://doi.org/10.1016/j.aap.2011.03.023 11. Mäntylä, T., Karlsson, M.J., Marklund, M.: Executive control functions in simulated driving. Appl. Neuropsychol. 16, 11–18 (2009). https://doi.org/10.1080/09084280802644086 12. Adrian, J., Moessinger, M., Charles, A., Postal, V.: Exploring the contribution of executive functions to on-road driving performance during aging: a latent variable analysis. Accid. Anal. Prev. 127, 96–109 (2019). https://doi.org/10.1016/j.aap.2019.02.010 13. Guerrier, J.H., Manivannan, P., Nair, S.N.: The role of working memory, field dependence, visual search, and reaction time in the left turn performance of older female drivers. Appl. Ergon. 30, 109–119 (1999). https://doi.org/10.1016/S0003-6870(98)00017-9 14. Baldock, M.R.J., Mathias, J., McLean, J., Berndt, A.: Visual attention as a predictor of onroad driving performance of older drivers. Aust. J. Psychol. 59, 159–168 (2007). https://doi. org/10.1080/00049530701458035 15. Peng, Q., Wu, Y., Qie, N., Iwaki, S.: Age-related effects of executive function on takeover performance in automated driving. Sci Rep. 12, 5410 (2022). https://doi.org/10.1038/s41598022-08522-4 16. Li, S., Blythe, P., Guo, W., Namdeo, A.: Investigation of older drivers’ requirements of the human-machine interaction in highly automated vehicles. Transport. Res. F: Traffic Psychol. Behav. 62, 546–563 (2019). https://doi.org/10.1016/j.trf.2019.02.009 17. Lee, S.C., Yoon, S.H., Ji, Y.G.: Effects of non-driving-related task attributes on takeover quality in automated vehicles. Int. J. Hum.-Comput. Inter. 37, 211–219 (2021). https://doi. org/10.1080/10447318.2020.1815361 18. Yoon, S.H., Lee, S.C., Ji, Y.G.: Modeling takeover time based on non-driving-related task attributes in highly automated driving. Appl. Ergon. 92, 103343 (2021). https://doi.org/10. 1016/j.apergo.2020.103343 19. Wu, Y., Kihara, K., Takeda, Y., Sato, T., Akamatsu, M., Kitazaki, S.: Effects of scheduled manual driving on drowsiness and response to take over request: a simulator study towards understanding drivers in automated driving. Accid. Anal. Prev. 124, 202–209 (2019). https:// doi.org/10.1016/j.aap.2019.01.013 20. Radlmayr, J., Gold, C., Lorenz, L., Farid, M., Bengler, K.: How traffic situations and nondriving related tasks affect the take-over quality in highly automated driving. Proc. Hum. Factors Ergonomics Soc. Ann. Meeting 58, 2063–2067 (2014). https://doi.org/10.1177/154 1931214581434 21. Wu, Y., Kihara, K., Takeda, Y., Sato, T., Akamatsu, M., Kitazaki, S.: Assessing the mental states of fallback-ready drivers in automated driving by electrooculography. In: 2019 IEEE Intelligent Transportation Systems Conference (ITSC), pp. 4018–4023 (2019). https://doi. org/10.1109/ITSC.2019.8917029 22. ISO/T. 14198: Road vehicles — Ergonomic aspects of transport information and control systems — Calibration tasks for methods which assess driver demand due to the use of in-vehicle systems (2012)

The Relationship Between Older Drivers’ Cognitive Ability

115

23. Naujoks, F., Befelein, D., Wiedemann, K., Neukum, A.: A review of non-driving-related tasks used in studies on automated driving. In: Stanton, N.A. (ed.) AHFE 2017. AISC, vol. 597, pp. 525–537. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-60441-1_52 24. Dumontheil, I., Klingberg, T.: Brain activity during a visuospatial working memory task predicts arithmetical performance 2 years later. Cereb. Cortex 22, 1078–1085 (2012). https:// doi.org/10.1093/cercor/bhr175 25. Woodman, G.F., Luck, S.J.: Visual search is slowed when visuospatial working memory is occupied. Psychon. Bull. Rev. 11, 269–274 (2004). https://doi.org/10.3758/BF03196569 26. Johannsdottir, K.R., Herdman, C.M.: The role of working memory in supporting drivers’ situation awareness for surrounding traffic. Hum Factors 52, 663–673 (2010). https://doi.org/ 10.1177/0018720810385427 27. Roche, F., Somieski, A., Brandenburg, S.: Behavioral changes to repeated takeovers in highly automated driving: effects of the takeover-request design and the nondriving-related task modality. Hum Factors 61, 839–849 (2019). https://doi.org/10.1177/0018720818814963 28. Endsley, M.R., Kiris, E.O.: The out-of-the-loop performance problem and level of control in automation. Hum Factors. 37, 381–394 (1995). https://doi.org/10.1518/001872095779 064555 29. Zeeb, K., Buchner, A., Schrauf, M.: Is take-over time all that matters? the impact of visualcognitive load on driver take-over quality after conditionally automated driving. Accid. Anal. Prev. 92, 230–239 (2016). https://doi.org/10.1016/j.aap.2016.04.002 30. Salubre, K.J., Nathan-Roberts, D.: Takeover request design in automated driving: a systematic review. Proc. Hum. Factors Ergon. Soc. Ann. Meeting 65, 868–872 (2021). https://doi.org/ 10.1177/1071181321651296

The After Effect of COVID-19 on Colour Perception in the Elderly Chuanke Qin1(B) and Ao Jiang2,3 1 The Glasgow School of Art, Glasgow, UK

[email protected]

2 Imperial College London, London, UK

[email protected] 3 EuroMoonMars ILEWG at ESA, Noordwijk, Netherlands

Abstract. Due to the high susceptibility of the elderly to infection, the lack of effective protection against the disease, and their declining physical functions, research on the consequences of COVID-19 in the elderly has remained limited since the disease outbreak. Through in-depth research, this study aims to discover what effect COVID-19 has on the colour perception of older individuals and to what extent. This study’s primary target population consisted of 60-year-olds and older who were infected with COVID-19. The first step was to use a large-scale questionnaire to determine older people’s age, gender, occupation, whether they had been infected with COVID-19, the length of time since their last infection, and whether they had any colour perception issues prior to infection. The FM100 and D15 colour vision tests, were then used to identify older individuals with current colour perception issues. The experimental results of this study indicate that COVID-19 does not affect the colour perception of older individuals. Keywords: COVID-19 · after effect · elderly · colour perception

1 Introduction In 2019 SARS-CoV-2 causes COVID-19, an infectious disease caused by the coronavirus virus. Most individuals with COVID-19 will experience mild to moderate symptoms and recover without treatment [1]. Nonetheless, some patients are gravely ill and require medical care. It is estimated that 600 million people have been infected with COVID19, many of them multiple times, and that over 6 million people have died as a result [2]. When an infected person coughs, sneezes, talks, sings or takes a deep breath, tiny liquid particles are expelled from his or her mouth or nose and spread the virus. A person can easily become infected if he or she inhales the virus from a person with COVID-19 or if he or she touches a contaminated surface and then touches his or her own eyes, nose, or mouth [1]. It is important to note that older individuals are highly susceptible to COVID-19 infection due to their diminished physical function, weakened autoimmunity, and many of them have underlying diseases. Statistics indicate that adults over 65 accounts for 80% of all hospital admissions and have a 23-fold higher risk of © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 116–127, 2023. https://doi.org/10.1007/978-3-031-34866-2_9

The After Effect of COVID-19 on Colour Perception

117

death than those under 65 [3]. In light of this, this thesis aims to raise social awareness of the impact of COVID-19 on the elderly by examining the consequences of COVID-19 in the elderly. There have been numerous studies on the consequences and effects of COVID-19 in older individuals. Rachel Ungar notes that COVID-19 increases stress and loneliness in the elderly due to decreased social interaction [4]. Marla K. Beauchamp has demonstrated experimentally that some older people infected with COVID-19 experience a decline in physical activity, with some unable to stand up or climb stairs [5]. However, there are few studies on the potential effects of COVID-19 on vision and colour vision in older individuals, with only Pierre Gascon [6] and Ilhan Bayazit reporting vision loss in middle-aged individuals [7]. It is evident from the survey that the majority of colour vision disorders are caused by heredity, disease, or injury. First, inheritance needs to be addressed in this paper. Secondly, it was discovered that certain eye diseases such as glaucoma and macular degeneration, specific brain and nervous system diseases such as Alzheimer’s disease or multiple sclerosis, specific medication effects such as Plaquenil (a drug used to treat rheumatoid arthritis), and eye or brain injuries could cause visual impairment [8]. COVID-19 has significant effects on the brain, including grey matter loss in the left parahippocampal gyrus, left orbitofrontal cortex, and left insula, as well as more significant grey matter loss in the cingulate cortex, central nucleus of the amygdala, and hippocampal cortex [9]. It has also been suggested that COVID-19 may contribute to the elderly’s elevated risk of Parkinson’s disease. [10] based on the causes of colour vision disorders, it can be concluded that COVID-19 is most likely to impact the colour vision of the elderly. In addition, studies on the ocular effects of COVID-19 have been conducted. For instance, Ann P. Murchison demonstrated that COVID-19 could cause acute vision loss in humans [11]. Vijairam Selvaraj suggested that patients infected with COVID-19 can experience ocular symptoms such as tearing, conjunctival congestion, and conjunctival oedema [12]. In addition, and Pierre Gascon discovered that the presence of COVID19 caused patients to experience acute vision loss, negative dark spots, and abnormal colour vision.6 Despite this, most studies have focused on the complications of COVID19 infection rather than its sequelae. In addition, little to no research has been conducted on the effects of COVID-19 on the colour vision of the elderly. The FM100 and Farnsworth D15 colour vision tests are commonly used to determine if something affects the subject’s colour vision because they are known to be simple and easy to administer [13]. Consequently, this experiment aims to determine whether COVID-19 affects colour vision in older individuals.

2 Method 2.1 Participants Ten older adults over 60 with a COVID-19 within the past year will comprise the experimental group, while ten older adults over 60 who have not got a COVID-19 within the past year will constitute the control group. It was essential to account for the fact that most of these older individuals resided in Scotland, United Kingdom, to prevent geographical differences from affecting their colour vision. It is also essential to ensure

118

C. Qin and A. Jiang

that the sex ratio of older people and the time to transition to COVID-19 are comparable to ensure that the variables are the same and that other factors have less of an impact on the experiment. In addition, it is essential that none of the test subjects has a history of colour vision disorder and that no family members have a history of colour vision disorder. They should not have any diseases that could cause colour vision disorders, nor should they have any habits, such as smoking or drinking, that could affect their colour vision. Before the experiment, subjects’ visual acuity was evaluated using the Snellen Visual Acuity Test Form, with 20/30 or better serving as the inclusion criterion for both groups [14]. Throughout the experiments, computer tests were utilized to ensure that the variables remained constant. 2.2 Equipment and Procedures Experimental Procedures. Before getting COVID-19 within a year, subjects were screened for colour vision disorders and a family history of inherited colour vision disorders, with clear medical reports and confirmation that the older control group had never had COVID-19. The FM100 Hue and Farnsworth D15 colour vision tests were then administered to the experimental and control groups, respectively, and the results were recorded. The test data were recorded, and a concluding comparison of value differences was conducted. FM100 Hue. The Farnsworth-Munsell 100 Hue Color Vision Test is a commonly used colour vision test for diagnosing colour blindness. It evaluates the ability to distinguish and align minute differences in various colour targets with constant values and chroma representing all the hues described by the Munsell colour system [15]. Before beginning the experiment, ensure that the light source in the test environment is constant, preferably using a D65-type light source. If testing online via a computer, ensure the screen monitor’s gamma is maintained at 2.2. Additionally, the subject should be devoid of visual impairments such as cataracts, eye surgery, etc. [16]. The most common test format consists of four rows of similar hues of various colours, with 25 distinct variations of each hue in each row. Each row’s most intense colour is fixed in the appropriate position to serve as an anchor point, and each colour block between the anchor points can be adjusted as the subject desires. The final configuration of the colour blocks represents the visual system’s capacity to distinguish between hue differences. Failure in the subject’s visual system can be measured by either the number of misplaced colour blocks or the degree to which the blocks are shifted (e.g., the distance between where the blocks should have been placed and where they were placed). The final statistic is the Total Estimated Score (TES), obtained by automatically counting the number of incorrectly placed blocks on the computer. The average TES score ranges between 30 and 40, and a score above 70 indicates colour blindness [16, 17]. Farnsworth D15 Test. The Farnsworth D15 is a classification test for colour vision disorders. One reference disc and fifteen numbered discs are included in each Farnsworth D15 test set. After the subject has attempted to arrange the discs in order, deficiencies in colour perception or deuteron, protan, or tritan discrimination are identified. The

The After Effect of COVID-19 on Colour Perception

119

subject can be classified as having severe/moderate colour vision or mild colour vision impairment based on the test results [18]. Before beginning the experiment, it is essential to ensure that the ambient light source is constant. The online test should ensure that the computer monitor’s gamma level is maintained at 2.2. During the experiment, the researcher must first place the reference palette (with a blank mark on the bottom) at the bottom of the box. The subject should then select the palette most similar to the reference palette and place it at the bottom of the box next to it. The subject selects the subsequent closest palette and positions it successively next to the reference palette [19]. Note that the participant should be given a reasonable amount of time to arrange the palettes and may be permitted to change the order before completion. However, the time should be limited to approximately 2 min. The score is completed by reading the reference palette and the number of the subject’s chosen palette and recording the order of the subject’s chosen palette on the scale. The palettes of subjects with colour vision disorders will be arranged differently than those with normal colour vision [20]. Any subject with a below-average score should be given a break and retested to ensure accurate results. On the second test sheet, the second test score must be recorded. Specifically, the Farnsworth D15 test enables the computer to automatically calculate the scores and form the following graphs, which differ significantly between those with and without colour vision impairment (see Fig. 1) [21].

Fig. 1. D15 color blindness test results (no date)

120

C. Qin and A. Jiang

Analysis. Due to the small sample size of the experimental population, non-parametric tests (Mann-Whitney U test and Wilcoxon matched pair test) [22] were utilized, and conclusions were drawn by comparing the results of the two tests for the experimental and control groups, taking the mean and median of the values and comparing their magnitudes. urate results, subjects with higher-than-average test scores should be given a second opportunity to take the exam. On the second test sheet, the second test score should be recorded.

3 Results Due to the physical limitations of older people compared to younger people, their poor eyesight, the limitations of comprehension, and the inability of many older people to concentrate for long periods, the FM100 Hue calculation for determining whether an older person has colour vision impairment would result in test results that are significantly different from usual when comparing subjects to controls, regardless of whether or not they have got COVID-19. In order to conclude, the FM100 Hue test was only used to compare the results of COVID-19 and non-COVID-19 seniors. It was determined that the test data for the non-COVID-19 older adults ranged from 20 to 60. It was confirmed that none of these older adults had a history of colour vision disorders or an inherited history of colour vision disorders. On this basis, the COVID-19 test was administered to older adults, and all results were within the normal range. The final results of the Farnsworth D15 test for COVID-19 and non-COVID-19 older adults were also comparable. Therefore, it was determined that COVID-19 does not affect the elderly’s colour vision. 3.1 FM100 Hue The results of the FM100 Hue experiment are shown. The following Table 1 and Table 2 gives the data. By comparing the test results of the control and experimental groups and taking the mean and median of the values of the two groups, it is concluded that the mean and median of the FM100 Hue test data of the control group is 35.6. The median is 34, and the mean and median of the FM100 Hue test data of the experimental group is 36.8. The median is 35, and based on this, it can be concluded that the experimental data of the two groups Based on this, it is clear that the data from the two groups are similar. Therefore the FM100 Hue colour vision test shows that COVID-19 does not affect the colour vision of the elderly. Table 1 FM100 test results for different elderly people n

Gender

Age

FM100 Hue

1

M

60

22

2

M

76

62 (continued)

The After Effect of COVID-19 on Colour Perception

121

Table 1 (continued) n

Gender

Age

FM100 Hue

3

M

75

24

4

F

68

44

5

M

64

26

6

F

78

42

7

F

74

48

8

M

63

24

9

F

64

22

10

F

76

42

Table 2 FM100 test results for different elderly people n

Gender

Age

Time since last COVID-19

FM100 Hue

1

F

61

6 months

28

2

M

60

3 months

60

3

M

77

5 months

42

4

F

69

3 months

24

5

M

75

7 months

24

6

F

77

3 months

48

7

F

67

6 months

22

8

M

69

6 months

28

9

F

74

4 months

48

10

M

79

5 months

46

3.2 Farnsworth D15 Below are the results of the Farnsworth D15 experiment (see Fig. 2 and Fig. 3). Comparing the results of the control and experimental groups, most individuals in both groups were found to have no colour vision impairment. Three individuals in the control group and four in the experimental group had minor errors that were not deemed colour vision impairment. The Farnsworth D15 colour vision test concluded that COVID-19 does not affect the elderly’s colour vision.

122

C. Qin and A. Jiang

1

Male 60

4. Female 68

7

Female 74

2. Male 76

3

Male 75

5 Male 64

6. Female 78

8. Male 63

9

10. Female 76

Fig. 2. Control group

Female 64

The After Effect of COVID-19 on Colour Perception

1. Female 61

2. Male 60

4. Female 69

5

7

Female 67

3. Male 77

Male 75

8. Male 69

10. Male 79

Fig. 3. Experimental group

6. Female 77

9

Female 74

123

124

C. Qin and A. Jiang

4 Discussion Using two colour vision tests, FM100 Hue and Farnsworth D15, this paper investigates whether COVID-19 can affect the colour vision of older people by causing colour vision problems such as colour vision impairment. Although previous cases have suggested that COVID-19 may cause colour vision disorders in older individuals, the investigation revealed that COVID-19 did not affect the colour vision of the vast majority of older individuals. This suggests that COVID-19 causing colour vision disorders in older individuals are not common and that the patients may have colour vision disorders or a gene for colour vision disorders. The current study demonstrates that COVID-19 does not affect the colour vision of older individuals who have recovered from the virus. Unlike previous studies, which were based on individual cases reported in hospitals and suggested that COVID-19 might affect older people’s colour vision, these effects were only observed during the period of COVID-19, and there were no studies to determine whether COVID-19 affected older people’s colour vision after recovery. This study, therefore, investigates whether COVID-19 causes colour vision sequelae in older individuals by recruiting 60-year-olds and older who have recovered from COVID-19 within the past year. It is important to note that colour vision impairment is irreversible, and it is crucial for older people to interact with others using their colour vision. Examples include common road signs, traffic lights, and studies indicating that many older people are unable to recognize when food changes colour during cooking, leading to situations such as undercooked food or fires [23]. In addition, due to cognitive decline in learning ability, short-term memory, attention and reaction time, the elderly have significant difficulty remembering maps, navigation, route learning, map learning, and place learning. Consequently, a study has been conducted on how older people perform in an unfamiliar environment to improve their ability to remember routes through colour interaction, as it has been discovered that colour is more valuable than form, particularly for special populations such as older people or people with dementia, for whom the colour advantage outweighs the form advantage. We are emphasizing this point by highlighting that distinguishing wall colours is an effective means of creating differentiation in a monochromatic environment [24]. Based on this, it can be determined that colour plays a greater role in the interaction of older people’s lives than it does in the lives of younger people, [25–28] therefore, if COVID-19 affects older people’s colour perception, it is crucial to be aware of this. It should be incorporated into future designs to help older people live better lives. However, there are some limitations to this study. First, this study has the problem of a small sample size because I the test is still somewhat difficult to administer in practice to older people because many of them are older, resulting in reduced vision and even hand tremors. Second, COVID-19 appeared so suddenly and spread so rapidly that many older people are infected with it. A more rigorous approach should have involved conducting a colour vision test before all the participants got COVID-19, to ensure they are not born with colour vision impairment [29–33]. Then a second colour vision test should be conducted after they got COVID-19, then record the test data, and comparing the results of the before & after experiments. The results of the two experiments were compared to reach more definitive conclusions.

The After Effect of COVID-19 on Colour Perception

125

Although there have been other cases of COVID-19 causing colour vision deficits in older individuals, these were complications rather than sequelae of the COVID-19 infection process. The fact that only one case has been made public is limiting. In addition, it is hoped that this study will raise awareness of the significance of colour vision disorders in the elderly and the significance of colour vision in their daily lives, where interaction with colour is essential [34, 35]. Future designs should therefore consider the interaction of the elderly with colour, as well as whether the elderly with colour vision impairment will behave and live differently than the regular elderly future designs can focus on how to interact with the elderly with colour vision impairment to improve their quality of life. In addition to colour blindness, older individuals with other visual impairments also require care. In addition, for future studies of a similar nature, care must be taken to conduct experiments both before and after the occurrence of a particular disorder so that more reliable conclusions can be drawn.

5 Conclusion This study establishes a control group and an experimental group to investigate if COVID-19 affects the colour vision of the elderly. The FM100 Hue and Farnsworth D15 colour vision tests reveal that COVID-19 does not affect the colour vision of elderly individuals. Although the final results indicate that COVID-19 does not affect the elderly’s colour vision, this study expands our understanding of the sequelae of COVID-19. It confirms that COVID-19 does not affect the elderly’s colour vision from a colour vision perspective.

References 1. Coronavirus disease (covid-19): How is it transmitted? World Health Organization. https:// www.who.int/emergencies/diseases/novel-coronavirus-2019/question-and-answers-hub/qa-detail/coronavirus-disease-covid-19-how-is-it-transmitted. Accessed 2023/02/03 2. Coronavirus (COVID-19) Google News. Google. https://news.google.com/covid19/map?hl= en-GB&gl=GB&ceid=GB%3Aen,last. Accessed 2023/02/03 3. Mueller, A.L., McNamara, M.S., Sinclair, D.A.: Why does covid-19 disproportionately affect older people? Ageing 12(10), 9959–9981 (2020) 4. Ungar, R., et al.: The impact of covid-19 on older adults: results from an annual survey. Geriatr. Nurs. 44, 131–136 (2022) 5. Beauchamp, M.K., et al.: Assessment of functional mobility after COVID-19 in adults aged 50 years or older in the Canadian Longitudinal Study on Aging. JAMA Network Open 5(1) (2022) 6. Jiang, A.O.: Effects of colour environment on spaceflight cognitive abilities during short-term simulations of three gravity states (Doctoral dissertation, University of Leeds) (2022) 7. Ilhan, B., Cokal, B.G., Mungan, Y.: Intracranial hypertension and visual loss following covid19: a case report. Indian J. Ophthalmol. 69(6), 1625 (2021) 8. Causes of colour blindness National Eye Institute. U.S. Department of Health and Human Services. https://www.nei.nih.gov/learn-about-eye-health/eye-conditions-and-diseases/color-bli ndness/causes-color-blindness. Accessed 2023/02/03 9. Douaud, G., et al.: SARS-CoV-2 is associated with changes in brain structure in UK Biobank. Nature 604, no. 7907 (2021)

126

C. Qin and A. Jiang

10. Meyer, P.T., et al.: Molecular imaging findings on acute and long-term effects of covid-19 on the brain: a systematic review. J. Nucl. Med. 63(7), 971–980 (2022) 11. Murchison, A.P., et al.: Monocular visual loss as the presenting symptom of covid-19 infection. Clin. Neurol. Neurosurg. 201, 106440 (2021) 12. Selvaraj, V., et al.: Acute vision loss in a patient with covid-19 (2020) 13. National Research Council (US) Committee on Vision.: CHAPTER 3COLOR VISION TESTS. In: Procedures for testing color vision: Report of Working Group 41. National Academy Press, Washington, D.C. (1981) 14. Huang, Z., Wang, S., Jiang, A., Hemingray, C., Westland, S.: Gender preference differences in color temperature associated with LED light sources in the autopilot cabin. In: HCI in Mobility, Transport, and Automotive Systems: 4th International Conference, MobiTAS 2022, Held as Part of the 24th HCI International Conference, HCII 2022, Virtual Event, June 26–July 1, 2022, Proceedings, pp. 151–166. Springer, Cham (2022). Doi: https://doi.org/10.1007/9783-031-04987-3_10 15. Kinnear, P.R.: New Farnsworth-Munsell 100 hue test norms of normal observers for each year of age 5–22 and for age decades 30–70. Br. J. Ophthalmol. 86(12), 1408–1411 (2002) 16. Zhang, N., Jiang, A.: Co-designing the next generation automatic driving vehicle HMI interface with lead-users. In: HCI in Mobility, Transport, and Automotive Systems: 4th International Conference, MobiTAS 2022, Held as Part of the 24th HCI International Conference, HCII 2022, Virtual Event, June 26–July 1, 2022, Proceedings, pp. 231–243. Springer, Cham (2022) 17. Farnsworth–Munsell 100 hue test Wikipedia. Wikimedia Foundation. https://en. wikipedia.org/wiki/Farnsworth%E2%80%93Munsell_100_hue_test#cite_note-6. Accessed 2023/02/03 18. Compact magnetic Farnsworth D-15 test (saturated) Gulden Ophthalmics. https://gul denophthalmics.com/product/compact-magnetic-farnsworth-dichotomous-d-15/. Accessed 2023/02/03 19. Hovis, J.K., Ramaswamy, S., Anderson, M.: Repeatability indices for the Farnsworth D-15 test. Visual Neurosci. 21(3) (2004) 20. Dain, S.J., Atchison, D.A., Hovis, J.K.: Limitations and precautions in the use of the farnsworth-munsell dichotomous D-15 Test. Optom. Vis. Sci. 96(9), 695–705 (2019) 21. Colorlite: D15 color blindness test: Color vision test Color Blind Glasses. https://www.colorl itelens.com/images/test/D15/D15.html. Accessed 2023/02/03 22. Nonparametric tests Corporate Finance Institute. https://corporatefinanceinstitute.com/resour ces/data-science/nonparametric-tests/. Accessed 2023/02/03 23. Jiang, A., Zhu, Y., Yao, X., Foing, B.H., Westland, S., Hemingray, C.: The effect of three body positions on colour preference: an exploration of microgravity and lunar gravity simulations. Acta Astronaut. 204, 1–10 (2023) 24. Süzer, Ö.K., Olguntürk, N.: The aid of colour on visuospatial navigation of elderly people in a virtual polyclinic environment. Color. Res. Appl. 43(6), 872–884 (2018) 25. Dewi, J., Prakoso, S., Heindri, N.: Integrated elderly park model with social interaction qualities. IOP Conference Series: Earth and Environmental Science 764(1), 012005 (2021) 26. Jiang, A., et al.: Short-term virtual reality simulation of the effects of space station colour and microgravity and lunar gravity on cognitive task performance and emotion. Build. Environ. 227, 109789 (2023) 27. Jiang, A., Yao, X., Westland, S., Hemingray, C., Foing, B., Lin, J.: The effect of correlated colour temperature on physiological, emotional and subjective satisfaction in the hygiene area of a space station. Int. J. Environ. Res. Public Health 19(15), 9090 (2022) 28. Jiang, A., et al.: Space Habitat Astronautics: Multicolour Lighting Psychology in a 7-Day Simulated Habitat. Space: Science & Technology (2022)

The After Effect of COVID-19 on Colour Perception

127

29. Yu, K., Jiang, A., Zeng, X., Wang, J., Yao, X. and Chen, Y.: Colour design method of ship centralized control cabin. In Advances in Human Aspects of Transportation: Proceedings of the AHFE 2021 Virtual Conference on Human Aspects of Transportation, July 25–29, 2021, USA, pp. 495–502. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-80012-3_57 30. Yu, K., Jiang, A., Wang, J., Zeng, X., Yao, X., Chen, Y.: Construction of crew visual behaviour mechanism in ship centralized control cabin. In: Advances in Human Aspects of Transportation: Proceedings of the AHFE 2021 Virtual Conference on Human Aspects of Transportation, July 25–29, 2021, USA, pp. 503–510. Springer, Cham (2021) 31. Lu, S., et al.: Effects and challenges of operational lighting illuminance in spacecraft on human visual acuity. In Advances in Human Aspects of Transportation: Proceedings of the AHFE 2021 Virtual Conference on Human Aspects of Transportation, July 25–29, 2021, USA (pp. 582–588). : Springer, Cham (2021) 32. Lu, S., Jiang, A., Schlacht, I., Ono, A., Foing, B., Yao, X., Westland, S., Guo, Y.: The effect on subjective alertness and fatigue of three colour temperatures in the spacecraft crew cabin. In Advances in Human Aspects of Transportation: Proceedings of the AHFE 2021 Virtual Conference on Human Aspects of Transportation, July 25–29, 2021, USA, pp. 632–639. Springer, Cham (2021) 33. Jiang, A., Yao, X., Hemingray, C., Westland, S.: Young people’s colour preference and the arousal level of small apartments. Color. Res. Appl. 47(3), 783–795 (2022) 34. Jiang, A., Foing, B.H., Schlacht, I.L., Yao, X., Cheung, V., Rhodes, P.A.: Colour schemes to reduce stress response in the hygiene area of a space station: a Delphi study. Appl. Ergon. 98, 103573 (2022) 35. Jiang, A., Yao, X., Schlacht, I.L., Musso, G., Tang, T., Westland, S.: Habitability study on space station colour design. In: AHFE 2020, pp. 507–514. Springer, Cham (2020). https:// doi.org/10.1007/978-3-030-50943-9_64

QR Codes as a Method for Older Adults to Access a Mobile Survey Alda G. Rivas1(B) and Anthony Schulzetenberg2 1 U.S. Census Bureau, Suitland, USA

[email protected] 2 LexisNexis, New York, USA [email protected]

Abstract. Current web surveys conducted by the U.S. Census Bureau require respondents to type in a URL and an authentication code. This task can be particularly difficult for older adults due to the fine motor movements required for this task. One potential solution may be to allow respondents to access web surveys through a Quick Response (QR) code. Although QR code adoption has increased since the onset of the COVID-19 pandemic, no standardized set of instructions has been developed to allow less-tech savvy individuals (e.g., older adults) to successfully use QR codes, and the usability of this technology for older adults has not been explored. In the present study (conducted in the United States between October and November 2022), 19 older adults (mean age = 63.68 years) attempted to access a survey by scanning a QR code in a paper invitation. This invitation also included step-by-step instructions on how to access the survey using the QR code. The results of this study showed an overall high level of usability (effectiveness, efficiency, and satisfaction) of QR codes. The most common usability problem was the lack of affordance of the tappable link that appears after scanning the QR code. Because different operating systems present the tappable link in different ways, and given the small space available in the invitation, it is not possible to provide detailed instructions to avoid this usability issue. Regardless of the usability issues encountered, most of the users stated that they would prefer to scan a QR code over typing a URL because it helped them avoid the typing errors that they usually commit when trying to type a web address in their smartphone. Given the high prevalence of QR codes, and the findings of the present study, including QR codes in survey invitations is a viable method to increase the effectiveness, efficiency, and satisfaction with which older adults access a survey on a mobile device. Keywords: Older adults · emerging technologies · usability · QR code · smartphones

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 128–141, 2023. https://doi.org/10.1007/978-3-031-34866-2_10

QR Codes as a Method for Older Adults to Access a Mobile Survey

129

1 Introduction 1.1 Background Currently, 93% of adults in the United States conduct activities on the internet, and 85% of adults have a smartphone that connects to the internet [6, 7]. Given these facts, and the efficiency of conducting activities online, it is not surprising that many adults choose to complete tasks online, when there is an online option. One example of this phenomena is the most recent decennial census in the United States. By May 27, 2020, approximately 80% of self-respondents had completed the census online [11]. Although the proportion of respondents who completed the online census on a smartphone versus personal computer is not yet released, the prevalence of smartphones in American households gives reason to suspect that a non-trivial proportion of respondents may have used their smartphones to complete the census. Recently, smartphones became the computing device owned by most households (84%) in the United States [12], and 71% of older adults ages 65–74 reported owning a smartphone [7]. In fact, for 15% of all adults and 12% of older adults (over 65 years of age), a smartphone is the only device available to connect to the internet [8]. When completing the decennial census in 2020, respondents received a letter in the mail which contained a website address and a Census ID (an alphanumeric string with 12 characters). To access their form online, responds were required to type the census URL and then enter their unique Census ID. These manual inputs could be problematic when attempting to complete the survey through a smartphone because of the small size of the keyboard. This activity may be particularly challenging for older adults due to the decrease in fine motor movements associated with healthy aging [2]. One potential solution to motivate older adults to complete surveys on a smartphone without having to type a web address or an authentication code, is to include a Quick Response (QR) code as an option to access a survey. QR codes are two-dimensional barcodes that store digital information about different types of products (e.g., a website) and are read through a scanning device (e.g., a smartphone camera). Although QR codes have existed for almost three decades, there has been an increase in QR code use since the COVID-19 pandemic in the United States, where in 2021, almost 76 million (25.5%) smartphone users scanned a QR code [10, 14]. QR codes have been explored before as a potential venue to access a survey through a smartphone. In 2018, a study [3] compared the response rates for three different conditions presented in survey invites: one condition included a URL, a second condition included URL and a QR code, and the third condition included only a QR code. All conditions also included the option to complete the survey on paper. Although the results did not find a statistically significant difference in response rates between the three conditions, the data indicated that only 4% of respondents had completed the survey through a QR code, and that the QR code-only condition showed the highest use of the paper alternative [3]. Here, it is worthy of note that the QR codes in the study described above only included the instruction to “Scan the QR code to access the website and complete the survey online.” Given that the study was conducted before the increase in QR code usage due to the COVID-19 pandemic, it is possible that a lack of familiarity with QR codes,

130

A. G. Rivas and A. Schulzetenberg

and the lack of specificity in the instructions led to the low response rates for the QR code condition in this study. Specificity in instructions is particularly important to promote a new technology when the potential users include older adults. This is because older adults show a higher level of aversion to technology than younger adults [4]. Although QR codes are more popular now than in recent years, their presence in everyday activities is relatively new. Therefore, it is possible that older adults have been reluctant to adopt this technology. Recently, researchers at the U.S. Census Bureau developed step-by-step instructions to access a survey using a QR code [9]. These instructions were presented to 20 experienced QR code users. The users were asked their opinion about the instructions and asked whether they thought that a person without QR code experience would be able to follow the instructions. Although all users described the instructions as easy and clear, most users specifically stated concern that older adults or less-tech savvy individuals may not be able to understand some of the wording in the instructions. Based on this feedback, the instructions were modified specifically with the aim to allow less-tech savvy individuals to access a survey through a QR code. The purpose of the present study is to explore the usability of QR codes as method for older adults to access an online survey, and to obtain feedback about the wording in the modified instructions. Specifically, we aim to answer the following two research questions: 1) What is the effectiveness, efficiency, and satisfaction with which older adults access a mobile survey through a QR code? And 2) Is the current wording in the instructions useful for a person without QR code experience?

2 Methodology Older adults (at least 55 years old) from a metropolitan area in Texas were recruited for this study. The rationale for the age requirement is that older age is associated with a decline in different cognitive abilities (e.g., processing speed, episodic memory) and this decline accelerates between the ages of 50 and 60 [5]. Recruitment took place between October and November, 2022, at different community centers that offered activities for senior citizens. Users were recruited through flyers and were also verbally invited to participate (intercepted) while they attended activities at the community centers. Requirements for participation included being at least 55 years old, being fluent in English, no diagnosis of dementia or Parkinson’s, having a smartphone, having accessed the internet through their smartphone during the last year, and having used the camera in their smartphone. Interested users that met the requirements for participation completed the usability session immediately after the screening process. In order to prevent naïve users (participants without previous QR code experience) from being discouraged to participate, no mention of QR code was made in the flyers, nor in the verbal invitation. The flyers and the verbal invitation only mentioned that their feedback was being requested about the instructions to access a survey through a smartphone. Sessions lasted approximately 20 min and were conducted individually: only one user participated at a time, with one researcher leading the session and taking notes. After completion of the usability session, users received a $20 cash incentive.

QR Codes as a Method for Older Adults to Access a Mobile Survey

131

2.1 Participants From 20 potential participants, 19 (3 males, 16 females) met the requirements for participation and completed the usability session. The average age of the sample was 63.68 years (SD: 7.8; range: 55–76 years). On average, for level of comfort using a smartphone, users reported a score of 2.05 (SD = 1.05, range = 1–4) on a 1–5 Likert scale (1 = “Very comfortable” – 5 = “Very uncomfortable”). For frequency of internet usage on their smartphone, users reported an average score of 4.1 (SD = 1.18, range = 2–5) on a 1–5 Likert scale (1 = “Never” – 5 = “Almost constantly”). Table 1 provides other basic demographic characteristics of the sample. Table 1. Demographic characteristics (race and education) of the sample. Race

Number of users

White

14

Black or African American

4

Asian

1

Hispanic Origin

6

Education Some High School

2

High School or equivalent

2

Vocational or Technical degree

2

Some college

5

Associate’s degree

1

Bachelor’s degree

4

Master’s degree

1

Doctorate degree

2

2.2 Design The design of this study was observational. The researcher took notes on outcomes of interest while the users completed the usability task. These notes included quantitative data for usability outcomes (e.g., effectiveness, efficiency, and satisfaction), as well as qualitative data describing behaviors that indicated a usability issue (e.g., users’ comments indicating confusion or inability to use the QR Code). The researcher also made notes on users’ answers to debriefing questions and general feedback about the wording of the instructions to access the mobile survey through the QR Code.

132

A. G. Rivas and A. Schulzetenberg

2.3 Measures and Outcomes Quantitative and qualitative data were recorded for each of the usability evaluations. Quantitative outcomes included effectiveness, efficiency, satisfaction, and previous QR code experience. The operational definitions of these outcomes are presented below: Effectiveness. Proportion of participants who successfully accessed the survey through the QR Code. This outcome was measured as a dichotomous variable (Yes/No) depending on whether the user was able to access the survey in their mobile device. Efficiency. The average amount of time (in seconds) for users who successfully accessed the survey through the QR Code. Measured from the moment the user opened the camera in their smartphone to the moment the survey appeared in the smartphone screen. Satisfaction. This outcome was recorded through users’ response to two different questions: Difficulty of Using a QR Code. Users’ response to the question “Overall, how easy or difficult was it to use the QR Code to access the survey in your smartphone?” Response choices for this item included a 7-point Likert scale (1 = “Extremely easy” and 7 = “Extremely difficult”). Likelihood of Using a QR Code in the Future. Users’ response to the question “If presented with the option, how likely are you to use a QR Code to access a survey on your smartphone?” Response choices for this item included a 7-point Likert scale (1 = “Extremely likely” and 7 = “Extremely unlikely”). Previous QR Code Experience. Recorded as users’ response to four different questions: 1) Before today, did you know what a QR code was? (Yes/No); 2) Before today, have you used a QR Code? (Yes/No). 3) In the past year, on average, how frequently did you use QR codes? Response options for this question included daily, weekly, monthly, twice per year, once per year, and open ended; 4) “In general, how comfortable are you using QR codes?” Response options for this question included a 7-point Likert scale (1 = “Very comfortable” and 7 = “Very uncomfortable.”). Besides the quantitative outcomes described above, qualitative data were also recorded. These data included users’ behaviors or comments signaling a usability issue, their experience with the QR process, and feedback about the wording in the instructions.

2.4 Procedure Users who met the requirements for the study signed an informed consent and completed the usability session. During each session, users were handed a paper invitation to complete the census. This invitation (Fig. 1) included a QR code and the modified instructions (to the left of the QR code). The QR code in the paper invitation measured one square inch. After receiving the invitation, users were told the following instruction: “Imagine that you received this paper in your mailbox and you wanted to complete the census using your smartphone. What would you do?” After reading the instructions, users described their interpretation of the instructions and then attempted to access the survey in their

QR Codes as a Method for Older Adults to Access a Mobile Survey

133

Fig. 1. Invitation used during the usability task.

smartphone. Users were instructed to think aloud while they completed these tasks. Successfully scanning the QR code led the user to a short survey with questions about the demographic information described in the “Participants” section of this paper.

3 Findings 3.1 Previous QR Code Experince Although only two users reporting knowing that the image was called a “QR code”, all users reported seeing QR codes before and being familiar with their purpose (accessing webpages) and the process of how to use it (requiring a phone). Approximately half of the sample (53.6%, 10 users) reported having used a QR code themselves at least once. Nine users reported not having used a QR code themselves, but having seen them before (e.g., at restaurants) and having seen someone else using them. Users that had previous experience reported an average of 3.3 for level of comfort using QR codes (SD = 2.3; range = 1–7) on a 1 -7 Likert scale (1 = “Very comfortable” - 7 = “Very uncomfortable”). For average use of QR codes in the past year, experienced users reported a wide range: two users reporting using them weekly, two users reported using them monthly, five users reported using a QR code twice, and one user reported not having used a QR code in the past year. 3.2 Usability Effectiveness. From the 19 users who completed the usability session, 14 users (74%) successfully accessed the survey after scanning the QR code. From these 14 users, 9 had

134

A. G. Rivas and A. Schulzetenberg

previous experience, and 5 did not have previous experience using QR codes. So, five of the nine users who had never used a QR code and nine of the 10 users who had used a QR code before, were successful in accessing the survey. On average, users took 1.6 (SD = 0.9; range 1–4) attempts at scanning the QR code before accessing the survey. Efficiency. Users who were able to access the survey did so in an average time of 31.2 s (SD = 33.3 s; range = 4 s – 128 s). Satisfaction. When rating the difficulty to access the survey through the QR code, users reported an average difficulty of 2.1 (SD = 1.3; range = 1–5) on a 1 – 7 Likert scale (1 = “Extremely easy” and 7 = “Extremely difficult”). About their preference to access a survey in the future, 84.2% of the sample (16 users) stated that they preferred to use a QR than typing; 10.5% of the sample (2 users) stated that they preferred typing than scanning a QR code, and 5.3% of the sample (one user) stated no preference. For the users who preferred typing, one user stated that they just did not like QR codes because they do not know how to use them. The second user stated they preferred typing because they do not have enough practice with QR codes but that they could see the advantage (faster, less errors) of QR codes over typing. For likelihood of using a QR code to access a survey in the future, users reported an average score of 2.21 (SD = 1.84, range = 1–7) on a 1 -7 Likert scale (1 = “Extremely likely” and 7 = “Extremely unlikely”).

3.3 Usability Issues During this evaluation, four main usability issues were observed. Some of these issues led to users not being able to complete the task, and some issues delayed users in task completion. Table 2 provides information on how these different issues affected users’ performance. The table also specifies whether the affected users did or did not have previous experience using QR codes. Description of each usability issue is presented after the table. The description of the usability issues is followed by a recommendation to improve the instructions in the invitation. Not Recognizing the Link. After focusing the QR code with the phone’s camera, a notification with the link appears on the screen. However, different operating systems present this notification in different ways. For some users, the notification was presented on top of the screen. For other users, the notification appeared at the bottom of the screen. Finally, for other users, the notification appears close to where the QR code is being focused on the screen. This latter notification is small (relative to the size on the screen) and users are not aware that the link is already available in the screen. This led to users thinking that they had not properly focused the QR code. Figure 2 presents a visualization of some of the different ways in which the notification with the link is presented across different operating systems/browsers (notification is outlined in red). Recommendation. In order to address this issue, a description could be provided for each of the possible ways in which the notification may be presented by different operating systems and browsers. However, this solution is unrealistic when considering the limited space in the paper invitation. The ideal solution to this issue would be for operating system developers to present a standardized notification across systems in a manner that

QR Codes as a Method for Older Adults to Access a Mobile Survey

135

Table 2. Usability issues by effect on performance and level of users’ experience Unable to complete task

Delayed in task completion

Usability issue

Experienced users

Inexperienced users

Experienced users

Inexperienced users

Not recognizing the link

1

1

5

3

1

1

3

3

Difficulty focusing QR code

2

Having to specify a browser

1

Taking a picture of QR code

Fig. 2. The notification with the link is presented in different ways across different operating systems/browsers (from left to right: at the top; at the bottom; close to QR code).

is easily identified by users (i.e., larger notification, underlining the address to make it look like a link, blue font). Difficulty Focusing the QR Code on the Camera. One user tried to focus the whole invitation in the camera, and one user physically centered the QR code right behind the phone (instead of centering the QR code in the phone screen). This led to the user focusing on the postage stamp (located above the QR code) instead of the QR code in the center of the screen. Recommendation. Include a label on top of the QR code to make more obvious the image that should be centered. Additionally, include in the instructions an image showing how the QR code would look like when focused on the center of the screen. Figure 3 provides an example of these recommendations.

136

A. G. Rivas and A. Schulzetenberg

Fig. 3. Visualization of recommendation to label the QR code and include a graphic depicting how a focused QR code would look in the smartphone screen.

Having to Specify a Browser. After a QR code is scanned, some operating systems provide a notification with the link (in light gray font) but do not allow the link to be opened until the user has selected a browser (the link is presented, but it is not active). The user must tap on “Show options” first and then select the browser before being able to open the link. Users who encountered this issue kept tapping in the inactive link and it was not obvious to them that they had to select a browser first. Figure 4 provides a visualization of this type of notification.

Fig. 4. Example of notification with inactive webpage link.

Recommendation. This issue is dependent on the settings that the user has specified in their smartphone. Including instructions to troubleshoot how to select a default browser in their smartphone settings is unrealistic because of the number of steps involved in that process, and the limited space available in the paper invite. One alternative solution would be for operating system developers to modify the “Show options” text to “Show options to open webpage.” Another potential solution is for developers to include text

QR Codes as a Method for Older Adults to Access a Mobile Survey

137

with instructions to select the browser (e.g., “Select a browser to open the webpage) along with icons for the browsers available to the user. Taking a Picture of QR Code. One user took a picture because they thought that it was part of the process. Five users took a picture even though they knew this was not part of the process. These five users did not see the notification with the link, so they stated that they took a picture because “there was nothing else to do.” Recommendation. A standardized presentation of the notification (as recommended above) may address the issue of users taking a picture. Unfortunately, standardizing the presentation of the link is up to the technology companies that own the operating systems. In the meantime, the instructions could specify to users that they do not need to take a picture. Although the main usability issues cannot be addressed with brief instructions, some of the issues (not knowing where the QR code is, how to focus the QR code on the screen, and not needing to take a picture) may be addressed without taking much space in the paper invitation. Figure 5 presents the instructions used in this study and the instructions with modifications to address some of these usability issues encountered in this study.

Fig. 5. Instructions used in the present study (top). Instructions modified with recommended changes (bottom).

3.4 Debriefing Data After completing the usability task, users were asked to describe the situation where they used or saw QR codes for the first time. User were also asked follow-up questions about their experience accessing the survey through the QR code. After this, users provided their feedback about the wording in the instructions and their opinion about whether they thought that someone without experience would be able to follow the instructions. Context Where Users Saw/Used QR Codes for the First Time. Users reported a wide range of activities as the context where they used or saw QR codes for the first

138

A. G. Rivas and A. Schulzetenberg

time. Two users reported using QR codes as part of their job to access different types of information, one user reported using QR codes to check in at the doctor’s office, one user reported using them for attendance in a continuing education class, and two users reported using them to access resources at a community center’s website. The highest frequency of QR code usage, however, were restaurants and grocery stores, with 14 users reporting being in one of these contexts the first time they either saw or used a QR code. Nine of the ten users with previous experience reported that someone had to show them how to use them. Only one user reported learning by themselves (by trial and error). Most users described the scanning process as “easy to learn” and only needing “a couple of tries” before being able to scan QR codes by themselves. Impressions About the QR Code Process. After being asked to describe their experience scanning the QR code during the usability task, users reported being “happy” or “excited” about having an alternative to typing. Most users stated that scanning was faster and much easier than typing. Five users spontaneously reported that scanning is better because it helps them avoid the typing mistakes that they usually commit when trying to type on a smartphone. However, three users specified that they would only scan a QR code if they knew that it would take them to a trusted website (e.g., government website) because they are aware that seniors are usual targets for scams. Feedback About Wording in the Instructions. After being asked their opinion about the instructions in the invitation, 16 users (84% of the sample) thought that the instructions were “clear”, “pretty simple”, “good”, and “easy to follow”. Twelve users (63% of the sample) thought that someone without QR code experience would be able to follow the instructions to use a QR code. Some recommendations to improve the instructions included changing the word “link” to “pop-up”, specify that a picture should not be taken, include a naming label either on top or bottom of the QR code, and specify to not put the phone too close to the QR code. Two users also stated concerns that people may not read the instructions carefully, or at all, because the instructions would have to be lengthy. Two users stated that some older adults may only attempt to use a QR code for the first time if someone is around to help them should they encounter issues. These users recommended a video tutorial that they could access in a page that is trusted by seniors (e.g., their community center’s page) or provide an in-person training at the community center.

4 Conclusions The present study aimed to answer the questions “What is the effectiveness, efficiency, and satisfaction with which older adults access a mobile survey through a QR code?” and “Is the current wording in the instructions useful for a person without QR code experience?”. Regarding the first question, we observed a high level of effectiveness, with most of the sample (74%) successfully accessing the survey through the QR code, including over half of the users without previous QR code experience. This level of effectiveness gives reason for optimism about the usability of QR codes, when considering that only four users (21% of the sample) reporting using a QR code more than twice during the

QR Codes as a Method for Older Adults to Access a Mobile Survey

139

previous year. We also observed a high level of efficiency. Previous studies exploring the usability of QR codes with a younger and experienced sample [9] showed that the average time for a successful scan was 12.4 s. This is noticeable less time than the average time for scan observed in this study (31.2 s). However, the larger time on task observed in the present sample is not surprising considering the changes in motor control associated with older age, and the fact that about 50% of the sample in the present study had never used QR codes before. Furthermore, the time on task was perceived by users as one of the main factors affecting satisfaction: Most users stated a high likelihood to use a QR code in the future and preferring to scan a QR code than to type a URL on their smartphone because the QR code process is easier and faster and helps them avoid typing errors. Given these findings, we conclude that the process of scanning a QR code to access a survey has a high level of usability for older adults and naïve users. Regarding the wording in the instructions, the instructions used in this study successfully guided about 50% of the naïve users through the QR code scanning process. Users described the instructions as “straight-forward”, “simple”, and “easy to follow”. The main shortcoming of these instructions was that they did not provide guidance to address the main usability issue observed in this study (lack of affordance of the tappable notification). Unfortunately, addressing this issue in the instructions would require including an exhaustive list of all the different presentations of the tappable link across different operating systems. Due to the limited space available for instructions in the invitation, addressing this usability issue in the instructions is not feasible. Other minor usability issues (e.g., specifying that a picture should not be taken) were addressed in the instructions taking into consideration the feedback from users. However, further testing is necessary to determine whether these changes can improve the performance of naïve users. Although the instructions we developed did not successfully guide all naïve users through the QR code scanning process, the data collected in this study indicates that QR codes are a useful tool for older adults to access a survey or website. Furthermore, when users recalled their own experience learning to use QR codes, they described the process as fast and easy to learn and expressed optimism about other older adults having little difficulty when learning the process. Older adults are sometimes described as “tech-averse” [1], and age has been shown to be a moderating factor when predicting acceptance of new technology [13]. However, all users in the sample reported, at a minimum, seeing QR codes before, and more than half of the sample reported having used them before. The results of the study indicated a high level of usability, easiness, and likelihood of using a QR code in the future for users in this sample. Additionally, most of the sample (including naïve users) reported preferring scanning a QR code over typing on their smartphone to access a website, including users who were unsuccessful at scanning the QR code during the usability task. Given the current prevalence of QR codes, this technology has become a familiar process for older adults. This, coupled with the high usability observed in this study, and its ease of learning, indicates that QR codes may be a viable alternative for older adults to access online surveys.

140

A. G. Rivas and A. Schulzetenberg

5 Limitations One limitation for this study involves the small sample of naïve users. From a sample of 19 users, only about 50% (9 users) reported not having using a QR code themselves prior to the study. Because of this, it is possible that the rates of effectiveness, efficiency and satisfaction for naïve users observed in this study may be different than the rates of these outcomes in the population. However, our inability to recruit a larger sample of inexperienced users may also serve as evidence of the prevalence of QR codes in the common activities of older adults. This is good news for entities that cater electronic products (e.g., websites, surveys) to older adults because it provides them with a tool to aid their users in accessing their products with high satisfaction. A second limitation of this study involved our inability to substantially improve the instructions. Because the main usability issues in this study stem from unstandardized practices across different operating systems and browsers, it was not feasible to address these issues in the limited space allowed for instruction in the invitation. These usability issues can be addressed by the owners of the technologies developing standardized practices (e.g., presenting the tappable link with a specified color, size, location, etc.) This would improve the usability of QR codes and eliminate the need for instructions to address the main usability issues observed in this study.

6 Implication for Future Research Future studies that wish to explore the usability of QR codes for older adults or naïve users may benefit from recruiting a larger sample. This would allow researchers to make more accurate inferences about the effectiveness, efficiency, and satisfaction with which this population interacts with QR codes. In the present study, we developed instructions to aid naïve users in accessing a survey using a QR code. However, as suggested by some of the users in this study, other methods to guide this population to access an electronic product may include short video tutorials or short in-person training community centers. Another possibility for future research may involve manipulating the characteristics (e.g., font size, color, location) of the tappable link that appears after scanning the QR code. This research may provide guidance for technology owners about the features in the QR code scanning process that show the highest usability and, hopefully, would spark a conversation regarding standardizing practices across operating systems and web browsers. Acknowledgements. Disclaimer: Any opinions and conclusions expressed herein are those of the author and do not reflect the views of the U.S. Census Bureau. Disclosure number CBDRBFY23-CBSM002-005.

QR Codes as a Method for Older Adults to Access a Mobile Survey

141

References 1. Haederle, M.: Technology fear stops older adults from logging on AARP (2011). https://www. aarp.org/technology/innovations/info-08-2011/elderly-fear-of-technology.html. Accessed 30 Jan 2023 2. Corti, E.J., Johnson, A.R., Riddle, H., Gasson, H., Kane, R., Loftus, A.M.: The relationship between executive function and fine motor control in young and older adults. Hum. Mov. Sci. 51, 41–50 (2017). https://doi.org/10.1016/j.humov.2016.11.001 3. Marlar, J.: Do Quick Response codes enhance or hinder surveys? Gallup (2018). https:// news.gallup.com/opinion/methodology/241808/quick-response-codes-enhance-hinder-sur verys.aspx. Accessed on 30 Jan 2023 4. Hanson, V.L.: Influencing technology adoption by older adults. Interact. Comput. 22, 502–509 (2010). https://doi.org/10.1016/j.intcom.2010.09.001 5. Park, D.C., Reuter-Lorenz, P.: The adaptive brain: aging and neurocognitive scaffolding. Annu. Rev. Psychol. 60, 173–196 (2009). https://doi.org/10.1146/annurev.psych.59.103006. 093656 6. Pew Research Center: Internet/Broadband Fact Sheet (2021). https://www.pewresearch. org/internet/fact-sheet/internet-broadband/?menuItem=9a15d0d3-3bff-4e9e-a329-6e328b c7bcce. Accessed 30 Jan 2023 7. Pew Research Center: Mobile Fact Sheet (2021). https://www.pewresearch.org/internet/factsheet/mobile/. Accessed 30 Jan 2023 8. Pew Research Center: Mobile Technology and Home Broadband 2021 (2021). https:// www.pewresearch.org/internet/2021/06/03/mobile-technology-and-home-broadband-2021/. Accessed 30 Jan 2023 9. Schulzetenberg, A.J., Rivas, A., Peterson, A.: Usability of QR codes as a method to access a mobile survey (2022). Manuscript in preparation 10. Ceci, L.: U.S. smartphone users scanning QR codes 2019–2025. Statista (2022). https://www. statista.com/statistics/1297768/us-smartphone-users-qr-scanner/. Accessed 30 Jan 2023 11. U.S. Census Bureau: Over 60 Percent of U.S. Households Have Responded to the 2020 Census. (2020). https://www.census.gov/newsroom/press-releases/2020/2020-census-60-per cent.html. Accessed 30 Jan 2023 12. U.S. Census Bureau: Computer and Internet Use in the United States: 2018 (2021). https:// www.census.gov/newsroom/press-releases/2021/computer-internet-use.html. Accessed 30 Jan 2023 13. Venkatesh, V., Morris, M.M., Davis, G.B., Davis, F.D.: User acceptance of information technology: toward a unified view. MIS Q. 27, 425–478 (2003). https://doi.org/10.2307/300 36540 14. Tiger, Q.R.: QR Code Usage Statistics 2022: 443% Scan Increase and 438% Generation Boost (2022). https://www.qrcode-tiger.com/qr-code-statistics-2022-q1. Accessed 30 Jan 2023

The Effects of an Art Program on Older Adults’ Cognition and Satisfaction with Life Fang-Wu Tung and Liang-Ming Jia(B) National Taiwan University of Science and Technology, No. 43, Keelung Rd., Sec. 4, Da’an Dist, Taipei 10607, Taiwan {fwtung,D10810801}@mail.ntust.edu.tw

Abstract. This study aimed to develop an art program to investigate the impact on older adults’ cognition and satisfaction with life. This study adopted the participatory case study approach and developed an eight-week art program (visual art - nostalgic topic, combination of visual art and music - linking color and music topic). The program was conducted at a community care station in Xinzhuang District, New Taipei. A total of 21 participants were recruited, 15 (aged 50–90) of whom participated in the entire eight-week program. A pretest and posttest were used to understand the program’s impact on participant cognition (MoCA) and satisfaction with life (SWLS). The results show that the participants significantly improved cognition after the art program. Their general cognition (p = 0.028) increased from 25.07 (3.43) to 27.47 (2.67); delayed recall (p = 0.005) increased from 2.87 (1.77) to 3.93 (1.53); and orientation (p = 0.028), improved from 5.53 (0.64) to 5.93 (0.26). There was no significant difference in the participants’ satisfaction with life scores, but there was a certain degree of increase from 27.93 (5.44) to 29.67 (4.42). The programs designed for this study can be applied to the learning programs of older adults to enhance their physical and mental health and quality of life. This study provides empirical evidence and practical experience for the intervention of art programs in older adults’ later life. Keywords: Art programs · Older adults · Recognition · Satisfaction with life · Visual arts · Music · Nostalgia · Association with color and music

1 Introduction Throughout this century, population aging has become a common phenomenon around the world; almost all areas of society, in both developing and developed countries, have been significantly affected by aging. Taiwan is no exception; according to the latest population projection report released by the National Development Council [1] in 2022, the total population of Taiwan reached a peak of 23.6 million in January 2022, after which it showed negative population growth. It is expected to enter a super-aged society in a few years (the proportion of older people will exceed 20% of the total population by 2025, entering an ageing society). The continuous aging of population will engender challenges in social development and become an important and urgent issue that needs © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 142–153, 2023. https://doi.org/10.1007/978-3-031-34866-2_11

The Effects of an Art Program on Older Adults’

143

to deal with. The degeneration of aging causes many inconveniences in the lives of older adults and threatens their physical and mental health [2]. Cognitive deterioration is a global health burden, and the effectiveness of the current medication is not satisfactory [3]. These are the negative factors affecting the satisfaction with life of older adults [4]. This requires the development of various programs to prevent cognitive impairment and enhance the physical and mental health and quality of life of older adults [5]. Among various therapies and approaches, there is evidence that active participation in creative activities brings benefits, including improved well-being, quality of life, health, and socialization. As an important component of creative activities, art programs are often used as an adjunct to general therapeutic measures as they could maintain the physical and mental health of older adults through non-pharmacological (non-invasive) approaches [2], and promote healthy behaviors to delay cognitive deterioration [6]. Art programs are available for a wide range of people. Erikson and Erikson [7] pointed out that art programs are very suitable for older adults, as they provide creative experiences and sensory stimulation to enrich their daily lives. Meanwhile, they can also enhance their quality of life and satisfaction [4]. Among the art programs for older adults, visual arts and music are the most popular [8], suggesting that the two art forms attain a higher degree of acceptance in older adults. Therefore, this study planned to design and develop an art program for older adults to investigate the impact on their cognition and satisfaction with life. The structure of the remainder of this paper is described below. In Sect. 2, we introduce a review of the literature on older adults’ cognition, life satisfaction, and art programs for older adults. Section 3 describes the research methodology of this study. Section 4 describes how we designed, developed, and implemented the program through a participatory case study, and obtained the impact of the program on participants’ cognitive and life satisfaction through pretest and posttest results. Finally, in Sect. 5, we discuss the results and describe the limitations and recommendations of this paper.

2 Literature Review 2.1 Cognition and Satisfaction with Life of Older Adults Cognitive function is a collective term for human information-processing abilities, including perception, memory, learning, reasoning, problem-solving, and verbal expression [9]. It is common for cognitive function to decline as the average lifespan increases in older adults and is associated with decreased performance in instrumental activities of daily living (IADLs) [10]. Cognitive functions, such as information processing speed, attention, episodic memory, spatial ability, and executive functions, show a continuous, regular decline with age [11]. It could result in emotional vulnerability, affect daily activities, and lead to stagnation of personal communication in older adults [12]. On the other hand, Khodabakhsh [13] pointed out that factors such as cognitive ability, aging, and daily living activities were associated with the life satisfaction of older adults and that declining cognitive ability would affect their life satisfaction. Satisfaction with life is an individual’s self-evaluation of life experiences and is one of the most critical indicators of subjective well-being. Participation in programs could thus positively affect the life satisfaction of older adults [14].

144

F.-W. Tung and L.-M. Jia

Among the various non-pharmacological interventions, art programs as a part of creative activities may be a practical intervention [15]. Hanna et al. [6] suggested that participation in art programs by older adults may encourage health-promoting behaviors (physical and mental stimulation, social engagement, stress reduction) that may delay cognitive decline. Moreover, it would enhance and encourage residual abilities in older adults by stimulating the visual cortex [16]. The studies by Brandmeyer [17], Hickson and Housley [18], and Adams-Price et al. [19] indicated that participation in art programs develops creativity in older adults as they age, and positively affects their satisfaction with life, happiness, and health. Çetinkaya et al. [4] evaluated the effect of an eight-week art program on the cognition and satisfaction with life of older adults in a community care center through a controlled experiment. The results showed a statistically significant increase in the cognitive level of the older adults in the experimental group (who participated in the art program) compared to the control group, and no statistically significant increase in satisfaction with life. However, it still had a positive effect on their quality of life. 2.2 Art Programs for Older Adults Among the different art forms often adopted in art programs for older adults, visual art and music are more common [5, 20–22]. Visual art includes painting, coloring, collage, art appreciation, sculpture, photography, and so on [8]. Meanwhile, in the development of art programs for older adults, the topic of nostalgia is often used, which could stimulate the long-term memory of older adults [23] and generate positive emotions through tangible and familiar connections to their past life [24]. It could help delay cognitive decline in older adults [25]. Forms of music participation for older adults range from listening, singing, and playing music [8]. Participation in music programs could positively impact the overall health of older adults [26, 27]. It may therefore be a useful and effective means of improving cognitive function and social skills in older adults [28]. Bugos et al. [29] found in a controlled trial that participation in music (experimental group) enhanced the attention, concentration, and planning abilities of older adults and reduced their normal age-related cognitive decline. On the other hand, with the development of various art programs for older adults, scholars [30] have suggested that older adults could try to experience various art form programs. Sandak et al. [31] suggested combining musical activities with artistic creation to enrich the experience of older adults through the synaesthesia of color and music [32]. Meanwhile, scholars also presented an integrated audio-visual way to develop art programs for older adults [31, 33]. For example, scholars Jia and Tung [33] explored the experience of audiovisual synaesthesia of older adults by designing a color-sensor play device called FEEL2 (FEEL2 is a musical instrument that is played by arranging the colors on a turntable) and corresponding activities involving music and art creation to enrich their daily lives. Therefore, this study intended to adopt visual art (nostalgic topic) and the combination of visual art and music (linking color and music topic) to develop an art program for older adults.

The Effects of an Art Program on Older Adults’

145

3 Research Methods This study adopted a participatory case study approach. It could cover participants, local groups, communities, and other stakeholders. Meanwhile, the participants in the case could also become contributory participants. Reilly [34] pointed out that this method could address specific issues in the development process, ideally through the participation of local people, and its findings could be directly applied to the problems encountered in the present. The researchers served as program designers, implementers, data collectors and analysts to enable the effective and smooth implementation of the program or the research team’s work. 3.1 Development of the Program Content As shown in Table 1, the research team designed and developed the program according to the two proposed topics. The research team developed four courses based on the nostalgic topic. They are as follows: “Childhood games” (adopt children’s games to evoke the childhood memories of older adults to strengthen the sense of distance from the first course and increase the sense of identity). “About me, self-portrait” (self-portraits are used to integrate the past and present of older adults, linking their internal and external experiences and enabling them to look back at themselves). “Memories of food” (to inspire older adults with eating experiences to increase their memories and interaction). “Flowers of the four seasons” (to evoke the memory of the seasons for older adults). Then, based on the topic of association of color and music, and the color-sensor play device (FEEL2), four courses were developed to enrich the audiovisual synesthetic experience of participants [35], namely: “Listen to music and paint with colors” (feel the mood variations and picture sense brought by music), “Create a mosaic collage” (tear and paste colored paper to create mosaic collages), “Compose music with colors” (feel the process of transforming colors into music by the FEEL2 musical instrument), and “Make a music plate” (listen to the prerecorded music in the DIY kit, and create and assemble music records by themselves). Taking into account the age of the older adults and the deterioration of their learning abilities, the research team made corresponding adjustments to the teaching process and materials including simplifying the creation process, reducing the difficulty of creation, setting certain creative tasks, and planning the corresponding knowledge in each course. It allowed them to create independently while completing creative tasks to stimulate participation and learn some knowledge.

146

F.-W. Tung and L.-M. Jia Table 1. Topics and course content.

Topic Week

One

Course

Aim

1

Childhood games

Remind older adults of their childhood games with children’s toys

2

About me, self-portrait

Allow older people to look back and confront their present selves

3

Memories of food

Stimulate the good past memories of food

4

5

Two

6 7 8

Main materials Stamps (children’s toys), marbles, gyroscopes Photographs, transparencies, brushes, paints Clay, molds and shaping tools (self-developed)

Flowers of the four seasons Listen to music and paint with colors Create a mosaic collage Compose music with colors

Learn mosaic art by making mosaic collages on line drawings Experience the process of transforming colors into music

Masterpieces (reproduction), FEEL2 device

Make a music plate

Create their own music plates with the DIY kit

DIY kits, paint brushes, paints

Evoke the memory of the seasons

Brushes, pigments

Express the perception of music through creation

Paint brushes, paints, acrylic discs Line drawings, colored paper

3.2 Program Implementation and Basic Information of Participants The eight-week art program was held from December 2, 2021, to January 20, 2022 (every Thursday) at a community care station. Each course was conducted from 1:30 to 3:30 p.m. The effectiveness of the teaching was evaluated at the end of the program. The research team recruited older adults to participate in the program through publicity at the institution, and the participants were given verbal and written instructions and signed informed consent forms prior to the administration of the program. This study was approved by the Institutional Review Board of National Taiwan University (Number: 202010ES005). The research team recruited 21 participants aged 50–90 years, of whom 15 participated in the entire eight-week art program. As shown in Table 2, the highest percentage of participants was female, reaching 80% (n = 12). The age group was mainly between 71–80 years old, with 53.3% (n = 15). Regarding educational background, the percentage of those with high school or higher education reached 60% (n = 9). The highest percentage of people living with family members was 60% (n = 9).

The Effects of an Art Program on Older Adults’

147

Table 2. Basic information of participants. Characteristics

Range

Quantity (%)

Gender

Male

3 (20%)

Female

12 (80%)

61–70

7 (46.7%)

Age

71–80

8 (53.3%)

Education Background

Elementary School / Junior High School

6 (40%)

High School / Vocational High School

7 (46.7%)

University or above

2 (13.3%)

Living Style

Living alone

2 (13.3%)

Living with spouse (or cohabitant) only

4 (26.7%)

Living with family

9 (60%)

3.3 Cognition and Satisfaction with Life Assessment The research team applied the MoCA (Montreal Cognitive Assessment) scale proposed by Nasreddine et al. [36] to the participants in a pretest and posttest to understand the variation in their cognition. The MoCA is a brief screening tool that measures seven domains of cognition, including visuospatial/executive, naming, attention, language, abstraction, delayed recall, and orientation. It takes approximately 15 min to administer and has a total scale score of 30 (range: 0–30), a score of ≥ 26 is normal, 18–26 is mild cognitive impairment (MCI), 10–17 is moderate, and less than 10 is severe (with an additional 1 point for those with less than 12 years of education), reflecting the overall cognitive performance of the individual. The Satisfaction with Life Scale (SWLS) proposed by Pavot and Diener [37] was adopted to estimate the variation in satisfaction with life (pretest and posttest) of the participants. The scale consists of five questions and is scored on a 7-point Likert scale. A total score of 26 or more indicates that the subject is satisfied with life, 20–25 indicates that the subject is basically satisfied with life, and 20 or less indicates that the subject is dissatisfied with life.

4 Research Results 4.1 Results of Program Implementation The research team went into the community and implemented the 8-week art program, recording the participation of the older adults through observation, video, and photographs. Participants created their work individually and in groups. They completed 1–3 pieces of artwork in each course, and the creation process and main outcomes are shown in Fig. 1 and Fig. 2.

148

F.-W. Tung and L.-M. Jia

Four weeks of nostalgia-themed courses: Week 1 (Childhood games): The participants improvised by shaking marbles in a cardboard box; they also created artwork using a toy stamp set that were designed by the research team. Week 2 (About me, self-portrait): The participants drew their silhouettes from a photo using a transparency overlay and then color the silhouettes to created their self-portraits on their own. Week 3 (Memories of food): The participants created their most memorable food using the clay and modeling tools (designed by the research team with 3D printing) or shaped by themselves. Week 4 (Flowers of the four seasons): The participants learned basic traditional Chinese painting skills to draw flowers of the four seasons.

Fig. 1. Process and main results of topic one

Four weeks of courses on the topic of linking color and music. Week 5 (Listen to music and paint with colors): The participants were encouraged to use simple tools to express their feelings while listening to the music. Week 6 (Create a mosaic collage): The participants collaborated with team members to create mosaic collages. Week 7 (Compose music with colors): The participants worked in groups. The groups can transform the paintings (copies) into mosaics through the FEEL2 device turntable and colored cylinders; they could then enjoy each other’s mosaic works and feel the FEEL2 device transforming the works into melodies. Week 8 (Make a music plate): The participants listened to pre-recorded music from the DIY kit and then created and assembled a paper music board using the elements from the kit.

The Effects of an Art Program on Older Adults’

149

Fig. 2. Process and main results of topic two

4.2 Variation of Participants’ Cognition and Satisfaction with Life The research team conducted paired sample t tests on the pretest and posttest results (cognition and satisfaction with life) of these 15 participants. As Table 3 shows, the participants produced significant differences in cognition total score (p = 0.000), delayed memory (p = 0.005), and orientation (p = 0.028). Their total cognitive score increased from 25.07 (3.43) to 27.47 (2.67), delayed memory increased from 2.87 (1.77) to 3.93 (1.53), and orientation improved from 5.53 (0.64) to 5.93 (0.26). There were no significant differences in other dimensions. This indicates that this art program benefits older adults’ general cognition, delayed memory, and orientation. Table 3. Dimensions and total score of MoCA for the paired sample t test (n = 15) Dimensions

Pre-test M (SD)

Post-test M (SD)

t

df

Sig. (2-tailed)

Total Score

25.07 (3.43)

27.47 (2.67)

−5.27

14

.000***

Visuospatial/Executive

3.93 (0.96)

4.27 (0.80)

−1.32

14

0.207

Naming

2.87 (0.35)

2.73 (0.46)

1.47

14

0.164

Attention

5.73 (0.46)

5.87 (0.35)

−0.81

14

0.433

Language

1.93 (0.96)

2.27 (0.80)

−1.78

14

0.096

Abstraction

1.33 (0.72)

1.53 (0.74)

−1.38

14

0.189

Delayed recall

2.87 (1.77)

3.93 (1.53)

−3.38

14

0.005**

Orientation

5.53 (0.64)

5.93 (0.26)

−2.45

14

0.028*

* p < .05; ** p < .01; *** p < .001

As shown in Table 4, there was no significant difference in the satisfaction with life scores of the participants, but there was a certain degree of increase from 27.93 (5.44) to 29.67 (4.42).

150

F.-W. Tung and L.-M. Jia Table 4. Paired t test on satisfaction with life (n = 15)

Dimension

Pre-test M (SD)

Post-test M (SD)

t

df

Sig. (2-tailed)

Satisfaction with life

27.93 (5.44)

29.67 (4.42)

1.70

14

0.112

* p < .05

5 Discussion and Recommendations This study investigated the effects on the cognition and satisfaction with life of 15 older adults after participating in an 8-week art program (nostalgia-themed visual art, combining visual art and music linking color and music-themed). The participants showed significant improvements in general cognition, delayed recall, and orientation after participating in the art program. Although there was no significant improvement in satisfaction with life, there was still some positive impact on their daily lives. This echoes the study by scholars Çetinkaya et al. [4], who found that participation in art programs could result in a statistically significant increase in the cognition of older adults and did not have a significant effect on their life satisfaction but still had a positive influence on the quality of life. There are several possible reasons for this: (1) Nostalgic topics were used in the program’s first four weeks, allowing the older adults to recall their past lives and evoking memories of their childhood, their lives, and the seasons. (2) This study adopted a combination of visual art and music for the program development of the last four weeks. Older adults could experience the sympathetic sensation of color and music, giving them a rich sensory experience while allowing them to train their hands and brains. The above two points may be ways to train and improve memory for older adults. (3) This program covered a variety of creative materials (e.g., marbles, gyroscopes, stamps, clay, FEEL2 musical instrument, DIY kits, etc.) and creation formats (individual and group). It promoted communication, interaction, and mutual support among older adults in the program and also allowed them to experience the current era and technology, which helped trigger their interest in participating and increased their sense of anticipation for the program, which in turn improved their orientation of cognition. (4) According to the community workers, most of the older adults who participated in the program had prior experience with the community care station program and may already have achieved a relatively high level of satisfaction with life before attending this program. Therefore, there may have been little room for improvement in the Satisfaction with Life Scale scores. The results suggest that the program designed for this study can be applied to the daily learning of older adults to enrich their later life and positively impact their physical and mental health and quality of life. However, there are some limitations to this study. First, as a case study, the sample size of participants in this program was relatively small. The sample size could be increased in the future to supplement the credibility of the study results. Secondly, this study’s target population and field came from northern Taiwan, so the study results are limited to that region. In the future, this program could be implemented in other regions to expand the impact of this study. Finally, this study adopted the topics of nostalgia, linking color and music for the program development, and there are still other topics

The Effects of an Art Program on Older Adults’

151

applicable to older adults. In the future, it is possible to try different topics to explore the effects on older adults’ cognition and satisfaction with life. Acknowledgments. We would like to thank our research team members Chen-Ling Hung, WeiJu Chiu, Chia-En Chen, Wei-Zhe Tzeng, Ryo-Wei Liu, and Chia-Hsuan Ku for their remarkable contributions to this study. We are also grateful to the staff and older adults of the Suang Line Social Welfare Foundation for their tremendous support and participation. This work was supported by a grant from the University Social Responsibility (USR) Program, Taiwan Ministry of Education (Grant No.: 110O5005; 111O5005).

References 1. National Development Council. Compendium of Population Projection Charts-2022 Edition (2022). https://www.ndc.gov.tw/Content_List.aspx?n=81ECE65E0F82773F 2. Gergen, M., Gergen, K.: Positive aging: reconstructing the life course. In: Goodheart, C. (ed.) Handbook of Girls’ and Women’s Psychological Health: Gender and Well-Being Across the Life Span, pp. 416–426. Oxford University Press, Oxford, UK (2006) 3. Zhang, Y., et al.: Does music therapy enhance behavioral and cognitive function in elderly dementia patients? a systematic review and meta-analysis. Ageing Res. Rev. 35, 1–11 (2017). https://doi.org/10.1016/j.arr.2016.12.003 4. Çetinkaya, F., Asiret, G.D., Direk, F., Özkanli, N.N.: The effect of ceramic painting on the life satisfaction and cognitive status of older adults residing in a nursing home. Topics Geriatric Rehabil. 35(2), 108–112 (2019). https://doi.org/10.1097/tgr.0000000000000208 5. Im, M.L., Lee, J.I.: Effects of art and music therapy on depression and cognitive function of the elderly. Technol. Health Care 22(3), 453–458 (2014). https://doi.org/10.3233/THC140803 6. Hanna, G., Patterson, M., Rollins, J., Sherman, A.: The Arts and Human Development: Framing a National Research Agenda for The Arts, Lifelong Learning, and Individual Well-Being. National Endowment for the Arts, Washington, DC (2011) 7. Fink, S., et al.: W. In: Massenkeil, G. (ed.) Metzler Sachlexikon Musik, pp. 1148–1158. J.B. Metzler, Stuttgart (1998). https://doi.org/10.1007/978-3-476-05199-8_23 8. Schneider, J.: The arts as a medium for care and self-care in dementia: arguments and evidence. Int. J. Environ. Res. Public Health 15(6), 1151 (2018). https://doi.org/10.3390/ijerph 15061151 9. Herrmann, D.J., Yoder, C.Y., Gruneberg, M., Payne, D.G.: Applied Cognitive Psychology: A textbook. Lawrence Erlbaum Associates Publishers, Philadelphia, PA (2006) 10. Rebok, G.W., et al.: Ten-year effects of the advanced cognitive training for independent and vital elderly cognitive training trial on cognition and everyday functioning in older adults. J. Am. Geriatrics Soc. 62(1), 16–24. Doi. https://doi.org/10.1111/jgs.12607 11. Park, D.C., Lautenschlager, G., Hedden, T., Davidson, N.S., Smith, A.D., Smith, P.K.: Models of visuospatial and verbal memory across the adult life span. Psychol. Aging 17(2), 299–320 (2002). https://doi.org/10.1037/0882-7974.17.2.299 12. Yang, H.K.: The effects of the recreational music therapy on the cognitive functional of the aged people with dementia. J. Korean Gerontol. Soc. 26(4), 749–765 (2006) 13. Khodabakhsh, S.: Factors affecting life satisfaction of older adults in Asia: a systematic review. J. Happiness Stud. 23(3), 1–16 (2021). https://doi.org/10.1007/s10902-021-00433-x1 14. George, L.K.: Perceived quality of life. In: Binstock, R.H., George, L.K. (eds.) Handbook of Aging and the Social Sciences, pp. 320–336. Academic Press (2006). https://doi.org/10. 1016/B978-012088388-2/50021-3

152

F.-W. Tung and L.-M. Jia

15. Blomdahl, C., Gunnarsson, A.B., Guregård, S., Björklund, A.: A realist review of art therapy for clients with depression. Arts Psychother. 40(3), 322–330 (2013). https://doi.org/10.1016/ j.aip.2013.05.009 16. Popa, L.C., Manea, M.C., Velcea, D., S, alapa, I., Manea, M., Ciobanu, A.M.: Impact of Alzheimer’s dementia on caregivers and quality improvement through art and music therapy. Healthcare 9(6), 698 (2021). https://doi.org/10.3390/healthcare9060698 17. Brandmeyer, D.D.: Higher education activity and life satisfaction in the older adult. [Unpublished doctoral dissertation]. University of Georgia (1987). https://www.elibrary.ru/item.asp? id=7527031 18. Hickson, J., Housley, W.: Creativity in later life. Educ. Gerontol. Int. Quarterly 23(6), 539–547 (1997). https://doi.org/10.1080/0360127970230604 19. Adams-Price, C.E., Nadorff, D.K., Morse, L.W., Davis, K.T., Stearns, M.A.: The creative benefits scale: connecting generativity to life satisfaction. Int. J. Aging Hum. Dev. 86(3), 242–265 (2018). https://doi.org/10.1177/0091415017699939 20. Flood, M., Phillips, K.D.: Creativity in older adults: a plethora of possibilities. Issues Ment. Health Nurs. 28(4), 389–411 (2007). https://doi.org/10.1080/01612840701252956 21. Mahendran, R., et al.: Art therapy and music reminiscence activity in the prevention of cognitive decline: study protocol for a randomized controlled trial. Trials 18(1), 1 (2017). https:// doi.org/10.1186/s13063-017-2080-7 22. Mahendran, R., et al.: Art therapy is associated with sustained improvement in cognitive function in the elderly with mild neurocognitive disorder: findings from a pilot randomized controlled trial for art therapy and music reminiscence activity versus usual care. Trials 19(1), 1 (2018). https://doi.org/10.1186/s13063-018-2988-6 23. Sarne-Fleischmann, V., Tractinsky, N.: Development and evaluation of a personalised multimedia system for reminiscence therapy in Alzheimer’s patients. Int. J. Soc. Humanistic Comput. 1(1), 81-96 (2008). Doi: https://doi.org/10.1504/IJSHC.2008.020482 24. Lee, H.M., Wu, H.T., Chou, M.J.: The effects of incorporating nostalgia activities into the classroom management course of in-service preschool teachers education program. Europ. J. Res. Reflection Manage. Sci. 4(3), 43–60 (2016) 25. Gonzalez, J., Mayordomo, T., Torres, M., Sales, A., Melendez, J.C.: Reminiscence and dementia: a therapeutic intervention. Int. Psychogeriatric Assoc. 27(10), 1731–1737 (2015). https:// doi.org/10.1017/S1041610215000344 26. Cohen, G.D., Perlstein, S., Chapline, J., Kelly, J., Firth, K.M., Simmens, S.: The impact of professionally conducted cultural programs on the physical health, mental health, and social functioning of older adults. Gerontologist 46(6), 726–734 (2006). https://doi.org/10.1093/ger ont/46.6.726 27. Cohen, G.D., Perlstein, S., Chapline, J., Kelly, J., Firth, K.M., Simmens, S.: The impact of professionally conducted cultural programs on the physical health, mental health, and social functioning of older adults—2-year results. J. Aging Humanities Arts 1(1–2), 5–22 (2007). https://doi.org/10.1080/19325610701410791 28. Habibi, A., Damasio, A.: Music, feelings, and the human brain. Psychomusicology: Music Mind Brain 24(1), 92–102 (2014). https://doi.org/10.1037/pmu0000033 29. Bugos, J.A., Perlstein, W.M., McCrae, C.S., Brophy, T.S., Bedenbaugh, P.H.: Individualized piano instruction enhances executive functioning and working memory in older adults. Aging Ment. Health 11(4), 464–471 (2007). https://doi.org/10.1080/13607860601086504 30. Fraser, K.D., O’Rourke, H.M., Wiens, H., Lai, J., Howell, C., Brett-MacLean, P.: A scoping review of research on the arts, aging, and quality of life. Gerontologist 55(4), 719–729 (2015). Doi: https://doi.org/10.1093/geront/gnv027 31. Sandak, B., Gilboa, A., Harel, D.: Computational paradigm to elucidate the effects of artsbased approaches: art and music studies and implications for research and therapy. Front. Psychol. 11(1200), 1–20 (2020). https://doi.org/10.3389/fpsyg.2020.01200

The Effects of an Art Program on Older Adults’

153

32. Curwen, C.: Music-colour synaesthesia: concept, context and qualia. Conscious. Cogn. 61, 94–106 (2018). https://doi.org/10.1016/j.concog.2018.04.005 33. Jia, L.M., Tung, F.W.: FEEL2: An Interactive Device for Older Adults to Experience Synesthesia and Age Creatively. Paper presented at the International Conference on Human-Computer Interaction 2022: Human Aspects of IT for the Aged Population. Technology in Everyday Living, Gothenburg, Sweden, June 2022. https://doi.org/10.1007/978-3-031-05654-3_17 34. Reilly, R.C.: Participatory case study. In: Albert, M., Durepos, G., Wiebe, E. (eds.) Encyclopedia of Case Study Research, pp. 658–660. Sage Publishing, Thousand Oaks (2010). https:// spectrum.library.concordia.ca/id/eprint/6420/ 35. Tung, F.W., Jia, L.M.: Developing a community-based art program for older adults that connects color and music. Educ. Gerontol., 1–15. Doi:https://doi.org/10.1080/03601277.2022. 2160573 36. Nasreddine, Z.S., et al.: The montreal cognitive assessment, MoCA: a brief screening tool for mild cognitive impairment. J. Am. Geriatr. Soc. 53(4), 695–699 (2005). https://doi.org/ 10.1111/j.1532-5415.2005.53221.x 37. Pavot, W., Diener, E.: Review of the satisfaction with life scale. Psychol. Assess. 5(2), 164–172 (1993). https://doi.org/10.1037/10403590.5.2.164

College Students’ Perceptions and Preferences Regarding Intelligent Advisory Systems in Multi-device Learning Environments Zhao Xu and Qingchuan Li(B) School of Humanity and Social Science, Harbin Institute of Technology, Shenzhen, China [email protected], [email protected] Abstract. With the emergence of e-learning systems, learning activities are no longer limited to a single fixed user interface. Nevertheless, college students’ multi-device learning behaviors are poorly understood. In addition, considering the possible crises brought about by the multi-device learning environment, there is also a necessity to investigate the effectiveness of intelligent advisory systems in the multi-device learning process. To address these problems, an online survey was designed and developed to understand Chinese college students’ multi-device learning behaviors and possible crises, as well as their perceptions and preferences of IAS virtual advisors. It also sought to discover the design opportunities of the IAS virtual advisor by uncovering the relationships between students’ personality traits, learning characteristics, learning patterns, and user preferences. The results yielded the college students’ usage habits with multiple devices during online learning, including the number and types of devices used, the frequency of use, the ways of interacting and combining different devices, as well as the possible learning crises and typical situations encountered. The study further explored the significant effects of students’ personality traits and learning characteristics on their learning behaviors, as well as perceptions and preferences of IAS virtual advisors, providing insights into how to design customized intelligent advisory systems based on students’ personal traits. Keywords: Multi-Device Learning · User Perceptions and Preferences · Intelligent Advisory System · Virtual Advisor · Personality Traits · Learning Characteristics

1 Introduction With an increasing number of technologies emerging in e-learning systems, learning activities are no longer limited to a single, fixed user interface such as an electronic whiteboard or computer. Students can utilize multiple devices, i.e., interactive whiteboards, desktops, smartphones, tablets, and wearable devices, to accomplish various learning activities such as online courses, practice and testing, reading literature, searching for information, and taking notes. Nevertheless, the effectiveness of multi-device The original version of this chapter was revised: the corresponding author has been corrected. The correction to this chapter is available at https://doi.org/10.1007/978-3-031-34866-2_47 © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023, corrected publication 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 154–166, 2023. https://doi.org/10.1007/978-3-031-34866-2_12

College Students’ Perceptions and Preferences

155

learning is highly questionable. First, the increasing amount of multimodal and distributed information could easily cause distraction, extensive cognitive load, and even learning errors or crises [1]. Second, the interaction between students and teachers is reduced to a great extent in e-learning environments; thus, student performance greatly varies depending on their abilities, such as self-control and self-management [2, 3]. To overcome the possible obstacles, intelligent advisory systems (IASs) have been developed by some researchers to guide and tutor students throughout e-learning activities, evaluate students’ learning status, and predict potential learning crises. For example, the Course Signals system developed by Purdue University can determine students who are in a learning crisis based on their status data and then remind the instructor to interfere in the students’ learning behavior via e-mail or instant message [4]. In addition, information visualization for e-learning activities can be also employed to display students’ status data. For example, the advisory system of Khan Academy’s learning dashboard sends appropriate warnings once a learning crisis has occurred [5]. More recently, the intelligent virtual advisor has become a more promising tool in IASs. It has been shown to be effective in providing immediate feedback regarding learning performance, guiding the process of solving problems step by step, and helping students reduce study stress [6, 7]. However, students’ learning habits and usage behaviors are poorly understood in multi-device learning environments. The effectiveness of such IASs in multi-device learning environments has been limitedly investigated. In addition, students with different personality traits usually adopt different e-learning patterns. Students’ perceptions and preferences of IASs may vary from person to person, which remains to be further studied. Thus, the objective of the current study was first to investigate students’ learning behaviors and habits when they are required to adopt multiple devices and utilize distributed interfaces for knowledge acquisition and generation. Second, the study aimed to evaluate students’ perceptions and preferences with various IAS virtual advisors, gaining insight into their needs and expectations when meeting with learning difficulties and crises. Third, this study has focused on the relationships between students’ personality traits, learning characteristics, multi-device learning behaviors, and perceptions of and preferences for IAS designs.

2 Literature Review Recently, with numerous variable learning devices and network platforms relatively suddenly appearing on the market, the implication of multiple devices has become a common phenomenon in the design of e-learning systems (ELSs) [8–10], particularly in the context of the ongoing COVID-19 pandemic. In particular, multi-device interaction should be concerned with the sequence of device usage, i.e., cross-device experiences and multi-device experiences, respectively referring to the use of multiple devices sequentially and simultaneously [11]. Moreover, the way devices are combined can also influence learners’ level of attention toward various content as well as task performance [12, 13]. Hence, to achieve a better educational outcome, it is vital to investigate how students make use of various devices by considering the devices’ portability, screen size, storage, capabilities, and limitations in multi-device learning environments.

156

Z. Xu and Q. Li

In addition, to ensure the effectiveness of an ELS, the personalization of learning mechanisms and strategies is particularly important, as this makes it possible for educators to automatically generate and customize course content across various devices [14]. For example, an IAS represents a promising solution for a personalized ELS following content-based, collaborative filtering or hybrid principles, thus enabling increased interaction between educators and students [15]. Specifically, the social presence of an intelligent virtual advisor plays a vital role in students’ acceptance and perceptions of the IAS [6, 16]. Students have been found to show more trust in and motivation toward an intelligent virtual advisor when learning and collaborating in the online environment [17]. However, there is little research focusing on the effectiveness of intelligent virtual advisors within multi-device learning environments. Considering that personalization is the main focus of IAS [18–20], it is necessary to investigate students’ perceptions of and preferences regarding IAS design in the multi-device learning process. Personality traits and learning characteristics are the two main factors to be considered in understanding students’ perceptions and preferences of IAS design because they can greatly influence students’ learning behaviors. First, it has been demonstrated that a student’s high level of agreeableness, conscientiousness, and openness to a new experience relates to better performance in online learning [21]. Moreover, several learning characteristics including academic self-efficacy, active procrastination, and self-handicapping have been suggested as influencing students’ learning performance and emotional states. For example, self-efficacy with technology and previous selfregulated learning experience are suggested to be highly associated with high grades in online courses [22]. Furthermore, IAS notifications with various contexts and from various devices elicit different emotional responses among students with different personality traits and learning characteristics [23]. In this vein, students who trend toward active procrastination may have more opportunities to reject frequent recommendations from an IAS. Overall, there remains a need to understand the underlying relationships between a student’s personality traits, learning characteristics, as well as perceptions and preferences in the context of IAS virtual advisor design. In summary, this research probes the following three questions. RQ1: How do the students use and manage their devices when they conduct online learning activities in a multi-device learning environment? RQ2: What are students’ perceptions and preferences for the IAS virtual advisor, especially when they encounter difficulties and crises? RQ3: What are the relationships between students’ personality traits, learning characteristics, multi-device learning behaviors, and their perceptions of and preferences regarding the IAS virtual advisor design?

3 Method 3.1 Instrument Development A survey was developed to investigate the above research questions. The first part of the questionnaire was designed to understand the students’ multi-device learning behaviors. Students were asked about the numbers, types, and frequency of various devices they usually used. The typical situations in which they needed to conduct multi-device learning and how they usually combined and interacted across various devices were

College Students’ Perceptions and Preferences

157

also investigated. The second part of the questionnaire mainly concentrates on students’ experiences of multi-device learning crises in terms of when and how they usually met with difficulties during the whole semester. The third part of the questionnaire investigates the details of students’ perceptions and preferences for IAS virtual advisor design. First, the students were asked about their general perceptions and preferences regarding the IAS virtual advisor by answering the following questions: “Who do you prefer to get help and assistance from in real-life situations?”; “Where do you prefer to receive reminders or messages from?”; “To what extent do you agree that the encouragement and assistance from the IAS virtual advisor could help you to overcome the possible learning difficulties?”; and “What kind of help and assistance from the IAS virtual advisor do you prefer when meeting with learning difficulties?” Then, we examined the students’ preferences and expectations for the design of the IAS virtual advisor in terms of their preferences about the virtual advisor’s roles, characteristics, appearances, and degree of customization. Ultimately, the questionnaire collected the participants’ demographic information, including their age, gender, education experience, and expertise. Moreover, we investigated the participants’ personality traits using the simplified scale of the Big Five factors of personality to evaluate students’ individuality [24]. As an explorative study into the underlying relationships between students’ personality traits and multi-device learning, we selected two items from each dimension of the simplified scale of the Big Five factors of personality. In total, 10 questions were developed and re-written in Chinese by two experts from the psychology and ergonomics fields. In addition, the measurement of students’ learning characteristics regarding academic self-efficacy, active procrastination, and self-handicapping were adopted from the Academic Self-Efficacy Scale (ASES) [25], the new active procrastination scale (NAPS) [26], and the Academic SelfHandicapping Scale (ASHS) [27]. Table 1 lists the measurement constructs and items for personality traits and learning characteristics used in this survey. The participants were asked to evaluate their personality traits and learning characteristics with a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). To ensure ideal data quality, an attention filter was designed to avoid invalid participation. Table 1. Personality traits scale used in the current study.

Personality traits

Construct

Item

Question

References

Extraversion

ET1

I’m energetic

[24]

ET2

I’m quiet.*

OE1

I’m intellectual

Openness to experience

OE2

Emotional stability ES1

I’m unimaginative.* I’m emotional.* (continued)

158

Z. Xu and Q. Li Table 1. (continued) Construct

Item ES2

I’m unenvious

Conscientiousness

CS1

I’m disorganized.*

CS2

I’m neat

AG1

I’m sympathetic

AG2

I’m unkind.*

SE1

I ask myself questions to [25] make sure I understand the material I have been studying

SE2

Even when I do poorly on a test, I try to learn from my mistakes

SE3

I find that when the teacher is talking, I think of other things and don’t really listen to what is being said.*

SE4

I often choose those topics that I can learn something from, even if they require more work

SE5

I always try to understand what the teacher said, even if it doesn’t make any sense

SE6

My learning skills are excellent compared with other students in this class

AP1

If I put things off until the last moment, I’m satisfied with the outcomes

AP2

It’s really a pain for me to work under upcoming deadlines.*

Agreeableness Learning characteristics

Academic self-efficacy

Active procrastination

Question

References

[26]

(continued)

College Students’ Perceptions and Preferences

159

Table 1. (continued) Construct

Academic self-handicapping

Item

Question

References

AP3

In order to make better use of my time, I intentionally put off some tasks

SH1

I would say I was nervous before the test, but I wasn’t as nervous as I said I was

SH2

I will set myself some unrealistic goals so that even if I don’t accomplish them, they won’t hit me particularly hard

SH3

When I was at school, I played more because I wanted to act like I was learning easily

[27]

Note: * indicates reverse items Table 2. Demographic characteristics. Frequency

Percentage (%)

76

42.7

Gender

Male Female

102

57.3

Age

17–20 years

102

57.3

21–23 years

70

39.3

Education stage Majors enrolled

24–27 years

6

Undergraduate

170

3.4

Postgraduate

8

4.5

Human science

28

15.7

Social science

44

24.7

Natural science

43

24.2

Engineering science

63

35.4

95.5

3.2 Data Collection and Analysis The survey was distributed through an online Chinese survey platform, Wenjuanxing (https://www.wjx.cn), in November 2022. Students who were older than 18 years and had

160

Z. Xu and Q. Li

online learning experience were recruited. Specifically, the participants were accustomed to conducting online learning activities, such as taking courses, completing exercises and assignments, taking examinations, searching for and reading literature, or taking notes. The whole survey took about 20 min to complete, and each participant received a reward of 7 RMB from the survey platform. We received a total of 250 responses, and 178 valid surveys were collected for further data analysis. Seventy-two invalid surveys were deleted because they failed the attention filter questions. The descriptive analysis was first performed to analyze the participants’ demographic information, multi-device learning behaviors, learning crisis experience, as well as perceptions and preferences regarding the IAS virtual advisor design. Furthermore, the Spearman’s rank-order correlation and Kruskal–Wallis H test were then conducted to determine if there were statistically significant relationships between participants’ personality traits, learning characteristics, multi-device behavior patterns, and perceptions and preference patterns of the IAS virtual advisor design.

4 Results 4.1 Participants One hundred seventy-eight students participated in this survey, including 76 males (42.7%) and 102 females (64.8%). They ranged in age from 17 to 27 years, and most were at an undergraduate education level. The participants studied in various majors including human science, social science, natural science, and engineering science. Table 2 provides their demographic characteristics in detail. 4.2 Multi-device Learning Behavior and Possible Crises Most participants (78.1%) utilized two to three devices in conjunction with one another for online learning, and only 21.9% possessed only one device for use during the online learning process. The most frequently used devices in multi-device learning included smartphones (96.6%), laptops (85.4%), tablets (66.3%), and earphones (64.5%). Of the participants, 96.6% had used smartphones in online learning, which usually served as a tool for social interaction (82.6%), word and voice recording (64.0%), and information searching (57.9%). Meanwhile, 85.4% of participants had adopted laptops in multidevice learning for content presentation (87.6%), typing or drawing tasks (75.8%), and information searching (70.8%). Furthermore, 66.3% of participants used tablets for multi-device learning, which usually served in content presentation (66.3%), coworking with electronic pens (48.9%), and information searching (47.8%). Earphones were found to accompany the process of online learning, with a 64.5% utilization rate; 53.9% of the participants used earphones during most of the online learning process, followed by participants who used earphones from time to time (25.8%) or during the entire online learning process (14.0%). Several situations were mentioned by the participants in terms of when they needed to learn across multiple devices, including watching online courses or viewing learning materials on screen (72.5%), conducting information searching (67.4%), transferring files (52.3%), and typing or creating (51.1%). As for the

College Students’ Perceptions and Preferences

161

ways of combining and interacting with multiple devices, 47.2% of participants were accustomed to keeping all the device screens active and operating each device successively. On the other hand, 32.0% of participants usually operated one main device and only activated the other devices as needed. Overall, the students’ current combining and interacting methods were quite time-consuming. Concerning the possible learning crises, we discovered several typical situations that could easily evoke learning difficulties and negative emotions. Specifically, 44.4% of participants reported that they have encountered crises in knowledge acquisition and generation such as when listening to courses, taking notes, and reading literature; 28.1% of participants encountered difficulties in collaborative and interactive activities such as group discussion, brain-storming, and collaboration. Furthermore, with the learning materials and knowledge becoming increasingly more in-depth, participants tended to have more negative feelings moving from the beginning of the semester (14.0%) to the middle (25.3%) and end (48.9%) stages of the semester. 4.3 User Perceptions and Preferences of IAS Virtual Advisors Four questions were asked about the participants’ perceptions and preferences regarding the IAS virtual advisor design. First, for those who preferred to receive help and assistance in person, students’ favorite help providers were reported to be their classmates (79.2%), teachers (69.7%), senior schoolmates (46.1%), online professionals (29.2%), and other instructors (16.9%). In addition, most of these students preferred to receive reminders and messages from online learning websites and applications (61.2%), as well as social media groups (34.3%). Regarding the IAS design, a majority of respondents (70.2%) believed that the virtual advisor could be useful in encouraging and helping them to overcome possible learning difficulties. More specifically, in terms of the help and assistance offered by the IAS, 42.7% of participants preferred that the IAS provide a general direction for problem-solving, while 35.4% preferred that the IAS offered some knowledge points and tips for problem-solving; moreover, 18.0% wanted specific guiding steps for problem-solving, and 3.9% wanted emotional support and encouragement. Participants’ preferences and expectations for the IAS virtual advisor were specifically investigated in this study. The participants expected the virtual advisor to play roles such as a facilitator (55.6%), companion (30.9%), inspiration (7.9%), and supervisor (5.6%). The most preferred personality of the virtual advisor included being sincere and reliable (43.3%), as well as sweet and kind (41.0%). Some participants (10.1%) expected the virtual advisor to be humorous and cheerful, and only a few of them preferred the virtual advisor to be cool-headed and strict (5.6%). In terms of the appearance of the virtual advisor, participants favored figures such as an anthropomorphic software agent (48.9%), cartoon character (29.2%), animal character (14.0%), and intelligent robots (7.9%). Moreover, participants preferred a moderate degree of customization with the figure of virtual advisors, such as customizing their preferred virtual agent by configuring them from serval recommended image sets (44.9%) or adjusting only a part of the virtual agent’s image (34.8%).

162

Z. Xu and Q. Li

4.4 Relationships Between Personality Traits, Learning Characteristics, Learning Behaviors, and User Perceptions The Spearman’s rank-order correlation was first conducted to determine the relationships between the participants’ personality traits, learning characteristics, multi-device behaviors, as well as perceptions and preferences of the IAS virtual advisor. The results indicated there were significant correlations between the participants’ general perceptions and their personality traits of ET1 (p = .002), OE2 (p = .023), CS1(p = .021), and CS2 (p = .007). To be specific, students who were more energetic, less open, and less conscientious were more likely to agree that the encouragement and help from the IAS virtual advisor could help them to overcome possible difficulties. Then, the Kruskal–Wallis H test was further employed to determine the significant differences in participants’ personality traits and learning characteristics between different multi-device learning behaviors and perceptions and preference patterns. First, the results showed that there was a statistically significant difference in participants’ personality traits of ET2 between different multi-device learning patterns—χ2 (1) = 6.661, p = 0.010—with a mean rank ET2 score of 107.23 for the participants who usually used only one device for online learning and 84.53 for the participants who usually adopted two to three devices for online learning. That is, the students who were more extroverted tended to adopt multiple devices for online learning. Second, a statistically significant difference was reported in participants’ learning characteristics of SE5—χ2 (3) = 12.761, p = .005—between different preference patterns for the kinds of IAS help and assistance, with a mean rank SE5 score of 111.57 for participants who wanted emotional support and encouragement, 100.95 for participants who preferred specific guiding steps for problem-solving, 99.37 for participants who preferred some knowledge points and tips, and 74.46 for participants who would like to receive general directions for problem-solving. Meanwhile, a statistically significant difference was also indicated in participants’ learning characteristics of AP3—χ2 (3) = 111.366, p = .010—between different preference patterns for the kinds of IAS help and assistance, with a mean rank AP3 score of 104.06 for participants who preferred some knowledge points and tips, 93.86 for participants who wanted emotional support and encouragement, 92.55 for participants who preferred specific guiding steps for problem-solving, and 75.75 for participants who would like to receive general directions for problem-solving. Moreover, the Kruskal–Wallis H test also revealed a significant difference in participants’ learning characteristics of SH1—χ2 (3) = 9.887, p = .020—between different preference patterns for the virtual advisor’s personalities, with a mean rank SH1 score of 108.78 for participants who preferred the virtual agent to be humorous and cheerful, 106.00 for participants who preferred the virtual agent to be cool-headed and strict, 95.82 for participants who preferred the virtual agent to be sweet and kind, and 76.86 for participants who preferred the virtual agent to be sincere and reliable. In addition, a significant difference was reported in participants’ personality traits of CS2—χ2 (3) = 10.959, p = .012—between different preference patterns for the virtual advisor’s role, with a mean rank CS2 score of 109.89 for the participants who preferred the virtual agent to be an inspiration, 96.76 for the participants who preferred the virtual agent to be a

College Students’ Perceptions and Preferences

163

companion, 86.30 for the participants who preferred the virtual agent to be a facilitator, and 52.70 for the participants who preferred the virtual agent to be a supervisor.

5 Discussion and Conclusion This research study aimed to investigate college students’ multi-device learning behaviors, as well as their perceptions and preferences of IAS virtual advisor design. An online survey was conducted to address the research questions. The results indicated that there were a majority of students who adopted multiple devices during the online learning process. The underlying relationships between the student’s personality traits, learning characteristics, multi-device learning behaviors, and perceptions and preferences of IAS virtual advisor design were also revealed. Overall, students tended to use and interact with a complex configuration of multiple devices in multi-device learning, including smartphones, laptops, earphones, and tablets [12]. RQ1 was thus addressed. Notably, the participants still adopted traditional methods to combine and interact with these devices, such as operating various devices successively or separately, which was quite time-consuming. Future studies are needed to design an effective transition method to support task-switching in the multi-device learning environment [28]. Additionally, a multi-device-based IAS virtual advisor could become a promising solution because it enables flexible multi-modal interactions for each of the devices [11]. We also discovered several typical multi-device learning crisis scenarios, such as knowledge acquisition and generation, as well as collaborative and interactive activities. Since the utilization of multi-devices may result in extra stress and pressure on students [29], it is necessary to pay attention to the possible crisis scenarios during this process. Regarding RQ2, we examined the students’ perceptions and preferences regarding the IAS virtual advisor. The results showed that it was practical to provide some reminders and suggestions for students via the IAS virtual advisor, such as general directions for problem-solving, knowledge points and tips, specific guiding steps for problem-solving, and emotional support and encouragement, depending on the students’ personality traits and learning characteristics. Generally, students preferred to get help from their classmates and teachers; thus, it was not surprising that they were more comfortable when the virtual advisors served as a facilitator or companion. Previous research has indicated that the humanness characteristics of the virtual agent are more effective in improving learning performance [6]. In addition to the anthropomorphic agent design, we suggest that a famous cartoon character may also be effective in the development of IAS virtual advisors. Based on the results of the current study, a virtual advisor that showcases a personality that is sincere, reliable, sweet, and kind may improve IAS acceptance and adoption among college students. Moreover, the students favored a moderate degree of customization with their IAS virtual advisor. This study also revealed the underlying relationships between college students’ personality traits, learning characteristics, multi-device learning behaviors, and perceptions and preferences with IAS virtual advisors. In summary, students who were more energetic, less open, and less conscientious tended to perceive virtual advisors as more helpful and inspiring than others, and extroverted students were more likely to adopt multiple

164

Z. Xu and Q. Li

devices in online learning. In addition, the students with a higher level of self-efficacy preferred emotional support and encouragement from the virtual agent rather than other kinds of help, and the students with a higher level of active procrastination preferred the virtual agent provided some knowledge points and tips. Furthermore, the students with a higher level of self-handicapping were more likely to choose a virtual agent with a personality with humor and cheerfulness; meanwhile, the students who were more conscientious preferred the virtual agent to be an inspiration. Therefore, design and development should also be considered in terms of the students’ personality traits and learning characteristics.

6 Conclusion The results of the current study revealed Chinese college students’ learning habits and possible crises encountered when accomplishing tasks with multiple devices and obtaining information across distributed user interfaces. The findings also revealed relationships between the college students’ personality traits, learning characteristics, multidevice learning behaviors, as well as the perceptions of and preferences for IAS virtual advisors, which can provide insights into how to design customized IAS according to the students’ personality traits and learning characteristics. However, this study should also be considered in terms of its limitations. Specifically, this survey study was an explorative step to investigate the relationships between students’ characteristics and multi-device learning behaviors, employing simplified versions of personality traits and learning characteristics scales. Future research could conduct more comprehensive studies to explore the influence of students’ personality traits and learning characteristics on their perceptions or preferences of IAS virtual advisors in depth. Acknowledgement. This research was supported by National Natural Science Foundation of China (Grant No. 62207008) and Shenzhen Educational Science Planning. Project (Program No. Zdfz20015).

References 1. Ayres, P.L.: Systematic mathematical errors and cognitive load. Contemp. Educ. Psychol. 26(2), 227–248 (2001) 2. Li, Q., Zhang, J., Xie, X., Luximon, Y.: How shared online whiteboard supports online collaborative design activities: a social interaction perspective. In: Markopoulos, E., Goonetilleke, R.S., Ho, A.G., Luximon, Y. (eds.) AHFE 2021. LNNS, vol. 276, pp. 285–293. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-80094-9_34 3. Kent, C., Laslo, E., Rafaeli, S.: Interactivity in online discussions and learning outcomes. Comput. Educ. 97, 116–128 (2016) 4. Arnold, K.E., Pistilli, M.D.: Course signals at Purdue: Using learning analytics to increase student success. In: Proceedings of the 2nd International Conference on Learning Analytics and Knowledge, pp. 267–270. ACM, New York (2012) 5. Khan, A.: Introducing…the Learning Dashboard, https://blog.khanacademy.org/introduci ngthe-learning-dashboard/. Accessed 2022/12/15

College Students’ Perceptions and Preferences

165

6. Huang, H., Chen, Y., Rau, P.L.P.: Exploring acceptance of intelligent tutoring system with pedagogical agent among high school students. Universal Access Inf. Soc. 21(2), 381–392 (2022) 7. Ranjbartabar, H., Richards, D., Bilgin, A., Kutay, C.: First impressions count! the role of the human’s emotional state on rapport established with an empathic versus neutral virtual therapist. IEEE Trans. Affective Comput., 1 (2019) 8. Jokela, T., Ojala, J., Olsson, T.: A diary study on combining multiple information devices in everyday activities and tasks. In: Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems, pp. 3903–3912. ACM, New York (April 2015) 9. Levin, M.: Designing Multi-Device Experiences: An Ecosystem Approach to User Experiences Across Devices. O’Reilly Media, Sebastopol (2014) 10. Sher, V., Hatala, M., Gaševi´c, D.: On multi-device use: Using technological modality profiles to explain differences in students’ learning. In: The 9th International Conference on Learning Analytics & Knowledge, pp. 1–10. ACM, New York (2019) 11. White, R.W., Fourney, A., Herring, A., Bennett, P.N.,… Encarnación, M.J.: Multi-device digital assistance. Commun. ACM 62(10), 28–31 (2019) 12. Heitmayer, M.: Patterns of multi-device use with the smartphone. A video-ethnographic study of young adults’ multi-device use with smartphones in naturally occurring contexts. Comput. Hum. Behav. Rep. 8, 100244 (2022) 13. Oulasvirta, A., Sumari, L.: Mobile kits and laptop trays: managing multiple devices in mobile information work. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 1127–1136. ACM, New York (2007) 14. Daniels, D., Lee, J.S.: The impact of avatar teachers on student learning and engagement in a virtual learning environment for online STEM courses. In: Learning and Collaboration Technologies. Novel Technological Environments: 9th International Conference, LCT 2022, Held as Part of the 24th HCI International Conference, HCII 2022, Virtual Event, June 26–July 1, 2022, Proceedings, Part II. Springer, Cham, Switzerland (2022) 15. Silveira, R.A., Vicari, R.M.: Improving interactivity in e-learning systems with multi-agent architecture. In: Adaptive Hypermedia and Adaptive Web-Based Systems: Second International Conference, pp. 466–471. Springer, Heidelberg (2002). Doi: https://doi.org/10.1007/ 3-540-47952-X_62 16. Chae, S.W., Lee, K.C., Seo, Y.W.: Exploring the effect of avatar trust on learners’ perceived participation intentions in an e-learning environment. Int. J. Hum.-Comput. Interact. 32(5), 373–393 (2016) 17. Wilton, L., Noël, T.: Avatars: usefulness in collaborative online learning environments. Collected Essays Learn. Teach. 4, 115–120 (2011) 18. Sancho, P., Martínez, I., Fernández-Manjón, B.: semantic web technologies applied to elearning personalization in. J. Univ. Comput. Sci. 11(9), 1470–1481 (2005) 19. Viet, A.N., Si, D.H.: Acgs: Adaptive course generation system-an efficient approach to build e-learning course. In: The Sixth IEEE International Conference on Computer and Information Technology (CIT 2006), p. 259. IEEE (2006) 20. Rosaci, D., Sarne, G.M.L.: Efficient personalization of e-learning activities using a multidevice decentralized recommender system. Comput. Intell. 26(2), 121–141 (2010) 21. Yu, Z.: The effects of gender, educational level, and personality on online learning outcomes during the COVID-19 pandemic. Int. J. Educ. Technol. High. Educ. 18(1), 1–17 (2021). https://doi.org/10.1186/s41239-021-00252-3 22. Wang, C.H., Shannon, D.M., Ross, M.E.: Students’ characteristics, self-regulated learning, technology self-efficacy, and course outcomes in online learning. Distance Educ. 34(3), 302– 323 (2013)

166

Z. Xu and Q. Li

23. Weber, D., Voit, A., Kratzer, P., Henze, N. In-situ investigation of notifications in multi-device environments. In: Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing, pp. 1259–1264 (2016) 24. Donnellan, M.B., Oswald, F.L., Baird, B.M., Lucas, R.E.: The mini-ipip scales: tiny-yeteffective measures of the big five factors of personality. Psychol. Assess. 18(2), 192 (2006) 25. Pintrich, P.R., Groot, E.: Motivational and self-regulated learning components of classroom academic performance. J. Educ. Psychol. 82(1), 33–40 (1990) 26. Choi, J.N., Moran, S.V.: Why not procrastinate? development and validation of a new active procrastination scale. J. Soc. Psychol. 149(2), 195–212 (2009) 27. Urdan, T., Midgley, C.: Academic self-handicapping: what we know, what more there is to learn. Educ. Psychol. Rev. 13, 115–138 (2001) 28. Nguyen, N.T., Lee, H.: Device transition: understanding usability issues in shifting a device during a task. In: Design, User Experience, and Usability. User Experience in Advanced Technological Environments: 8th International Conference, DUXU 2019, Held as Part of the 21st HCI International Conference, HCII 2019, Orlando, FL, USA, July 26–31, 2019, Proceedings, Part II 21, pp. 178–191. Springer Cham (2019) 29. Barber, L.K., Santuzzi, A.M.: Please respond ASAP: workplace telepressure and employee recovery. J. Occup. Health Psychol. 20(2), 172 (2015)

Experimental Design and Design Methodology of Smart Screen Ageing Based on Emotion Regulation Yan-Min Xue1 , Meng-Ting Xu1(B) , Bao-Xin Xu1 , Chang Ge1 , and Sui-Huai Yu2 1 College of Art and Design, Xi’an University of Technology, Xi’an 710000, China

[email protected] 2 Institute of Industrial Design, Northwestern Polytechnical University, Xi’an 710000, China

Abstract. In order to improve the user experience of the elderly and alleviate their negative emotions caused by the lack of experience, the mapping influence between the emotional state of the elderly and the design elements of the interactive interface is obtained through human-machine experiments, and an age-appropriate design method for the interface is proposed with the introduction of emotion regulation. Firstly, we use semi-structured interviews to obtain the problems of elderly users in using smart products, focus on the influence of users’ emotions on the use of operating interfaces, and complete the analysis and extraction of interface design elements; secondly, we use the China Emotion Picture System (CAPS) to establish a negative emotion-evoking sample database, and use wearable devices and eye-movement devices to monitor the visual characteristics of elderly people’s interaction interface design elements under different emotions. The mapping relationship between the emotional state of the elderly and the design elements of the interactive interface was obtained. Based on the analysis of the experimental results, we identified the direction of product optimisation and proposed a design method based on emotion regulation. Conclusion The research on the ageingfriendly design of smart screens aims to alleviate the negative emotional problems of the elderly, provide a more convenient and comfortable life for elderly users, and provide a reference for other ageing-friendly product designs. Keywords: emotion regulation · interactive interface · ageing · smart screens · user needs

1 Introduction In the “14th Five-Year Plan” for the development of the digital economy, the State Council proposed to guide the interconnection of smart home products, promote the intelligent interaction between home products and the home environment, and create a new digital life goal of wisdom and sharing [1]. Nowadays, smart screens as a comprehensive interaction platform bring users a more comfortable and convenient home life, becoming the main terminal form of the future family wisdom centre and the most feasible form of wisdom elderly terminal. However, the current design of suchproducts is mostly based © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 167–180, 2023. https://doi.org/10.1007/978-3-031-34866-2_13

168

Y.-M. Xue et al.

on the habits and preferences of young people, and focuses on the use of functions [2], while ignoring the psychological needs and changes in the emotional state of older people, resulting in information cognitive impairment and poor emotional experience for older users, who are at a loss for products and also generate a lot of negative emotions. Studies have shown that the emotional state of the elderly is an important measure of their physical health, and if negative emotions such as anxiety and loss cannot be regulated and channeled in time, they can easily lead to various psychological and physical diseases [3]. Therefore, using emotion regulation as a perspective and smart screen devices as a carrier, we find the mapping relationship between the emotional state of the elderly and interface design elements through experiments, and determine the direction of product optimisation based on the analysis of the results, so as to improve the operating experience and emotional state of elderly users, which is of great significance to the research on the ageing-friendly design of relevant interaction interfaces.

2 Overview of the Application of Emotion Regulation and Emotion Elicitation Methods Emotion regulation is a process in which individuals use certain behavioural strategies to influence and control their emotions based on the monitoring and assessment of their internal and external environment, and is a functional response of individuals to maintain internal and external adaptation [4]. Changes in mood are generally caused by external environmental factors, accompanied by individual representations and physiological responses, so emotion recognition methods are used to measure changes in people’s mood and to obtain objective mood data. The process of regulation begins with identifying the user’s emotions through speech, movement, facial expressions at the behavioural level and physiological data at the physiological level to obtain whether the user’s emotions are positive or negative, and for negative emotions, especially for special groups, active emotion regulation needs to be incorporated. Among the existing studies, the SEED dataset created by Lu Bao-gong’s team is dedicated to safe driving, and Susan W. White et al. have developed a facial emotion expression training system for children with autism based on the principle of facial emotion recognition to enhance the early emotional output of children with autism [5–7]. The product is able to recognise and respond to the emotions generated by the user, which is a good way to proactively regulate emotions. In this study, the difference in the effect of emotions on the user is obtained, which requires the use of emotion elicitation to obtain user data. Emotion elicitation is the process of using psychological methods and a series of procedures to induce a specific target emotional state in an individual [8], and is divided into two types: material elicitation and contextual elicitation. The common ones are material evocation, which are utterances, pictures, videos, music and smells. The establishment of a library of standard emotion materials is of great importance in user emotion research [9]. Among them, picture elicitation is the most commonly used method of emotion elicitation, through visual stimulation capable of presenting strongly emotionally coloured pictures and words to produce the emotional state required by the experimental target [8], and a more complete standard library of material stimuli has been formed [13].

Experimental Design and Design Methodology

169

Therefore, this paper chooses the method of picture emotion elicitation to conduct interaction experiments with older people in different emotions. Integrating emotion regulation into the design of the interaction interface and the form of interaction can help to achieve breakthroughs and innovations in existing products, if functional products can recognise and give corresponding feedback, making them emotional and changing from passive emotion de-escalation to active emotion communication, thus achieving the purpose of active emotional interaction.

3 Analysis of Smart Screen Elderly Users’ Usage Needs and Extraction of Interface Design Elements 3.1 Semi-structured Interviews to Obtain Needs and Questions Ten elderly users of smart screens in Xi’an (aged 60–75) were selected for semistructured interviews in the first stage of the study, and were informed of their problems and needs in using the product, as shown in Table 1. Analysis of the interview results showed that the respondents’ needs for the interface of the smart screen were: simpler operation and less distracting information; timely and effective guidance and feedback in case of operation errors; and a less colourful interface, with frequently used functions concentrated and placed in a position that can be easily captured by the eye. Table 1. User requirements and existing problems Usage issues

Usage requirements

Too many functions, too complicated

Removing unnecessary functions that Meet the perceptions and interests of older people

Colourful interface prone to visual fatigue

Simple and easily recognizable colours to avoid attentional distractions

Not skilled in operation, no guidance

Reduce the difficulty of operation, preferably with detailed step-by-step guidance

Too many messages to distinguish correctly sometimes

Add reminders, such as weather reminders, medication reminders, etc

3.2 Analysis and Extraction of Smart Screen Interface Design Elements Interface design elements cover structural elements, interactive elements and visual elements [10], which in smart screen interface interaction design correspond to interface layout, information architecture and colour preference, and are therefore analysed and extracted from these three aspects. Interface Layout. The layout of a product’s interface is a combination of the user’s needs to design the overall architecture, according to the content and form to divide

170

Y.-M. Xue et al.

the structure and modules, so that users can quickly find effective information in the interface [11]. At the same time, taking into account the user’s cognitive style, usage habits and thinking logic, the main functions should be highlighted and the interface information should be accurately conveyed in order to reduce the user’s threshold and enhance the user experience. In this paper, we have researched nearly 20 smart screen products such as Xiaodu Smart Screen and Xiaomi Smart Screen, and analysed their interface layouts. The main layout elements include toolbars, status bars and layout forms, and there are six common layouts as follows: card layout (upper navigation bar ➀), (lower navigation bar ➁), (left navigation bar ➂), palace format layout (upper navigation bar ➀), (lower navigation bar ➁), (left navigation bar ➂) for later experiments, as shown in Fig. 1.

Fig. 1. Sample interface layout

Information Architecture. “Shallow” and “deep” indicate the number of levels of information, the more levels the deeper the level, the more associations the user needs to remember, while the opposite is true for shallow levels. “The wider the level, the greater the number of sub-options and the richer the information available to the user, which in turn puts a strain on the user’s memory. This is why it is important to design a hierarchy that is appropriate to the user’s cognitive ability, so that the user experience is enhanced and the user can efficiently and quickly achieve their goals in the interface”. Colour Preference. Older people’s vision deteriorates with age and they perceive different colours differently. In accordance with the basic principle of the “anti-colour channel” in the colour opposition theory proposed by Hering [12], five pure colours red, green, blue, yellow and black - were selected as typical colour samples. On the basis of the typical colour samples, variations in lightness and purity are used to carry out subsequent experiments. The colour combinations bring about different visual and psychological effects and increase the accuracy of the experimental results.

Experimental Design and Design Methodology

171

4 An Experimental Study of the Relationship Between Negative Emotion Elicitation and the Influence of Smart Screen Interactive Interface Elements 4.1 Experimental Apparatus The ErgoLAB wearable electrodermal and pulse wave recording module and Tobii Fusion oculomotor were used for data acquisition, as shown in Fig. 2. The collected data was processed through the ErgoLAB V3.0 human-computer environment test cloud platform, and the experimental procedure was recorded using the camera.

Fig. 2. Skin electricity, pulse wave, Tobii fusion eye tracker

Fig. 3. Caps negative pictures

4.2 Purpose of the Experiment Older people may have difficulty in searching and make mistakes when using smart screen devices depending on their emotions. The aim of this experiment is to investigate the relationship between emotion and interface design elements through the construction of an emotion-evoking material library, an experiment on the relationship between negative emotion-evoking and interface design elements, the use of physiological data to support the effectiveness of emotion-evoking and eye-movement data to analyse the changes in the visual characteristics of the elderly in different emotional states, and to investigate the mapping relationship between emotion and interface design elements. 4.3 Experimental Subjects Experiment 1 selected 10 young subjects and 10 older subjects aged 60–70 years. In Experiment 2, 10 staff members aged 60–70 were selected as experimental subjects, 5 of each sex. The young subjects and the elderly subjects all had varying degrees of experience in computer operation. The subjects had normal eyesight, no strabismus, colour vision or colour blindness, and were in good health and did not exercise strenuously on the day of the experiment. 4.4 Experimental Stimulus Material The experimental material for the emotion elicitation part of Experiment 1 was 288 negative images from the China Emotion Picture System (CAPS) [13], as shown in Fig. 3.

172

Y.-M. Xue et al.

The experimental material for the interface design elements testing part of Experiment 2 was the interface layout, colour perception, information hierarchy (three layers) pictures extracted from Fig. 1, and negative emotion elicitation samples as the screening results of Experiment 1. 4.5 Experimental Procedure The experimental procedure was divided into two parts, namely, the experiment on the construction of an emotion-evoking material library and the experiment on the relationship between negative emotion evocation and the influence of interface design elements. In order to avoid experimental errors, the experimenter introduced the experimental process and precautions for the subjects, and the whole experimental process is shown in Figs. 4 and 5.

Fig. 4. Flow chart of Experiment 1

Fig. 5. Flow chart of Experiment 2

Experiment 1: 288 negative pictures from the China Emotion Picture System (CAPS) were sorted and screened by young subjects to build a sample bank, and then the sample

Experimental Design and Design Methodology

173

bank was checked for reasonableness by older subjects. This was used to prepare the ground for Experiment 2 and to ensure that the subjects’ emotional states were effectively induced. Experiment 2: (1) Wear physiological equipment and adjust the seat and computer position according to the subject’s visual acuity and other conditions. The sampling values were calibrated to ensure that the equipment could accurately capture the physiological data. (2) To begin the experiment, the subject simply sits still and follows the prompts on the computer. The person in charge of the experiment observes from the right side, at least 1 m apart, to avoid disturbance to the subject during the experiment. (3) At the end of the experiment, the subject stops recording the equipment and removes the physiological equipment for the subject. The subject is asked to complete a subjective questionnaire, asking questions about the subject’s feelings during the experiment, the operation of the experiment and other relevant questions to assist in the subsequent analysis of the experimental results. (4) Repeat the experimental process by replacing the subject. Figure 6 shows multiple experimental scenarios for the elderly subjects in the experiment.

Fig. 6. Scenario diagram of the tested operation

Fig. 7. Negative emotion inducing material

4.6 Experimental Results and Analysis Analysis of the Results of the Construction of a Library of Emotion-Evoking Material. Due to the large number of pictures in the Chinese Emotion Picture System and the fact that some of the pictures may cause discomfort to the elderly, the young subjects were asked to make a preliminary selection of 40 pictures to build a sample of negative emotion pictures. Based on the classification results of the young subjects, the reasonableness of the 40 pictures in the negative emotion picture sample was tested by the older subjects’ perception of the emotion of the pictures and the pleasantness of each picture [14]. Eight photos of feelings of loneliness, anxiety, loss and insecurity were obtained through comprehensive analysis as the negative emotion eliciting material in the subsequent Experiment 2, as shown in Fig. 7. According to the results of Experiment 2, it was verified that the material library was reasonable and could effectively induce the subjects’ emotional states. Analysis of Physiological Signalling Experimental Results. At the end of the experiment, the physiological signal data were exported, and the data were filtered and processed. One group of data was excluded due to equipment failure, and nine groups of pre

174

Y.-M. Xue et al.

and post-experimental electrodermal (EDA) and pulse wave (PPG) data were obtained, which were subjected to descriptive statistical analysis using SPSS software to obtain the before and after comparison analysis graph shown in Fig. 8 (EDA is the pre-experimental EDA Data, EDA’ is the post-experimental data).

Fig. 8. Descriptive statistical analysis

The correlation test was performed on the data of the two pairs of groups (EDA, EDA’) (PPG, PPG’) before and after the experiment using SPSS, and the output results are shown in Table 2. The correlation coefficients of the two sample groups were 0.988 and 0.929, indicating that the correlation between EDA and EDA’ and PPG and PPG’ was extremely high, and according to the correlation coefficients and the probability of significance, indicating that the difference was significant at the 98.8% confidence level and that the difference in the values of the subjects on EDA was significant [15]; 92.9% of the subjects were numerically significant on PPG, meeting the prerequisites for the paired samples t-test. As shown in Table 3 the probability of significance for EDA was 0.031 < 0.05 and for PPG was 0.023 < 0.05 (significance level of 0.05), indicating that there were significant differences in human electrodermal signals and pulse waves before and after the emotion-induced experiment, and that the physiological indicators of the elderly subjects decreased after the emotion-induced experiment, and the manipulation of the interface was significantly disturbed, verifying that the effect of negative emotions on the elderly is extremely serious.

Experimental Design and Design Methodology

175

Table 2. Correlation coefficient of paired samples N

Correlation coefficient

Sig.

To 1

EDA&EDA

9

.988

.000

To 2

PPG(bpm)&PPG

9

.929

.000

Table 3. Paired sample t-test Paired difference mean value

t

Standard Standard 95% confidence value error of interval of mean difference lower limit

To EDA&EDA 1

−.30667 .35132

To PPG(bpm)&PPG 2.55556 2

df Sig.(bilateral)

2.74368

upper limit

.11711

−.57671 −03662 −2.619 8

.031

.91456

.44658

.023

4.66453 2.794

8

Eye Movement Trajectory Map and Hot Spot Map Analysis. The trajectory maps and hotspot maps of the nine subjects were processed by overlaying the data [16] to obtain some of the interface gaze trajectory maps and hotspot maps before and after the emotion elicitation as shown in Fig. 9. From the figure, it is intuitively clear that the subjects’ eye movement trajectories for a given target basically showed a top-to-bottom and left-to-right order due to the subjects’ visual habits. Secondly, it is also clear from the diagram that the focus of attention is concentrated on the key information area, and the attention paths of the palisade list are more scattered and jumpy, which increases the visual load on the elderly users who need to filter the information on the interface in order to find the required content, especially in the case of multiple assignments where distraction is more likely to occur. By comparing the time taken by the subjects to complete the tasks, the advantages and disadvantages of the corresponding functional operation design can be compared. The card layout is horizontally distributed, the gaze track has a certain pattern and there is little diversion and distribution of attention, which effectively avoids distraction. This shows that the horizontal card layout with a more focused gaze track is more in line with the visual characteristics of the elderly, regardless of the emotional state, and can effectively improve the visual search efficiency of elderly users and reduce the distraction and attention load of elderly users. Eye movement interest areas were classified according to colour characteristics. The 5 colour variations were pictured in 5 corresponding interest zones, as shown in Fig. 11 as red AOI-1, green AOI-2, blue AOI-3, yellow AOI-4, and black AOI-5. The mean values of the 3 oculomotor indicators for the 9 subjects in the 5 AOI interest zones classified by SPSS were obtained and are shown in Table 4. The gaze duration and gaze duration

176

Y.-M. Xue et al.

Fig. 9. Emotional evoked locus and hot spot diagram

percentage of the first gaze point were selected as The first gaze time and the percentage of gaze duration were selected as indicators of colour preference in the elderly user interface. From the mean value of the first gaze time, we can see that the first gaze points are mostly in the yellow AOI-4 and red AOI-1 areas. The yellow and red colours are more likely to attract the visual attention of older users, so when designing important information, such as hints and feedback, red and yellow can be used to play a prominent role. From the total percentage of time spent looking at them, it can be concluded that users are most interested in AOI-3, AOI-2, AOI-1, AOI-4 and AOI-5. It can be deduced that blue is the most interesting colour for older people, which is close to the results in the colour preference selection. The comparison of the pre- and post-hotspot diagrams shows that emotion elicitation has essentially no effect on the colour preferences of older people. Analysis of the hotspot diagrams (saturation and lightness) of the elderly after the emotion elicitation in Fig. 10 shows that the oculomotor hotspots of the elderly are in the middle of saturation and lightness. The use of overly saturated colours can cause psychological anxiety and discomfort, so it is important to use colours with moderate saturation and brightness when designing the interface colour scheme. To explore the effect of emotion on interface manipulation, a comparison table of the experimental time spent targeting the information hierarchy section before and after the subjects’ emotion elicitation is shown in Table 5, from which it is visually evident that the older subjects’ manipulation time in the information hierarchy section generally increased after the negative emotion elicitation. By means of the downgrading approach, it is clear that older subjects spend less time operating. Therefore, in the age-appropriate design of the interactive interface, it is necessary to avoid excessive cognitive memory burden and to prevent the elderly from getting lost in the operating system, so it is necessary to integrate emotions and emotional design into the interactive interface design, not only to increase the self-confidence of the elderly, but also to help them integrate into the smart era and enjoy the convenience of smart life.

Experimental Design and Design Methodology

177

Fig. 10. Eye movement area of interest

Table 4. Mean value of AOI eye movement index Eye movement index

AOI-1

AOI-2

AOI-3

AOI-4

0.43

0.61

0.58

0.51

The fixation time of the first fixation point/s Total fixation duration/s Percentage of fixation duration/%

AOI-5 0.73

1.66

1.85

2.27

1.53

0.61

14.93

15.26

20.37

13.62

5.59

Table 5. Partial experimental time of information level before and after emotion induction Subjects

Gender

Age

Pre-emotional triggers

After an emotional trigger

After an emotional trigger

Level 3

Level 3

Level 2

Mr. Tang

Male

60

29 s

32 s

25 s

Ms. Yang

Female

64

18 s

25 s

18 s

Mr.Dong

Male

64

40 s

48 s

26 s

Mr.Ji

Male

63

36 s

33 s

17 s (continued)

178

Y.-M. Xue et al. Table 5. (continued)

Subjects

Gender

Age

Pre-emotional triggers

After an emotional trigger

After an emotional trigger

Level 3

Level 3

Level 2

Ms. Zhang

Female

68

35 s

43 s

27 s

Ms. Huang

Female

65

31 s

40 s

18 s

Mr. Wang

Male

68

37 s

41 s

21 s

Mr. Zhang

Male

66

23 s

29 s

23 s

Ms. Wang

Female

62

36 s

41 s

20 s

Ms. Bai

Female

66

Eye movement data acquisition failed due to interference from physiological equipment monitoring

5 Discussion From the above experimental results and analysis, we know that the significance probability of EDA is 0.031 < 0.05 and the significance probability of PPG is 0.023 < 0.05, both of which are lower than the significance level (0.05), indicating that there are significant differences in human electrodermal signals and pulse waves before and after the emotion-evoking experiment, and the physiological indicators of elderly subjects are significantly lower after emotion-evoking, and the operation of the later interface is obviously disturbed. This demonstrates that negative emotions have a serious impact on the elderly, and therefore the feasibility of using physiological acquisition devices to obtain the mapping between the emotional state of the elderly and the interface design elements to enhance the operational experience and emotional state of the elderly users. Emotions not only affect other types of mental activity and fluctuate with them, but the core of emotion regulation is the interactive process of emotion and cognition. Traditional interaction interface design has been based on the habits and preferences of young people and has focused on functionality, while ignoring the psychological needs and emotional states of older people. In recent years, researchers have been looking for interaction interface designs that can enhance user experience through emotional research from both practical and aesthetic perspectives, but the research is more subjective and lacks objective data support and verification. As a research on the age-appropriate design of interactive interfaces, it is important to consider not only the operational experience of elderly users, but also to integrate the research on emotional experience, so as to improve the usability of interactive interfaces and avoid the rejection and resistance to intelligent interactive interfaces due to the information cognitive impairment and poor emotional experience of elderly users, which may even aggravate the digital divide between the elderly and the existing society. In summary, this paper proposes an experimental design and design research method for smart screen ageing with emotion regulation as the research perspective, as shown in Fig. 11. Firstly, through user demand analysis and interface design element extraction, a library of experimental interface design elements is established. Secondly, based

Experimental Design and Design Methodology

179

on experiments to investigate the influence between emotions and design elements, physiological data is used to support the validity of emotion-evoking experiments, and eye-movement data is used to analyse the changes in visual characteristics of the elderly in different emotional states. Finally, we determine the direction of product improvement based on the influence relationship model and the cognitive, visual characteristics and usage habits of elderly users, and build an emotional interaction interface design to enhance the operational experience and emotional state of elderly users. The conclusions of this paper are based on two experiments: emotion elicitation and emotional interface mapping, and the experimental data and design process can be used as a reference for the design of relevant interaction interfaces for ageing.

Fig. 11. Flow chart of the experimental design method for smart screen ageing from an emotion regulation perspective

6 Conclusion In the paper, the smart screen interaction interface is used as a research case to analyse the needs and usage of elderly users and extract interface design elements from three aspects: interface layout, information architecture and colour preference. Secondly, the negative images from the China Emotion Picture System (CAPS) were used to filter and classify the images and construct a library of emotion-evoking experimental materials. The physiological equipment was used to obtain physiological and eye-movement data of the elderly in the experiment, and the mapping relationship between emotional

180

Y.-M. Xue et al.

changes and interface design elements was analysed, verifying that the impact of negative emotions on people is extremely serious. The design methods are proposed from the perspectives of emotional changes, visual characteristics and cognitive characteristics of elderly users, aiming to alleviate the negative emotion problem of the elderly and provide a more convenient and comfortable life for elderly users, as well as providing a reference for future interactive interface age-appropriate design. Fund Project. National Social Science Foundation of China General Project 22BSH122: Study on the Mechanisms of Perceived Restriction and Negative Emotions on the Integration of the Elderly into the Digital Society and Countermeasures.

References 1. The State Council issued the “14th Five-Year Plan for the Development of Digital Economy”. New Money 20(Z1), 8–11 (2022) 2. Hu, L., Zhang, Z.: Current status and prospects of research on emotions in human-computer interaction. Hum. Ergon. 16(4), 87–90 (2010) 3. Liu, H., Hu, Z., Peng, D.: Theory and research on the relationship between positive and negative emotions. Adv. Psychol. Sci. 26(02), 295–301 (2008) 4. Zhao, Q.: Overview of research on emotion regulation. Sci. Popular (Sci. Educ.) 76(10), 141 (2012). https://doi.org/10.16728/j.cnki.kxdz.2012.10.136.KOELSTRA 5. Koelstra, S., Muhl, C., Soleymani, M., et al.: DEAP: a database for emotion analysis using physiological signals. IEEE Trans. Affect. Comput. 3(1), 18–31 (2012) 6. Zheng, W.L., Lu, B.L.: A multimodal approach to estimating vigilance using EEG and forehead EOG. J. Neural Eng. 1(2), 026017 (2017) 7. White, S.W., Lynn Abbott, A., Wieckowski, A.T., et al.: Feasibility of automated training for facial emotion expression and recognition in autism. Behav. Therapy 49(6), 881–888 (2018) 8. Liu, Y.: A review of research on emotion elicitation methods for emotional materials. Talent 18(02), 232 (2018) 9. Fu, Y.: A review of research on emotion elicitation methods. Hubei Agric. Mech. 41(16), 43–45 (2019) 10. Wang, S.-L.: Computer Graphical User Interface Design and Application. Peking University Press, Beijing (2014) 11. Hu, Z.: Research on the interaction design of elderly smart TV interface based on cognitive psychology. Shandong University, Shandong (2019) 12. Hering, E.: Outlines of a Theory of the Light Sense. Harvard University Press (1864) 13. Bai, L., Ma, H., Huang, Y., Luo, Y.: Development of the Chinese emotion picture system. In: Proceedings of the 10th National Psychology Academic Conference Abstracts, p. 595 (2005) 14. Xue, Y., Dai, Y.: Research on the influence of web design elements on the emotional experience of PAD. Decoration 61(02), 124–125 (2018). https://doi.org/10.16272/j.cnki.cn11-1392/ j.2018.02.029 15. Jiang, M.Y., Li, Q., Wang, D.X., Sun, D.J., Xu, X.: Study on usability of user interface design elements of radiotherapy software based on eye movements and physiological signals. Packag. Eng. 43(04), 163–168 (2022). https://doi.org/10.19554/j.cnki.1001-3563.2022.04.019 16. Tian, Q., Lv, S.: Optimal design of refrigerator operation interface based on illumination and eye-movement experiment. Packag. Eng. 42(24), 230–236 (2021). https://doi.org/10.19554/ j.cnki.1001-3563.2021.24.027

Aging and Social Media

Strategies for Fostering Senior Citizens’ Active Participation in Online Communities Rafaela Carrilho1

, Liliana Vale Costa2(B)

, and Ana Isabel Veloso2

1 Education and Psychology Department, University of Aveiro, Aveiro, Portugal

[email protected]

2 DigiMedia, Department of Communication and Art, University of Aveiro, Aveiro, Portugal

{lilianavale,aiv}@ua.pt

Abstract. The aging of the world population and growing increase in the number of information and communication technology users have posited great challenges in addressing interface design and finding strategies for fostering their online participation. Although the benefits and recommendations for developing online communities for active ageing have been widely recognized, there has been far too little attention to empirical studies that analyze a set of recommendations to undertake these online communities within institutional practices underlying the different pillars of active ageing established by the World Health Organization, i.e., Health, Safety, Participation in Society, and Lifelong Learning. The aim of this paper is to identify a set of strategies for fostering senior citizens’ participation based on a seven-session activity package involving 12 senior citizens using the online community “miOne.”. These recommendations included trustworthiness in web design, active listening, and incorporation of the social networks in the physical space within the online community. Keywords: Online communities · Senior Citizens · Active Participation · Information and Communication Technologies

1 Introduction The past years have seen increasingly rapid growth of Internet users in the 65+ age bracket (HelpAge International & Fundo de População das Nações Unidas [UNFPA] 2012; Nimrod 2014; World Health Organization 2005), amplified with the COVID-19 pandemic and need for social connectedness, access for information, and social support (Bailey and Ngwenyama 2010; Czaja et al. 2018; Nyman and Isaksson 2015). Although some studies (e.g., Nimrod 2009, 2014; Veloso et al. 2020) highlight the benefits and recommendations for the development of online communities for active aging, there has been a general lack of empirical studies on the senior citizens’ participation strategies adopted in the context of online communities. Online communities can be defined as virtual and regulated spaces of social interaction, in which members gather around a common purpose, and thus presenting an enormous potential to strengthen communication and a sense of belonging that goes © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 183–193, 2023. https://doi.org/10.1007/978-3-031-34866-2_14

184

R. Carrilho et al.

beyond the physical space (Nimrod 2014; Preece and Maloney-Krichmar 2003; Preece et al. 2003). In this sense, this research set out to identify the strategies adopted to foster senior citizens’ active participation in online communities. For that, current strategies adopted in online communities were mapped, and others were proposed considering the example of the online community “miOne.” (mione.altice.pt). This paper is structured into the following sections, including the Introduction and Conclusion: Sect. 2 reviews the use of online communities for active ageing and the example of “miOne.”; Sect. 3 describes the research method, recruitment procedures, and participants; and Sect. 4 discusses the results.

2 Online Communities for Active Ageing Online communities tend to play a great potential in establishing and maintaining social ties that may impact active ageing. For instance, Nimrod (2014) highlights that developing products aimed at the aging population may help overcome geographical and temporal boundaries, possibly stirring moments of joy, stimulation, and companionship. Indeed, online communities enable emotion expression, relationship building, and role-making that are relevant to identity formation (Kim 2000; Nimrod 2009). In this sense, Kim (2000) draws our attention to the following roles that a community can foster: (a) support providers (provide feedback and problem-solvers); (b) hosts (keep the activities running); (c) greeters (welcome newcomers); (d) cops (regulate the community and content checkers); (e) event coordinators (plan, and coordinate events); (f) teachers (tutor the community members); and (g) merchants (feed the community economy). When considering the typical life cycle of an online community (inception > creation > growth > maturity and death) (Iriberri and Leroy 2009), these may advance through the following stages: visitors; novices; regulars; leaders; and elders (i.e., long-time regulars or leaders, who share their experience and knowledge) (Kim 2000). These online communities may be defined by its structure, rules, size, culture dependent on the members’ social interactions and the policies that guide them, i.e., sociability (Preece et al. 2003). In this sense, a common goal, interest, or activity provides a sense of purpose and reciprocity of information among members. For that, Preece and Maloney-Krichmar (2003) highlight the necessity to determine different community roles, mechanisms to exhibit members’ reputation, and ensure novelty in events or rituals. It is worth remembering that the term community has evolved over the years as it is part of the human condition – i.e., working in groups and community to survive and reproduce. With the advancements of information and communication technologies (ICT), the social dynamics, and relationships proper of communities (Kim 2000; Preece and Maloney-Krichmar 2003) extended to the online space (Nimrod 2014). Indeed, online communities are virtual environments delivered using digital platforms regulated with social norms and policies (Preece and Maloney-Krichmar 2003; Preece et al. 2003; Nimrod 2014). It is important to notice, however, that online communities differ from social network sites. Whereas online communities refer to online spaces

Strategies for Fostering Senior Citizens’ Active Participation

185

in which users gather towards common interests using computer-mediated communication, social network sites are organized around users (Preece and Maloney-Krichmar 2003), extending offline relationships. Considering what has been mentioned, one may suppose that online communities may play a key role in maintaining social connections crucial to aging well, overcoming geographical distance, and fostering social contact (Leist 2013). Other documented benefits of these contributions include the following: fostering self-confidence, maintaining a sense of usefulness; identifying with peers; providing an opportunity for self-knowledge, growth, and change; and compensating for the loss of roles, affecting online and offline well-being; and promote autonomy (Nimrod 2009, 2014). Using ICT, these may contribute to the sense of independence (Azevedo 2019), empowerment (Nimrod 2014), peer interaction for information exchange and mutual support (Burmeister 2012) through learning and digital literacy programs. Major limitations seem to lie, however, in the lack of time, followed by the lack of energy and other offline responsibilities. Limitations with the system and other users also stand out (Nimrod 2014). In all, the online communities seem to very likely contributing to active ageing whether the users are digitally included and involved in the design process. Before moving to presentation of the empirical work, the next subsections cover digital inclusion, and an example of an online community that involved the end-users in the process and served as the basis for this study, i.e., the online community “miOne.”. 2.1 Being Digitally Included in a Later Age The Internet has reconfigured how society functions, and the proliferation of low-cost Internet access has facilitated connectivity, enabling it to reach new users (Plant 2004). As information is transmitted more quickly and social connections are reinforced, there has been an increasing demand to disseminate goods and services from young to older age cohorts (International Longevity Centre Brazil 2015). According to the Organization for Economic Cooperation and Development (2014) and Pew Internet and American Life Project (2010), senior citizens are the fastest growing age group among Internet users (Nimrod 2014). The National Institute of Statistics (INE) reports that, the proportion of internet users between 55 and 64 years old corresponded to 71% and 65 and 74 years old was 47.7% in 2021 (Instituto Nacional de Estatística 2021). In fact, learning ICT in a later age tend to be dependent on the older adult users’ acceptability of digital devices, teaching method, alternative forms of communication with family and friends, intergenerational relationships, and access to entertainment, leisure, and cultural context (Dias 2012). In the current days, being involved and present in the digital space has become imperative especially to older adults given that access to many goods and services has been made online, as well as reinforce contact with both weak and strong social ties. Still, the exposition to certain dangers (e.g., scam, behavior addictions) or lack of access and exclusion are some of the barriers to overcome. Having presented the importance of being digitally included in a later age, the next subsection presents the “miOne.” online community.

186

R. Carrilho et al.

2.2 The Online Community for Active Ageing “MIOne” “miOne.” is the name of an online community developed under the research project SEDUCE 2.0.1 – Use of Communication and Information in the “miOne.” online community. The purpose of this project is twofold: (a) to assess the impact of psychosocial variables and Online Sociability of senior citizens using ICT; and (b) contribute to the growing development of the “miOne.” community with the participation of senior citizens. This community involved the end-users in the requirement definition and design process, following a participatory design approach, in which the end-users interacted with the platform and a set of services (Fig. 1). In the later stages of its development, it attracted different generations, and the community extended the contents and services to an intergenerational target group – i.e., communication (synchronous and asynchronous communication services); news (access to news information); games (virtual games); share (users’ publications); and health (health information).

Fig. 1. Comunidade Online “miOne.” https://mione.altice.pt

In addition, this community is divided into private and public groups that constitute an online space for social interactions and share content. Users can share and react the news, videos, links, and other relevant pages and files beyond expressing their opinion through comments on the publications. In the sections that follows, the procedures undertaken to propose the strategies that can foster the end-users’ participation and results are presented.

1 Https://www.seduce.pt 20/02/2023.

Strategies for Fostering Senior Citizens’ Active Participation

187

3 Method A qualitative paradigm was adopted to answer the research question What are the strategies to foster the active participation of senior citizens in online communities? and, as such, identify a set of strategies for encouraging the senior citizens’ active participation in online communities. This aim encompasses the following specific objectives: • Identification of a set of strategies and activities to promote senior citizens’ active participation in online communities, grounded on the literature review and perspectives from different practitioners that develop initiatives to encourage their active role in the community. • Dynamization of activities with the Ílhavo Ageing Laboratory, using direct observation and a field diary. • Evaluation of the effectiveness of the proposed activities with the application of a post-activity questionnaire. 3.1 Participants’ Selection and Procedures To address the strategies adopted to foster the senior citizens’ participation in the online community “miOne.”, the Ageing Lab – Laboratório de Envelhecimento in Ílhavo was contacted. This laboratory is a partner of the research project SEDUCE 2.0, aiming to promote knowledge about the ageing process and strengthen a network of researchers, students, enterprises, among other organizations with communal activities and active role in society. There was a total of 12 participants in a two-hour activity distributed in seven weeks to foster the participants’ confidence in the Internet, community knowledge, games, and content sharing. The selection criteria for participation are (a) interest in using ICTs; (b) age 50 or older, people who are preparing to cease professional activity and thinking about active aging; (c) voluntary participation; and (d) interest in being a user of the “miOne.” community. The empirical work relied on direct observation and field notes, in which the participants’ interactions, difficulties, and lessons learned were registered. Then, the participants were surveyed about their attitudes towards their expectations toward the activities, usage of the online community “miOne.” and the services offered, and what their perspective relative to the contribution of the activities performed for active ageing. 3.2 Ethical Considerations The data collected in this study follow the procedures approved by the Ethics and Deontology Council of the University of Aveiro for the project “SEDUCE 2.0 - use of Communication and Information in the “miOne.” online community by senior citizens” (FCT Project POCI-01-0145-FEDER-031696). These ethical procedures include (a) consensus for all participants; (b) voluntary participation in the study; (c) confidentiality and anonymity in data collection and analysis. The application of the questionnaires follows the General Data Protection Regulation.

188

R. Carrilho et al.

4 Results A set of activities distributed in seven weeks were proposed to foster active ageing addressing the following recommendations and topics based on the literature search (Baez et al. 2019; Kope´c et al. 2017; Gschwind et al. 2015; Chao Scherer et al. 2013; Mayas et al. 2014; Wollersheim et al. 2010; Schell et al. 2016; Sauvé et al. 2015; Zhang et al. 2017).and perspectives from practitioners in the field reported in the second author’s dissertation (Carrilho 2022): • Enable scaffolding in the activities and gradually introduce complex tasks. • Encourage mentorship among senior citizens with different levels of digital literacy skills. • Create communal challenges and activities that may reinforce story sharing and relations. • Foster cooperative and competitive dynamics in terms of learning content delivered through these communities. • Consider the following activities, i.e., Physical Activity; Cognitive and Sensory Activities; Cultural Activities; Plastic Expression Activities; Playful Activities; Religious Activities; Social Activities; Educational Activities, and Miscellaneous Activities. Based on these recommendations, the activities presented in Table 1 were proposed. There was a total of 12 participants involved these activities, in which 58.33% (n = 7) were female and 41,67% (n = 5) male. Most of the participants aged between 70 and 79 years old, and have the fourth year of school (58.33%, N = 7). Table 1. Comunidade Online “miOne.” https://mione.altice.pt Activity Description

Procedures

Equipment

Internet Game This activity aims to inform participants about the Internet and how to use it. It will allow the acquisition of knowledge and cognitive stimulation at the level of attention and memory

This activity aims to attract new Computer members to the community and Projector encourage the participation of regular Pressing Button members, enabling mentorship. A quiz was used, encouraging both team collaboration and competition

Internet Safety Workshop Following the Internet game, this activity aims to provide participants’ confidence while using the Internet, and specifically the “miOne.” online community

Lecture sessions with practical Computer activities, and a short quiz are Projector provided for participants to assess their knowledge and earn a member distinction. Post-session challenges relative to the topic of Internet Safety are, then, shared in the community (continued)

Strategies for Fostering Senior Citizens’ Active Participation

189

Table 1. (continued) Activity Description

Procedures

Equipment

Health Information Sources In this session, participants acquire knowledge on how to search for credible health information sources. This action stimulates the participants’ self-expression on health in the online community groups

Lecture session with practical activities. Throughout the week, participants have a quiz on the “miOne.” community so that they will be able to apply what they have learned and recall the learning content

Computer Projector

Artistic club This activity aims to contribute to the maintenance of the community and strengthen interpersonal relations in the community; acquisition and transmission of knowledge; cognitive stimulation, and creativity

Regular members can actively Computer; participate through suggestions, Mobile phone; selection, and analysis of the activities. This activity will constitute a cultural and leisure space where various themes and issues can be debated, considering each participant’s perspectives and context

Cooking Game This activity aims to contribute to the maintenance and strengthening of interpersonal relations in the community; acquisition and transmission of knowledge; and cognitive stimulation at the level of attention and memory In this activity, participants discuss different proverbs, dishes, recipes, among other subjects related to the theme

This leisure activity allows the discussion of different food-related topics, keeping the “miOne.” community active and fostering members’ relationships related with common interests

Health Bingo This activity aims to spread knowledge about health, through the Bingo game This activity aims to attract new members to the community and foster the participation of regular members

This leisure activity is divided into Bingo game; several rounds. After three numbers Computer; come out each participant who has Mobile phone these numbers on his card, should access the “miOne.” community and answer the question with the same number. Whoever completes the card first and gets the answers right in the “miOne.” community wins

Computer; Mobile phone;

(continued)

190

R. Carrilho et al. Table 1. (continued)

Activity Description

Procedures

Equipment

Plant-miOne In this activity, participants share the process of caring for a plant This is activity that aims to reinforce the interrelatedness of daily activities and sharing content in the community “miOne.”

Participants share the process of caring for a plant and the evolution of the plant in the “miOne.” community to ensure their regularity and “check-in” in the online community

Computer; Mobile phone; Earth; Pots: Plants;

When surveyed about the activities they enjoyed the most in their life with multioptions to select, the majority of female participants mentioned gardening, followed by traveling (67.67%, n = 8), watching television, reading, and walking (58.33%, n = 7), playing traditional games and cycling (50%, n = 6), walking (33.33%, n = 4), playing computer games (16.67%, n = 2), and socializing and singing (8%, n = 1). For males, 16.67% (n = 2) indicated that they liked to go for walks, garden, ride a bicycle, play computer games, go for walks, travel, and watch television, whereas 8.33% (n = 1) indicated that they like to play traditional games, read, cultivate the vegetable garden, listen, and play music, practice gymnastics, and swim. Concerning the access to ICT, most of the participants rarely use it owing to difficulties on how to use the devices. Nevertheless, the activities they would like to perform included: (i) Be present in social networks and read the news; (ii) send emails; and (iii) learn. After assessing the participants’ context, the participants were surveyed about the regularity and activities in the online community “miOne.” and, indeed, most of the participants started to use the community once a week and their motivation relied the most on its potential to make or meet new friends (91.67%, N = 11), participate in community groups (83.33%, n = 10). While performing the activities, the participants had difficulty in accessing to the instant messages and joining groups and, as such, these element visibility and navigability had to be considered in the design of the platform. When asked whether the “miOne.” community, the participants agreed that this contributed to maintaining/establishing contacts, combating isolation, learning, and promoting intergenerational relationships, and well-being. For that, these were the activities most highlighted to foster their participation: (i) Organization of regular activities and help in accessing the community (83.33%, n = 10); (ii) Access using own computer/mobile phone (50%, n = 6) following a Bring Your Own Device strategy; and (iii) Integrate game activities (33.33%, n = 4).

5 Discussion The aging process is multidimensional and idiosyncratic. Given the population’s growth of this age group of the population, finding the strategies adopted to foster senior citizens’ active participation in online communities has been more and more essential. Whilst

Strategies for Fostering Senior Citizens’ Active Participation

191

reviewing the literature on the subject, one may suppose that online communities seem to promote the end-users’ independence, socialization, education, physical and cognitive maintenance. These strategies are even more important given that the need of older target groups for primary and instrumental daily living support have been increasing very recently with the outbreak (Pimentel et al. 2021). Whereas ICT sound very promising to overcome the barriers of lack of connectedness and enforcing the frequency of contact and overcome physical distances, its lack of access, lack of knowledge on how to use these, accessibility and disjoint between the services/activities offered and motivations of the target group pose great challenges to the gerontechnology field. In this paper, a set of activities were performed aiming to attract new members for an online community for active ageing entitled “miOne.” and ensure regular activities. In specific, the health theme was outlined given the end-users’ motivation to find credible sources of information relative to the outbreak and many playful games and artistic clubs were performed to incite self-expression within the diverse community groups as well as an interlink between what is shared and impact on physical communities (e.g., planting – last session). In all, the sessions corresponded to the expectations of the participants. Throughout the sessions, there was an increase in the sense of mutual help, from accessing the community to solving the different challenges. Also, the participants showed greater autonomy in accessing and using the community. Therefore, considering the activities accomplished and to answer the research question “What are the strategies to foster the active participation of senior citizens in online communities?” the following recommendations to foster senior citizens’ active participation in online communities can be, summarized as follows: (a) Include trustworthiness in web design respecting the end-users’ context and motivations, providing credible information, respecting their identity and be comprehensive of the provided context; (b) active listening of the end-users’ concerns and ensure instrumental and social support in group activities; and (c) incorporation of the social networks in the physical space within the online community. The most important limitation in this study lies in the fact that many institutions were contacted in these periods of the COVID-19 outbreak but only analyzing the case of the ageing lab was possible with a convenient sample of 12 participants. As such, the recommendations proposed may be circumscribed to this context and cannot be applied to other contexts. Acknowledgments. This work is financially supported by national funds through FCT – Foundation for Science and Technology, I.P., under the project UIDB/05460/2020. Thanks are due to FCT/MCTES for the financial support to DIGIMEDIA, through national funds. This work was funded by DigiMedia Research Center, under the project o GIP3_2022 ICSeniorX - Impact of Information and Communication on Senior’s eXperience in “miOne.” online community and DigiMedia ResearchCenter, under the project UIDB/05460/2020.

192

R. Carrilho et al.

References Azevedo, C.: Demasiadovelho para o digital? Envelhecimento ativo e os usos das TIC por pessoas mais velhas no Brasil e em Portugal (Instituto de Comunicação da Faculdade de Ciências Sociais e Humanas/Universidade NOVA de Lisboa, Ed.). Lisboa (2019) Baez, M., Nielek, R., Casati, F., Wierzbicki, A.: Technologies for promoting social participation in later life. In: Neves, B.B., Vetere, F. (eds.) Ageing and Digital Technology, pp. 285–306. Springer, Singapore (2019). https://doi.org/10.1007/978-981-13-3693-5_17 Bailey, A., Ngwenyama, O.: Bridging the generation gap in ICT use: interrogating identity, technology and interactions in community telecenters. Inf. Technol. Dev. 16(1), 62–82 (2010). https://doi.org/10.1080/02681100903566156 Burmeister, O.K.: What seniors value about online community. J. Community Inform. 8(1), 1–12 (2012). https://doi.org/10.15353/joci.v8i1.3053 Carrilho, R.: Estratégias de Participação Ativa do Cidadão Sénior em Comunidades Online. Dissertação de Mestrado, Universidade de Aveiro (2022). http://hdl.handle.net/10773/36197 Chao, Y.Y., Scherer, Y.K., Wu, Y.W., Lucke, K.T., Montgomery, C.A.: The feasibility of an intervention combining self-efficacy theory and Wii Fit exergames in assisted living residents: a pilot study. Geriatr. Nurs. 34(5), 377–382 (2013). https://doi.org/10.1016/j.gerinurse.2013. 05.006 Czaja, S.J., Boot, W.R., Charness, N., Rogers, W.A., Sharit, J.: Improving social support for older adults through technology: findings from the PRISM randomized controlled trial. Gerontologist 58(3), 467–477 (2018). https://doi.org/10.1093/geront/gnw249 Dias, I.: O uso das tecnologias digitais entre os seniores: motivações e interesses, pp. 51–77 (2012). https://doi.org/10.7458/SPP201268693 HelpAge International, & Fundo de População das Nações Unidas (UNFPA): Envelhecimento no Século XXI: Celebração e Desafio. In: Fundo de População Das Nações Unidas (UNFPA); HelpAge International, Londres (2012) Instituto Nacional de Estatística (INE): Hácada vez mais utilizadores do comércio eletrónico, principalmente mulheres. Sociedade Da Informação e Do Conhecimento Inquérito à Utilização de Tecnologias Da Informação e Da Comunicação Nas Famílias, 1 (2021). https://www.ine.pt/xportal/xmain?xpid=INE&xpgid=ine_destaques&DESTAQUES dest_boui=473557834&DESTAQUESmodo=2&xlang=pt International Longevity Centre Brazil: EnvelhecimentoAtivo: Um Marco Político em Resposta à Revolução da Longevidade (2015). https://prceu.usp.br/usp60/wp-content/uploads/2017/07/ Envelhecimento-Ativo-Um-Marco-Politico-ILC-Brasil_web.pdf Iriberri, A., Leroy, G.: A life-cycle perspective on online community success. ACM Comput. Surv. (CSUR) 41(2), 1–29 (2009) Kim, A.J.: Community building on the web: secret strategies for successful online communities. In: Widman, J., Morgan B. (eds.) European University Institute. Peachpit Press (2000) Kope´c, W., et al.: A Location-based game for two generations: teaching mobile technology to the elderly with the support of young volunteers. In: Giokas, K., Bokor, L., Hopfgartner, F. (eds.) eHealth 360°. LNICSSITE, vol. 181, pp. 84–91. Springer, Cham (2017). https://doi.org/10. 1007/978-3-319-49655-9_12 Leist, A.K.: Social media use of older adults: a mini-review. Gerontology 59(4), 378–384 (2013). https://doi.org/10.1159/000346818 Mayas, J., Parmentier, F.B.R., Andrés, P., Ballesteros, S.: Plasticity of attentional functions in older adults after non-action video game training: a randomized controlled trial. PLoS ONE 9(3), 1–10 (2014). https://doi.org/10.1371/journal.pone.0092269 Nimrod, G.: Seniors’ online communities: a quantitative content analysis. Gerontologist 50(3), 382–392 (2009). https://doi.org/10.1093/geront/gnp141

Strategies for Fostering Senior Citizens’ Active Participation

193

Nimrod, G.: The benefits of and constraints to participation in seniors’ online communities. Leis. Stud. 33(3), 247–266 (2014). https://doi.org/10.1080/02614367.2012.697697 Nyman, A., Isaksson, G.: Togetherness in another way: Internet as a tool for togetherness in everyday occupations among older adults. Scandinavian J. Occup. Therapy 22(5), 387–393 (2015). https://doi.org/10.3109/11038128.2015.1020867 Organization for Economic Co-operation and Development (OECD, 2014): How Was Life? In How Was Life?, 23 February 2023. https://doi.org/10.1787/9789264214262-en Pew Internet and American Life Project (PIALP): Four in ten seniors go online (2010). https:// www.pewresearch.org/internet/2017/05/17/technology-use-among-seniors/ Pimentel, M.H., Pereira, F., Teixeira, C.: Impacto da covid-19 em idosos institucionalizados em estruturas residenciais para pessoas idosas. Revista INFAD de Psicología. Int. J. Dev. Educ. Psychol. 1(1), 477–490 (2021). https://doi.org/10.17060/ijodaep.2021.n1.v1.2129 Preece, J., Maloney-Krichmar, D.: Online communities. In: Jacko, A.J., Sears, A. (ed.) Handbook of Human-Computer Interaction. Lawrence E, pp. 596–620 (2003). https://doi.org/10.1145/ 215585.215969 Preece, J., Maloney-krichmar, D., Abras, C.: History of online communities. Encyclopedia of Community: From Village to Virtual World, January 2003, pp. 1023–1027 (2003). http://cit eseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.91.3155 Sauvé, L., Renaud, L., Kaufman, D., Duplàa, E.: Validation of the educational game for seniors: “live well, live healthy!” Procedia. Soc. Behav. Sci. 176, 674–682 (2015). https://doi.org/10. 1016/j.sbspro.2015.01.526 Schneider, R.H., Irigaray, T.Q.: O envelhecimento na atualidade: aspectos cronológicos, biológicos, psicológicos e sociais. Estudos de Psicologia (Campinas) 25(4), 585–593 (2008). https:// doi.org/10.1590/s0103-166x2008000400013 Zhang, F., Kaufman, D., Schell, R., Salgado, G., Seah, E.T.W., Jeremic, J.: Situated learning through intergenerational play between older adults and undergraduates. Int. J. Educ. Technol. High. Educ. 14(1), 1–16 (2017). https://doi.org/10.1186/s41239-017-0055-0 Wollersheim, D., et al.: Physical and psychosocial effects of Wii video game use among older women. Aust. J. Emerg. Technol. Soc. 8(2), 85–98 (2010)

Facebook Community Lurking and Non-participation: Sociotechnical Barriers Among Older Adults in Online Participation Ryan Ebardo(B) and Merlin Teodosia Suarez De La Salle University, Taft Avenue, 2401 Manila, Philippines {ryan.ebardo,merlin.suarez}@dlsu.edu.ph

Abstract. Online communities based on social media platforms such as Facebook Communities are social spaces where older adults participate in various forms. A body of work has investigated the adoption, continued use, and motivations of social media and online communities. However, published research relied vastly on online data generated by conversations or qualitative data from those who regularly participate. In this exploratory study, we invited 12 members of older adults at least 60 years old from a Facebook community who passively observed or were previously active but disengaged to identify the sociotechnical barriers to online social participation. To identify our participants, we observed the online community of Filipino older adults from 2019 until 2020. This study presents an alternative lens in late-life research through a contextualized understanding of Facebook community non-participation. Thematic analysis of in-depth interview transcripts revealed that the sociotechnical barriers are challenges to keeping up with technology, prior negative experiences with social media, and issues with the community. We discuss the implications of our study to various stakeholders of late life and social media research. Keywords: Online community · Social media · Facebook · Age · Sociotechnical · Barriers · Older adults

1 Introduction The uninterrupted growth of social networks globally for all segments of society is undeniable. Social networks allow users to express themselves online, share personal milestones and achievements, and seek and maintain connections [11]. Social networks are sustained by the consistent flow and consumption of information across all sectors worldwide and by customizing user feeds based on profile and context to promote interactivity. Because of its pervasiveness, researchers are drawn to understanding the nature of complex interactions within the network, its usefulness, and its effects [22, 29, 35]. Lurkers, or those who observe, and dormants, or those who disengage, are considered the silent majority of an online community [20, 39]. This is primarily due to the popularity of archival data generated by online platforms that can be collected and analyzed. Despite their sheer number, prior investigations favored heavily active older adults in © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 194–204, 2023. https://doi.org/10.1007/978-3-031-34866-2_15

Facebook Community Lurking and Non-participation

195

online communities as their research subjects [6, 13]. While such practices uncovered previously undiscovered insights into the online social lives of older adults or those who are at least 60 years old, they also neglected research subjects who did not participate or stopped their engagement with social media primarily due to the difficulty of identifying these individuals. To address this underrepresentation, the study’s objective is to identify the barriers that prevent older adults from socially participating in an online community. Using netnography [19] as our methodology, we initially adopted a passive approach to identify our target participants. We transitioned to active netnography by reaching out to our participants and conducted interviews to identify the reasons behind their non-participation to answer our research question: What are the sociotechnical barriers to participation by older adults in an online community? In the following sections, we discuss prior investigations related to our inquiry, detail our research methodology and relate our findings to the context of older adults and social media. Our concluding section summarizes our results, acknowledging the study’s limitations and implications for research.

2 Older Adults and Social Networks Early studies on social networks view them as a knowledge-sharing platform, trying to understand knowledge contribution behavior [11], and what influences user satisfaction within the network [35]. The findings from these and similar work indicate that social factors such as ethical culture, social ties, and sense of belonging within one’s network positively influence knowledge contribution behavior. This is especially true for virtual communities formed towards special interests (like cooking, sports) or specific profiles (such as a community for teachers, or retirees). Members of virtual communities subscribe to certain conduct within it and benefit from positive emotions elicited from being part of the group. User satisfaction, on the other hand, is influenced by their perceived ease of use of the network, and the perceived usefulness of the network, whether to keep themselves updated with news, express their thoughts, or be entertained. While its usefulness is well-accepted, studies have also been conducted on the dangers of social networks. It has been used to spread fake news [24, 25], “houses” trolls who influence opinions and create artificial popularity[26], and is addicting [29]. The active and passive dichotomy phenomenon in social networks is also widely studied. Early studies indicate that active users benefit from social network engagements, while passive ones’ well-being is harmed because it elicits tends to elicit envy [28]. An active social network user posts and interact with content by liking, commenting, and resharing activities. A passive user limits engagement to browse other people’s posts, profiles, or comments. The popularity of social networks such as Facebook, Instagram, Twitter, and Tiktok has produced abundant, varied explorations of its complexity, effects, and expediency. However, little attention has been given to its nature, i.e. it is an assimilation of both social and technological [15]. Moreover, most of these studies focus on typical users of social networks – teenagers and adults. Very little work has been done to understand social networks’ behavior, influences, and effects on older adults.

196

R. Ebardo and M. T. Suarez

3 Methodology Online communities provide rich insights into older adults’ lived realities, and netnography is a practical approach to depict the online culture in digital spaces [40]. Understanding the world beyond the physical and social reality through analysis of naturally occurring data in online communities is the primary goal of netnography which was found effective by prior studies in the ethnographic research of online communities by older adults [3, 18, 41]. In netnography, the site provides the field where the researcher observes and collects data on the everyday social realities of its community members. The Site is a private Facebook community established in 2017 where membership is limited to older adults who are based in the Philippines, at least 60 years old and membership application was approved by the administrators. Membership as of the end of 2022 is 69,700 community members. The first author, RE, was embedded within the Facebook community for two years from 2019 until 2021. Despite studies showing that lurking or listening are two of the most common behaviors in online communities, research is yet to unravel why older adults refuse to contribute, post, or interact [20, 39]. Our recruitment strategy involved observing the Facebook community and identifying members who did not post nor commented in any post on the platform as “lurkers” like the classification used by prior studies of social media use by older adults [6]. We also identified members who have previously contributed a few posts or comments but registered no activities for at least a year in what literature referred to as “dormants” [20]. Using direct messages through Facebook messenger, we sent invitations to eighteen (18) participants, and twelve (12) agreed to participate in the study. All participants are older adults or 60 years and above, provided their informed consent, and have been in the community for at least a year. The profiles of the participants are shown in Table 1. Table 1. Participants Participant ID

Number of Months in the Community

Type

Gender

001

35

Dormant

Male

002

12

Dormant

Female

003

23

Lurker

Female

004

18

Lurker

Female

005

15

Dormant

Male

006

21

Lurker

Male

007

22

Dormant

Female

008

13

Lurker

Female

009

15

Dormant

Male (continued)

Facebook Community Lurking and Non-participation

197

Table 1. (continued) Participant ID

Number of Months in the Community

Type

Gender

010

18

Dormant

Female

011

17

Lurker

Female

012

23

Dormant

Female

While most of the studies using netnography collected social media data such as profile descriptions, posts, and comments as its primary data source, interviews allow researchers to capture community members’ online and offline experiences [13]. In addition, interviews allow researchers to capture a neglected facet of online communities by capturing the reasons behind non-participation by directly interviewing the participants beyond what they do inside the community [14]. Aligned with the objectives of our study, we conducted our interviews, which ranged from 17 min to 56 min. All interviews were transcribed and compiled in a dataset uploaded to a qualitative software called Dedoose [1, 10]. RE, assisted by a research assistant, JBT, applied codes to portions representing meaning in the transcript using a predetermined codebook. In addition, for every code application, both coders used memos to capture their reflections for excerpts that are difficult to code or may represent a new code. These memos were used during coding sessions between RE and JBT, where agreements and disagreements in the application of codes were discussed. The second author, MTS, assisted in the finalization of codes and the labeling of the themes. Associations between codes represented implicit meanings in the interviews and formed the themes using the code co-occurrence technique of Dedoose [2, 36]. The process resulted in 5 updates to the codebook, 11 codes, and 12 memos. This study ensured trustworthiness in the analysis stage by using a codebook to ensure replicability, adhering to consensus code application, transcribing the interviews to establish credibility, used memos in reflexive coding for validity [17, 32, 33]. We also identified exemplars for each theme and identified the specific participant and the line in the transcript to establish traceability. In the next section, we present our major themes, along with its attached codes and exemplars from the transcript, and provide an explanation of our findings.

4 Discussion of Results Analysis of the interview transcripts revealed three major themes, namely 1) Keeping up with technology, 2) Prior negative experiences, and 3) Community issues. We discuss each theme and the corresponding associated codes. For transparency, we lifted the description of each code from the codebook and provided three (3) exemplars for each theme.

198

R. Ebardo and M. T. Suarez

4.1 Keeping up with Technology To keep up with technology and threats, social media platforms constantly evolve, and changes to their features and login processes deter older adults from actively participating in online communities. Our first theme is “Keeping up with technology” and consists of four (4) codes as shown in Table 2 – Theme 1. Table 2. Theme 1 and corresponding codes Theme 1 - Keeping up with technology Code Name

Description

001

Confusion with new features Refers to the various features of Facebook that participants find overwhelming

002

Login complexity

Difficulty in the way users access their Facebook accounts due to multiple steps of accessing their accounts

003

Lack of kin assistance

The absence of kin ties to assist older adults in using Facebook

004

Lack of technology access

Limited resources to connect to the Internet

The participants shared that they always forget their passwords and will require assistance from their grandchildren. For Participant 005, his grandson used to assist him and type his comments on the platform. He stopped engaging in conversations since his grandson moved away for college. Participant 001 shared that he cannot afford a tablet and is limited only to his smartphone, which was already slow. He also shared that he has limited connections due to financial constraints. He would instead use his Internet to communicate with family members rather than talk with strangers in the Facebook community. Participant 011, a lurker, was unaware she belonged to a Facebook community of older adults. However, she recalls that a friend invited her to join a group chat but ended up in the private Facebook community. The following are excerpts from interviews for Theme 1 - Keeping up with technology: “I used to enjoy chatting in the community before. My grandson used to read the comments on my posts and will ask him to type my reply. He also maintains my password which I always forget because Facebook keeps on asking me to change it. Since he moved away, I am not updated anymore. I tried to read the conversations but I already have a poor eyesight.” – Participant 005 (Codes: Login complexity, lack of kin assistance) “I am on prepaid Internet which my children provide so I can use Facebook. I have no means of income. Whenever I’m connected, I chat with my grandchildren so that I get updated of their lives and consumes my prepaid account. I think this is better than seeing strangers in the Facebook community whom I don’t know.” Participant 001 (Codes: Lack of technology access, lack of kin assistance). “I am surprised I am a member of that group. All the while, I thought that is part of Facebook. I was looking for my children in the members section because I

Facebook Community Lurking and Non-participation

199

thought that was the friend’s list. My children used to teach me Facebook but the new features are just overwhelming.” – Participant 011 (Codes: Confusion with new features, lack of kin assistance) The continuous changes in the interface and the addition of features make Facebook a popular social media feature. For most users, using new versions of social media platforms is easy. Learning to use and navigate in a different environment may take time for older adults, resulting in distress and technology avoidance [37]. Prior research supports this theme as older adults experience cognitive and physical decline and may require support from family members, referred to as intergenerational support in technology use [4, 9]. In online participation, the absence of this support may prevent older adults to interact with others as no household members are present to provide training and guidance in using their smartphones. This kin support is also crucial to ensure that older adults adhere to social media security measures to help them remember their login details, such as their passwords [30, 38]. Lastly, technology access, such as new devices and connectivity, will require financial resources. Facebook and active participation in the community may require devices to be updated and subscribed to faster Internet access. Older adults from developing economies have limited access to these technologies, resulting in lack of online participation [16]. 4.2 Prior Negative Experiences The community feature of Facebook is heavily reliant on the platform itself, and it is common for users to associate their experiences from mainstream social media with their behavior within the community. For example, a friend’s posts can be easily forwarded to a private community. Our second theme is “Prior negative experiences” with three (3) codes, as shown in Table 3 – Theme 2. Table 3. Theme 2 and corresponding codes Theme 2 – Prior negative Experiences Code Name

Description

005

Online scam

Refers to activities that threaten or harm social media users

006

Misinformation

Unverified information circulating online and in social media

007

Unhealthy discourses Conversations online that cause psychological distress, such as arguments, sensitive topics, and politics

Participant 008 shared that she was a victim of an online scam by accidentally clicking a link on Facebook shared by a friend. She refused to interact with the community members and prefers reading conversations during her free time. Participant 009, a dormant, stopped communicating with community members when he shared a post that turned out to be misinformation. This resulted in a discussion thread that made him feel uncomfortable. Participant 004 prefers reading conversations and observed that fake

200

R. Ebardo and M. T. Suarez

news and posts soliciting donations had proliferated the platform since she joined. The following are excerpts from interviews for Theme 3 – Prior negative experience: “I was directed to a dating group because a friend posted a link in the community. I thought it was legitimate because there were reviews. Sadly, I almost lost money and I was traumatized from the experience” – Participant 008 (Codes: Misinformation and online scam) “My friend shared information in his feed and I thought it was a valid news. People started to say negative things about me” - Participant 009 (Codes: Misinformation and unhealthy discourses) “The group has become an online source of fake news and scammers. I am thinking of leaving the group” – Participant 004 (Misinformation, online scam) The proliferation of misinformation and online scams has been widely researched in the purview of older adults [8, 12, 34]. A decline in the cognitive faculty due to late life may impede adequate discernment on what to share in the community, resulting in unhealthy conversations with other members. Prior research revealed that while most conversations afford older adults social connection and participation, some discussions may cause severe stress due to differences in opinions and values [5, 31]. Past stressful conversations in the online public sphere may discourage older adults from further engaging in meaningful discussions with other community members. These past negative experiences affect how they behave online and may result in isolating themselves from the online community or stop engaging in other conversations. 4.3 Community Issues A Facebook community requires administrators to manage membership, enforce community rules, and to some extent, moderate content shared on the platform. Our site was founded and controlled by Mr. J, who assigns moderator roles to various members based on their geographical areas. Politics and jealousy among members resulted in a breakaway group that formed a separate community where they had broader control and power. Our last theme is Community issues with four (4) codes, as shown in Table 4. Table 4. Theme 3 and corresponding codes Theme 3 – Community Issues Code Name

Description

008

Changing community rules Refers to enforcement of new community rules by the site administrators

009

Leadership issues

Trust issues and disagreement with the site administrators

010

Community hopping

Refers to membership to other Facebook communities

011

Privacy

Preference of members to private conversations rather than

Facebook Community Lurking and Non-participation

201

The constant changes in the community rules imposed by the leadership of the community affected some members resulting in disengagement from conversations or transfer to another Facebook community. There needed to be more clarity on the types of content that can be shared. As a result, some members prefer conversing privately rather than being subjected to strict moderation. Participant 007 used to be active but refrained from posting when her attention was called by posting something about religion. She eventually transferred to another Facebook community where she posts regularly. Participant 006 never posted in the community. He finds the rules to be too strict and enjoys the liberty of chatting with other members directly through Facebook messenger. Participant 010 believes there is favoritism in the appointment of site moderators and has stopped being active in the community. She eventually joined another community upon the invitation of a friend she met from the group. “Posting bible verses is natural for me because I’m a devout catholic. When they told me that it was not allowed, I questioned the rule because it appears that they are selective in who to moderate.” – Participant 007 (Codes: Changing community rules, leadership issues) “Chatting privately is enough for me. I am concerned with what people will think if I post it in the community. The community will kick you out if you post something that they don’t like because you violated their rules” – Participant 006 (Codes: Privacy, changing community rules) “Despite being helpful to others, my application to be a site moderator was declined. I think it’s because I had a debate with the site founder. I decided just to be quiet and transferred to another community where they encourage healthy discussions.” – Participant 010 (Codes: Leadership issues, community hopping) Although the size of the online community creates opportunities for more social interactions, it can likewise create problems such as internal politics and difficulty in managing the members. It has been found that group dynamics is an essential ingredient for community cohesion in online communities of older adults to encourage participation [18, 21]. Privacy is also something that older adults value when interacting online. Disclosing something with strangers may be helpful to others but may also make some older adults uncomfortable [27, 37]. Fair moderation practices are also vital to encourage conversations among members, highlighting the community leaders’ vital role in these online social spaces [7].

5 Conclusion, Limitations, and Implications for Future Research To identify the sociotechnical barriers to the participation of older adults in a Facebook community, we used a qualitative approach to uncover three sociotechnical barriers to why older adults adopt lurking and dormant behaviors, namely 1) Keeping up with technology, 2) Prior negative experiences and 3) Community issues. While our findings contribute to the scant literature on lurking and dormancy in online communities of older adults, we were limited by the number of participants who provided insights into their online lives. However, due to the exploratory nature of our investigation, our findings

202

R. Ebardo and M. T. Suarez

open exciting frontiers in social media research from a late-life perspective, given that we have captured a perspective from an underrepresented cluster of older adults in online communities. The findings of our study highlight the crucial roles of intergenerational family support and community leaders in encouraging social participation in an online community. Family members provide an essential support mechanism to educate and assist older adults in learning new features, validate misinformation, and provide resources to access technology. Some of our barriers may not necessarily be a complete disengagement from social media use but merely a transfer to another community or a change in communication modality with more privacy and intimacy. Our study also demonstrated that lurking does not equate to problematic behavior. As our study uncovered, lurking may result from a preference to converse privately or mean that the older adult is unaware that he or she is part of an online community. Lastly, group cohesion is a critical factor in an active online community. Fair practices in content moderation and power distribution may encourage more members to participate and establish healthy group dynamics, volunteerism, and inclusive group norms. While our study did not account for the differences in perceptions based on age and treated our participants as a single group, future research may conduct a similar investigation to a larger sample. A recent study [23] revealed differences in media use technology generations in older adults. It would be interesting to see whether this specific framework applies to non-participation and disengagement in new forms of media. Another research opportunity worth investigating is to capture the perspective of community administrators. How they craft policies, respond to criticisms, and justify their content moderation practices will benefit technology designers in developing tools to automate platform management tools. Lastly, a quantitative approach using our sociotechnical barriers may complement our findings to reach out to more lurkers and dormants for an inclusive understanding of Facebook community use among older adults. Acknowledgments. Our research was made possible due to the participation of older adults from The Site. We also thank the community administrators for allowing the researchers to observe and recruit participants from The Site. This study is part of the dissertation of the first author. We thank the panel members and the conference reviewers for the helpful comments that improved the quality of the paper.

References 1. Andalibi, N., Flood, M.K.: Considerations in designing digital peer support for mental health: interview study among users of a digital support system (buddy project). JMIR Ment. Health. 8(1), e21819 (2021). https://doi.org/10.2196/21819 2. Armborst, A.: Thematic proximity in content analysis. SAGE Open 7, 2 (2017). https://doi. org/10.1177/2158244017707797 3. Balcerzak, B., Nielek, R.: Golden years, golden shores: a study of elders in online travel communities. In: 2017 7th International Conference on Computer and Knowledge Engineering, ICCKE 2017, pp. 199–204. IWW (2017). https://doi.org/10.1109/ICCKE.2017.8167875 4. Beaunoyer, E., et al.: COVID-19 and digital inequalities: reciprocal impacts and mitigation strategies. Comput. Hum. Behav. 111(May) (2020). https://doi.org/10.1016/j.chb.2020. 106424

Facebook Community Lurking and Non-participation

203

5. Berdychevsky, L., Nimrod, G.: “Let’s talk about sex”: discussions in seniors’ online communities. J Leis Res. 47(4), 467–484 (2015). https://doi.org/10.1080/00222216.2015.119 50371 6. Berdychevsky, L., Nimrod, G.: Sex as leisure in later life: a netnographic approach. Leis. Sci. 39(3), 224–243 (2017). https://doi.org/10.1080/01490400.2016.1189368 7. Bizzotto, N., et al.: Misinformation in Italian online mental health communities during the COVID-19 pandemic: protocol for a content analysis study. JMIR Res. Protoc. 11, 5 (2022). https://doi.org/10.2196/35347 8. Brashier, N.M., Schacter, D.L.: Aging in an era of fake news. 29(3), 316–323 (2021). https:// doi.org/10.1177/0963721420915872.Aging 9. Casanova, G., et al.: New older users’ attitudes toward social networking sites and loneliness: the case of the oldest-old residents in a small Italian City. Soc. Media Soc. 7, 4 (2021). https:// doi.org/10.1177/20563051211052905 10. Catedrilla, J.M.: Exploring micro-entrepreneurs’ trust on customers in social commerce : perspective from an emerging economy. In: PACIS 2022 Proceedings, p. 232 (2022) 11. Chai, S., Kim, M.: A sociotechnical approach to knowledge contribution behavior: an empirical investigation of social networking sites users. Int. J. Inf. Manag. 32(2), 118–126 (2012). https://doi.org/10.1016/j.ijinfomgt.2011.07.004 12. Choudrie, J., et al.: Machine learning techniques and older adults processing of online information and misinformation: a covid 19 study. Comput. Hum. Behav. 119(July 2020) (2021). https://doi.org/10.1016/j.chb.2021.106716 13. Costello, L., et al.: Netnography: range of practices, misperceptions, and missed opportunities. Int. J. Qual. Methods 16(1), 1–12 (2017). https://doi.org/10.1177/1609406917700647 14. Duggan, M.: Questioning “digital ethnography” in an era of ubiquitous computing. Geogr Compass. 11(5), 1–13 (2017). https://doi.org/10.1111/gec3.12313 15. Ellison, N.B., et al.: Scholarship on well-being and social media: a sociotechnical perspective. Curr. Opin. Psychol. 46, 101340 (2022). https://doi.org/10.1016/j.copsyc.2022.101340 16. Fang, Y.X., et al.: Digital Divide: an inquiry on the native communities of Sabah. Societies 12(6), 1–22 (2022). https://doi.org/10.3390/soc12060148 17. Fereday, J., Muir-Cochrane, E.: Demonstrating rigor using thematic analysis: a hybrid approach of inductive and deductive coding and theme development. Int. J. Qual. Methods 5(1), 80–92 (2017). https://doi.org/10.1177/160940690600500107 18. Ivan, L.: Using netnography to research older adults’ online communities. In: Neves, B., Vetere, F. (eds.) Ageing and Digital Technology, pp. 133–150. Springer, Singapore (2019). https://doi.org/10.1007/978-981-13-3693-5_9 19. Kozinets, R.V., et al.: Understanding culture through social media data. In: Kozinets, R.V., Dolbec, P.Y. and Earley, A. (eds.) Netnographic Analysis : Understanding Culture through Social Media Data, September 2017 20. Liao, R., et al.: Social support and social satisfaction for older adults in online virtual communities1. In: AMCIS 2017 - America’s Conference on Information Systems: A Tradition of Innovation. 2017-August, vol. 2009, pp. 1–5 (2017) 21. Litchman, M.L., et al.: The diabetes online community: older adults supporting self-care through peer health. Patient Educ. Couns. 101(3), 518–523 (2018). https://doi.org/10.1016/j. pec.2017.08.023 22. Lombardo, G., et al.: Adoption of social media in sociotechnical systems: a survey. Information (Switzerland) 12, 3 (2021). https://doi.org/10.3390/info12030132 23. Loos, E., Ivan, L.: Not only people are getting old, the new media are too: technology generations and the changes in new media use. New Media Soc. (2022). https://doi.org/10.1177/ 14614448221101783

204

R. Ebardo and M. T. Suarez

24. Loos, E., Nijenhuis, J.: Consuming fake news: a matter of age? The perception of political fake news stories in Facebook ads. In: Gao, Q., Zhou, J. (eds.) HCII 2020. LNCS, vol. 12209, pp. 69–88. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-50232-4_6 25. Malik, A., et al.: Antecedents and consequences of misinformation sharing behavior among adults on social media during COVID-19. SAGE Open 13(1), 1–14 (2023). https://doi.org/ 10.1177/21582440221147022 26. Mazza, M., et al.: Investigating the difference between trolls, social bots, and humans on Twitter. Comput. Commun. 196(December 2021), 23–36 (2022). https://doi.org/10.1016/j. comcom.2022.09.022 27. McNeill, A.R., et al.: Privacy considerations when designing social network systems to support successful ageing. In: CHI 2017, pp. 6425–6437 ACM (2017). https://doi.org/10.1145/302 5453.3025861 28. Meier, A., Krause, H.V.: Does passive social media use harm well-being?: an adversarial review. J. Media Psychol. 0 (2022). https://doi.org/10.1027/1864-1105/a000358 29. Miranda, S., et al.: Addiction to social networking sites: motivations, flow, and sense of belonging at the root of addiction. Technol. Forecast. Soc. Change 188(December 2022) (2023). https://doi.org/10.1016/j.techfore.2022.122280 30. Nimrod, G.: The benefits of and constraints to participation in seniors’ online communities. Leis. Stud. 33(3), 247–266 (2014). https://doi.org/10.1080/02614367.2012.697697 31. Nimrod, G., Berdychevsky, L.: Laughing off the stereotypes: age and aging in seniors’ online sex-related humor. Gerontologist 58(5), 960–969 (2018). https://doi.org/10.1093/ger ont/gnx032 32. Nowell, L.S., et al.: Thematic analysis: striving to meet the trustworthiness criteria. Int. J. Qual. Methods 16(1), 1–13 (2017). https://doi.org/10.1177/1609406917733847 33. Roberts, K., et al.: Attempting rigour and replicability in thematic analysis of qualitative research data; a case study of codebook development. BMC Med. Res. Methodol. 19(1), 1–8 (2019). https://doi.org/10.1186/s12874-019-0707-y 34. Shang, Y., et al.: The psychology of the internet fraud victimization of older adults: a systematic review. Front. Psychol. 13 (2022). https://doi.org/10.3389/fpsyg.2022.912242 35. Shipps, B., Phillips, B.: Social networks, interactivity and satisfaction: assessing sociotechnical behavioral factors as an extension to technology acceptance. J. Theor. Appl. Electron. Commer. Res. 8(1), 35–52 (2013). https://doi.org/10.4067/S0718-18762013000100004 36. Talanquer, V.: Using qualitative analysis software to facilitate qualitative data analysis. In: ACS Symposium Series, vol. 1166, pp. 83–95 (2014). https://doi.org/10.1021/bk-2014-1166. ch005 37. Waycott, J., et al.: Not for me: older adults choosing not to participate in a social isolation intervention. In: Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems, pp. 745–757 (2016). https://doi.org/10.1145/2858036.2858458 38. Willard, S., et al.: Development and testing of an online community care platform for frail older adults in the Netherlands: a user-centred design. BMC Geriatr. 18(1), 1–9 (2018). https:// doi.org/10.1186/s12877-018-0774-7 39. Winter, R., Lavis, A.: Looking, but not listening? Theorizing the practice and ethics of online ethnography. J. Empir. Res. Hum. Res. Ethics 15(1–2), 55–62 (2020). https://doi.org/10.1177/ 1556264619857529 40. Zhao, D., et al.: What is discussed about eldercare?: a netnography study on a Chinese online community for older adults. Electron. Libr. 38(2), 239–255 (2020). https://doi.org/10.1108/ EL-10-2019-0225 41. Zhao, X., et al.: Online health information seeking using “#COVID-19 patient seeking help” on Weibo in Wuhan, China: descriptive study. J. Med. Internet Res. 22, 10 (2020). https://doi. org/10.2196/22910

Optimizing WeChat User Experience for Older Adults in China Nan Hu(B) Minnesota State University, Moorhead, MN 56563, USA [email protected]

Abstract. The population of older adults in China has been growing over the past few decades. Urbanization, as part of China’s modernization, prompted young Chinese adults to migrate to large cities. This may give rise to a stratified social structure where the older adults were left feeling isolated. To cope with loneliness and distance during such an unprecedented shift in society, older adults have begun to use social media, including WeChat. However, WeChat’s interface design, in primarily targeting young adults, has made using the app challenging for older adults, whose decreasing cognitive and physical performance due to aging will increasingly plague them with confusion and frustration while using WeChat in their daily communication. In this paper, we sought to identify older adults’ points of tension with using WeChat and apply these insights to improve the WeChat UX for older adult users. Through a user survey and data analysis, we investigated two categories: (1) older adults’ UX with WeChat, including their main motivation for using the app and their preference for using the chat functions, and (2) their problems with using WeChat, including legibility of typography and icons recognition. This paper presents older adults’ social, cognitive, and physical conditions that serve as the driving force behind our identification of UX shortcomings in the WeChat app. We hope that our findings yielded from research and rigorous data analysis can contribute to better UX and service-based design for older adults and direct more attention to their communities from researchers. Keywords: WeChat · older adults · UX design

1 Introduction China’s urbanization, as part of the country’s modernization, has led to the expansion and massive economic growth of urban areas, where the increasing demand for labor has encouraged young people to migrate to larger cities for better education, job opportunities, and medical care. Such rural-to-urban labor migration, however, may give rise to new social changes, distancing young adults from their older parents, breaking family bonds, and isolating communities of older adults. Social media can restore family bonds and improve the overall quality of life (Song et al. 2021). As the most rapidly growing segment of the population in China, older adults have increasingly come to use WeChat, a social app, to remain connected with © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 205–219, 2023. https://doi.org/10.1007/978-3-031-34866-2_16

206

N. Hu

young family members and pursue a better quality of life during their retirement (2021). However, WeChat’s interface design, with features mostly targeting young users, often leaves older adult users confused and challenged due to their decreased cognitive and physical performance as they age. Against that background, the purpose of our study was to analyze data from a survey inquiring into older adults’ user experience (UX) with WeChat. Our hypothesis addresses the app’s frequently used chat functions, typography, and icons that might pose cognitive and physical challenges for older adults. In our study, older adult users refer to individuals at least 55 years old, who are susceptible to declining cognitive and physical functioning. The findings of our data analysis confirmed our hypothesis and revealed the potential problems in WeChat’s tool hierarchy and navigation, typography, and icon design for older adult users. For an alternative, we propose an enhanced interface able to improve WeChat’s UX by making the most-used tools in the app more intuitive and user-friendly for older adults in China.

2 Related Work 2.1 Internet, Smart Phones, and Growing Community of Older Adults in China The development of technology has been a priority in China’s modernization process. In the last two decades, internet coverage and usage in China have seen substantial growth. Internet connectivity has been made available across the country since 1994 (Yang 2011). As of 2019, the China Network Information Center (CNNIC 2019) reported 854.49 million internet users in China, indicating the rise of dynamic internet culture. In 2022, the number of internet users in China reached remarkably 1.05 billion, as reported by CNNIC. Smartphones and Chinese social media, enabled by this technology and cyber-active culture, have become ubiquitous in the daily life of Chinese people in both urban and rural areas. According to data released by China’s National Bureau of Statistics, the average lifespan of individuals in China has increased, from 40.8 years old in 1950 to 74.92 years old in 2010. The 60 years old and older population reached 153 million in 2007. Within just one decade, the numbers spiked again by 33.2% from 2007, which reached 229 million in 2016. One of contributing factors for such expansion, according to literature, is the One-Child Policy (Zhang and Goza 2006). One-Child Policy has led to a 4-2-1 (4 grandparents, two parents, one child) family structure, staggering the growth of the older adult (i.e., ≥60 years old) population. Other factors, such as much-improved medical care, technological advancements, and rising childcare costs, have also contributed to an increase in the percentage of older adults in China. By 2030, it is estimated that the number of seniors (65 years and older) will make up around 23% of the total population in the country (BBC 2012, citing UN 2010). As the chief part of China’s modernization, urbanization has led to the growth of China’s cities and, in turn, greatly heightened the demand for labor there. Better education, more job opportunities with competitive pay and fringe benefits, and advanced medical care constantly encouraged the One-Child generations, including “farmers, laidoff employees, and college students”, to migrate from small towns and the countryside to major cities (Wei and Gao 2017; Liu 2011; Ke 2015). However, as young adults

Optimizing WeChat User Experience for Older Adults in China

207

have increasingly migrated to large cities, family values and bonds derived from a traditional agriculture-driven society have been challenged. Such rural-to-urban migration has separated young people from their older relatives, thereby intensifying loneliness not only among young migrants (Crampton 2022) but also among their older parents. The One-Child policy and the relatively recent surge in labor migration have both meant that many older adults in China now live far from their children, which has resulted a more stratified social structure. In a parallel trend, many older adults confront anxiety and lack family support during retirement. Social media on smartphones contributes to a better quality of life for elderly users (Kiat and Chen 2015). As an antidote, older adults began using the WeChat app on smartphones, not only because the app offers a wide spectrum of chat functions but also because older adults can use it to remain connected with young family members anytime, anywhere. Older adults can also use WeChat as a channel through which to engage with their local communities. Thus, despite being a rather new technology for older adults in China, WeChat has been largely adopted in such communities and used daily as a means to “build new patterns of family bonding” (Ke 2015, p. 1). 2.2 WeChat and Its Frequently Used Functions Released by Tencent in 2011, WeChat offers an extensive range of social tools and features for day-to-day communication. The New York Times has described the app as “a Swiss army knife” (Kessel and Mozur 2022). As Table 1 shows, the WeChat app contains four major sections: Chats, Contacts, Discover, and Me. Table 1. The WeChat app’s four major sections and their features, listed according to hierarchy in the app. Screen shots of four major sections are attached in the appendix section. Chats

Contacts

Discover

Me

Typing

New friends

Moments

WeChat ID

Hold-to-speak

Group chat

Scan

Wallet

Emojis

Tags

Shake

Favorites

⊕ More

Official accounts

Search

My posts

·Album

Contacts (A-Z listed)

people-nearby

Cards & offers

· Camera

Mini programs

Sticker gallery

· (video) Call

WeChat out

Settings

· Voice call · Location · Red packet · Transfer (continued)

208

N. Hu Table 1. (continued)

Chats

Contacts

Discover

Me

· Contact card · Favorites · Files

The Chats section is the main part of the app for communication, offering users all possible forms of instant messaging. Its prioritized default chat functions include typing, hold-to-speak, and emojis. These multiple chat tools are playful and certainly appeal to young users. The app also provides users with more options to choose their favorite chat functions, for the Chats section also has a “More” button: ⊕, which grants users access to (video) Call, Voice call, and some non-speaking features that allow taking photos and short videos, sending red packets (monetary gift), transferring money, and more (see Table 1). The Contacts section primarily displays a list of saved contacts, organized alphabetically. It allows users to participate in exclusive group chats, create personal tags, and set up an official account similar to a blog. It also allows users to search and add new friends by invitation after finding them by using WeChat ID, radar, QR code, or their phone numbers. The Moments feature in the Discover section, a popular spot for both young and older adult users, allows users to check their friends’ statuses, post and share photos and/or short videos, read others’ posts, give likes, and leave comments. Aside from its great appeal among young users, Moments gives older adult users a place to enjoy virtual life without much distraction. Such users enjoy following the statuses of younger family members and to learn what they are doing (Ke 2015). The Discover section also offers users various ways, such as Scan and Shake, of finding potential friends, although that feature appeals most likely to young users only. Last, the Me section allows users to update their personal settings. The highlight of the Me section is perhaps the digital wallet, which supports online banking. Users can sync their bank cards with their digital wallet to pay or receive money. Some third-party add-ons can also be synced with the wallet function, thereby allowing users to pay utility bills, book flights and hotels, reserve cars, purchase movie tickets, and more. 2.3 Aging Due to a natural aging process, older adults experience a decreasing capacity in cognition, mobility, and sensation that can greatly impact their experiences with using social media on smartphones. In the following subsections, we discuss the aging-related characteristics of older adults to provide insight and a deeper understanding of older adults

Optimizing WeChat User Experience for Older Adults in China

209

as our study’s target population. For designers, understanding the core concept of measuring “an individual’s ability to engage in activities” (Blazer et al. 2015, p. 2) is crucial when exploring user-centered design solutions. Cognition in Aging. Among older adults, cognition—that is, a process of human mental reaction to the external world—experiences a decline in memory, attention, thinking, and learning. Approaching research on the aging-related cognitive decline from multiple perspectives, researchers have suggested, based on observations of older adults with dementia, that cognitive decline derives from neuronal cell loss (Price et al. 1998). Other research has added that cognitive decline among healthy older adults can be attributed to disruptions in the neural network other than the loss of cells (Stern and Carstension 2000, p. 15). Among common findings of cognitive decline among healthy older adults relevant to our study, it is widely recognized that older adults generally have less cognitive capacity than younger adults, for research has shown that information processing in mind slows with age (Moscovitch 1982). Many older adults complained that they had the difficulty of remembering things (Herzog 2002). The relative memory decline among older adults compared with their younger counterparts suggests that older adults are slow in reacting to information, “less able to retrace and navigate a rout” (Zajicek 2001, p. 61) and relatively unable to multitask (Salthouse 1982; Wright 1981; Burke and Light 1981, p. 528). Due to such cognitive decline, social media with complicated architecture and flow can be difficult to learn for older adult users (Kiat and Chen 2015). For them, more brief navigation through information can greatly reduce their cognitive load when using apps. Mobility of Hands (Haptic Deterioration). In our work, mobility refers to the finemotor finger motions required to use the touch screen on a smartphone. Decreased hand mobility appears to be common among older adults, most of whom experience decreasing agility in their fingers as they age. Added to that, scientists have noted that hand performance gradually declines in fine-and gross-motor tasks with age (Iwase and Murata 2003; Lepicard and Vigouroux 2010) and eventually becomes ineffective for performing everyday tasks such as typing, writing, and holding objects (Beek 2019; Shiffman 1992). Numerous studies on muscle properties have shown that with age, muscles become stiffer (Tseng et al. 1995), muscle mass shrinks (Keller and Engelhardt 2013), and the amount of non-contractile tissue within the muscle belly decreases (Power et al. 2013). Other studies linking human mobility with the central nervous system (Vijay et al. 2016; Sorond et al. 2015) have suggested that the human central nervous system changes neurologically with aging, which can result in the loss of finger functions (Park et al. 2018). In their study addressing aging and sequential finger movements among young, middle-aged, and older adults, Cacola et al. (2013) found that older adults experience significant declines in their ability to mentally imagine their finger movements, which underscores how declining cognition during aging indeed decreases finger mobility. When using smartphones, older adults may experience that their finger performance is inept, which presents them with challenges in executing prehensile actions such as grasping, gripping, pinching, and holding. Vision and Eyesight. With aging—more specifically, by the age of 45 years—people’s eyesight begins to decline. The cornea is no longer sensitive to stimuli, which renders

210

N. Hu

the eyes unresponsive to external changes. Beyond that, as the lens in the eye loses its flexibility, older adults increasingly experience problems with differentiating colors (Brabyn et al. 2000). Other common vision problems among older adults include a reduced field of vision, problems with glares, and difficulties with reading small text, perceiving depth, and adjusting to dark and light conditions. Because poor vision upon encountering images and text leads to problems with comprehension (Dodd et al. 2017), small text and small Ui graphics can delay comprehension among older adults and significantly increase the time needed to perform tasks. Hearing and Speaking. Aging causes hearing loss, which can be partial or complete. The most common age-related hearing losses are noise-induced hearing loss, presbycusis, and tinnitus. Presbycusis and tinnitus occur as a natural aging process. One noticeable symptom of presbycusis is older adults experiencing slowly diminished hearing. This gradual loss of hearing happens because older adults’ eardrums grow thicker as they age. In addition, increased earwax with aging can also affect the auditory nerves, which produce fewer signals and impulses for the brain to interpret the sound message accurately. Although decreased hearing among older adults is not a significant factor in interface design (Dodd et al. 2017), it should not be ignored when evaluating sound-based tools such as the Hold-to-speak feature in the WeChat app.

2.4 Typography and Icons “Font size is a common app design problem” (Watkins et al. 2014, p. 119). Typography is what people directly read to acquire information. Poor legibility leads to misunderstanding the information. For older adults, a larger type size is recommended for reading. Evidence has shown that older adults, especially those with presbyopia, encounter problems focusing on small text (Beaver 2010). Chisnell (2005) recommended 12-point or bigger for the text on AARP web pages designed for older adult users. The National Institute on Aging of the National Institute of Health (NIH) also suggests at least a 12-point type size make the text senior-friendly. For apps and tablet interface design, some even recommend a 14-point type size (Watkins et al. 2014; Demiris et al. 2001) to achieve good usability for older adult users. Icon interpretation is another significant interface consideration. Studies have revealed that the semantic gap between the icon and its meaning prevents older adults from recognizing the icon’s function (McDougall et al. 1999), for the icon that is symbolically designed has a distance from its semantic meaning. The triangle symbol for “play” in interface design may make no sense to older adults with little computer experience (Xie et al. 2011). To prevent icon misinterpretation, Chisnell (2005) suggests that the images/symbols on icons should be “task-relevant.” Otherwise, icons should be labeled with text. This recommendation is supported by other researchers, including Demiris et al. (2001), who suggest adding text to icons for greater clarity for older adult users.

Optimizing WeChat User Experience for Older Adults in China

211

3 Methodology 3.1 Survey Preparation Based on the findings of our literature review, the questionnaire used in our survey aimed at collecting data related to our hypothesis and revealing points of frustration with the frequently used chat features in WeChat for daily communication (i.e., not the entire app’s interface) among older adults and problems with usability related to typography and icon design. The questionnaire was prepared in both Chinese and English, and a senior scholar in the Department of World Language and Culture at Iowa State University was invited to ensure that no disparities in translation manifested between the two versions of the survey. The questionnaire in English was approved by the university’s Institutional Review Board (IRB) for human research. Because the primary users in our study were older adults living in China, the questionnaire in Chinese was used to collect data. 3.2 Recruitment and Demographic Data To ensure maximum diversity in the sample and the accuracy of data, we aimed to recruit 100 participants. The survey was distributed to potential participants via WeChat and shared in WeChat Moments, and more than 200 people visited the survey link on their smartphones. Because older adults in our study were defined as individuals at least 55 years old, younger participants (n = 92) were excluded from the sample. Ultimately, we received 107 valid responses with the consent form signed. The participants included 47 men, 55 women, and two others who did not disclose their gender. Most participants (89.4%) were urban residents, whereas participants from suburban and rural areas accounted for 5.78% and 4.8% of the sample, respectively. The majority of participants (57.1%) were 55–60 years old, and the percentage of older participants decreased as their age increased. Many participants were retired (69.2%), and although a few had master’s-level education, most had a bachelor’s degree (47.6%) or a middle or high school education (44.7%). We did not inquire about the number of years but the daily hours of using WeChat to make sure participants were active users and meanwhile to get a sense of the frequency of using WeChat among the older adult users. In the survey, most participants reported spending 2–4 h on WeChat per day. 3.3 Data Analysis Quantitative analysis is a “strategy of using a survey to collect data on predetermined instruments that yield statistical data” (Creswell and Clark 2018, p. 18). After collecting data with the survey, we conducted statistical analysis on the data and sought to test our hypothesis and identify the problems that the participating older adults have experienced in their daily use of WeChat. Each survey question served as a variable in the dataset. It is necessary to note that some questions were left unanswered, resulting in missing responses and invalid entries in certain variables. We removed these invalid entries during the data cleaning process to address this issue. As a result, the total number of participants (n) mentioned in the “Findings” section may vary based on the variables used in statistical analysis.

212

N. Hu

4 Findings In view of the results of data analysis, we testified our hypothesis in two categories: (1) older adults’ UX with WeChat, including their primary reason for using the app, user preference of chat functions in Chats, and (2) their problems with using WeChat, including typographic legibility and icon design. 4.1 WeChat’s UX Among Older Adults Purpose of Using WeChat. We hypothesized that older adults primarily use WeChat to maintain close relationships with their family members. Of 73 participants, 48 reported primarily using WeChat to chat with family and friends, and only 12 participants considered using WeChat to chat with their circle of friends as being their primary purpose in using the app. Meanwhile, 13 participants rated using WeChat for online payment as their highest priority when using the app. As the data shows above, our assumption is valid. Rank of Preferential Chatting Tools in Chats Section. Older adults, under the influence of age, lose finger mobility over time. We hypothesized that older adults would prefer to use video call or voice call instead of typing or using the hold-to-speak tool, both of which require multiple finger tasks and relatively high physical effort (Fig. 1).

Fig. 1. A statistic chart showing the rank of preferential chatting tools in Chats section.

However, when we sampled 95 participants, data analysis surprisingly revealed that typing was the mostly used chat function among the participants (37.25%), which clearly contradicts our hypothesis. Such a result could be explained by the fact that many participants were on the younger side of our study’s target age range. For those younger older adults, typing may not yet pose an obvious challenge. At the same time, level of education also arguably affects how people use chat functions in WeChat (Chen 2020), and more than half (55%) of participants, as shown by their demographic data, had completed higher education. To further explore the result, we performed the Pearson correlation test in SPSS to identify the correlation between a preference for using the

Optimizing WeChat User Experience for Older Adults in China

213

hold-to-speak function and age group. Results indicated a significant positive association between those two variables, r(94) = 0.256 p = .012, and a p value far less than .05, which supports our hypothesis that the older the participants are, the less likely they were to use chat functions that demand more physical effort, including the hold-to-speak function that requires holding a button while speaking or chatting on WeChat. Data analysis also elucidated the hierarchy of participants’ preferences for all 13 functions in the Chats section, for we hypothesized that the hierarchy of those functions for older adults would differ from that among younger users, whom the various functions favor. Using a frequency test in the data, we found that video call and voice call were equally popular among our participants and ranked second only to the typing tool. The Camera was also reported to be a frequently used function among older adults but along with the hold-to-speak function was used less than typing, video call, and voice call. Those findings can contribute to designing a brief tool navigation, prioritizing the tools in Chats section that are preferred and frequently used by older adult users.

4.2 Interface Issues Typography and Font Size. Older adults generally have poor eyesight. Difficulty seeing text can lead to problems with comprehension, as we mentioned in Sect. 2.3 Aging. We assumed that older adults would prefer a larger text size for reading. The finding showed that of 100 participants, 57% considered the font size in WeChat as being too small to read, whereas 43% said that the font size was acceptable. Beyond that, 42% indicated knowing how to enlarge the font size in WeChat, whereas 58% indicated not knowing to do so. In general, we can therefore conclude that more than half of the participants (57/100) considered the font size in WeChat too small to read and that nearly the same number (58/100) did not know how to reset the font size (Fig. 2).

Fig. 2. Distribution of answers to what the icon (post in Moments icon) means

Another tension point for older adults using WeChat is recognizing icons. The post in Moments icon (i.e., ), located in the WeChat’s Discover section that older adults enjoy visiting, is the only icon that users can click to initiate the process of making a post (i.e., but not necessarily taking photos) in Moments. The survey revealed, as we assumed,

214

N. Hu

that 82% of participants misread the icon as a button for taking photos or videos; only 13% correctly read it as the means to make a post in Moments.

5 Discussion To serve older adult users with optimized usability, designers should attend to the abilities of older adults as target users and how they interact with the interface in question, and their unique needs “must be examined in light of user interface advances” (Hutchison et al. 1997, p 497). In our study, the finding concerning older adults’ UX in the Chats section of WeChat provides a picture of how chat functions and features in Chats can be and should be re-prioritized based on older adult users’ preferences and needs to provide them a more friendly user experience. As expected, the older adults’ preference for functions in the Chats section, as revealed through the survey, significantly differed from the default setting - the preferences of younger users, who favor certain playful social tools (e.g., hold-to-talk) that older adults do not because the tools require multitasking. Given the fact that deeper navigation in apps is not friendly for older adult users (Rot et al. 2017), rearranging the rank of functions in Chats and prioritizing the chatting tools in WeChat based on older adults’ preferences can greatly shorten their navigation distance and reduce the cognitive and physical effort needed to navigate the tools in Chats. The font size is a critical and often overlooked factor in app design for older adult users. With less sensitive corneas and lenses, older adults have difficulty reading small text and distinguishing colors. Younger users can easily zoom in on small text by pinching their fingers, but this might pose a challenge for older adults. Research results have indicated that a font size of 12 points or larger is necessary for good legibility among older adult users. While a smaller font size might result in a more aesthetically pleasing and trendy look on small screens, it’s important to prioritize functionality for older adults by using a larger font size. In our study, the tested icon “Post in Moments”, as a feature older adults enjoy visiting, was misunderstood by most participants. The original Post in Moments icon is represented by a camera symbol , which does not indicate a direct relevance to posting content. As Chisnell (2005) mentioned that an icon that is symbolically designed may not accurately represent its semantic meaning, designers should aim at making the icon “task-relevant,” thereby making the icons less symbolic and more concrete in the presentation of their function. Otherwise, icons should be labeled. Specifically for older adults, adding a label or text to icons can effectively improve the icons’ usability, as suggested by Demiris et al. (2001). Based on those scholars’ work and our survey results about icon recognition, the Post in Moments icon should be redesigned to be task-relevant and adding a text label for the icon may reduce the cognitive effort needed by older adults to understand symbolic icons and buttons.

6 Design Prototype We developed a high-fidelity prototype based on the data analysis and findings that modifies the function hierarchy in Chats, namely, to reduce the cognitive effort needed to navigate functions and tools, and the icon design, namely for legibility and thus easy

Optimizing WeChat User Experience for Older Adults in China

215

comprehension. Table 2 below shows the different hierarchies of functions (tools) in the Chat section between WeChat version 6.6.1, and the suggested hierarchy of functions for older adults. Table 2. Preferential hierarchy between WeChat version 6.6.1 and the proposed tool hierarchy for older adults Hierarchy of function preference in Chats

Default tool hierarchy in Chats section (WeChat version 6.6.1)

Proposed tool hierarchy in Chats sec. for older adult users

1

Hold-to-speak

Typing

2

Typing

Voice call

3

Emoji

Video call

4

Album

Camera

5

Camera

Hold-to-speak

6

Video call

Red packet

7

(voice) Call

Transfer money

8

Location

Albumen

9

Red packet

Location

10

Transfer money

Emoji

11

Contact card

Contact card

12

Favorite

Favorite

13

Files

Files

Following the proposed hierarchy for older adult users, we designed the prototype shown in Fig. 3. The proposed hierarchy of functions affords older adults quick navigation to the chat functions that they frequently use in the Chat section, which, as mentioned in Sect. 2.3 “Aging,” will reduce the cognitive effort needed from older adult users. As stated by Watkins et al. (2014), “Font size is a common app design problem” (p. 119). Typography is what people directly read to acquire information, and if illegible, it can cause the information to be misunderstood. Evidence shows that older adults, especially ones with presbyopia, encounter problems with focusing on small text (Beaver 2010). Our findings add that the font size in WeChat is too small for older users. As we explained in the Sect. 5 “Discussion”, we used a 12-piont font size in the prototype, functional and legible to older adult users. The typeface used is Huawen Heiti (华文 黑体) for the Chinese version, given its excellent legibility, and the font Myriad for English. The template mockup in Fig. 3 is the iPhone 6s, which was released in 2006 and provides a 5.5-in., 1080p screen.

216

N. Hu

Fig. 3. A high-fidelity visual prototype showing the reprioritized ranking of chat functions for older adult users.

Fig. 4. Revised “Post in Moments” for making the icon task relevant.

Beyond that, we focused on making the “Post in Moments” icon task-relevant for the reasons mentioned in Sect. 2.4 “Typography and Icons”, and Sect. 5 “Discussion.” Fig. 4 shows the redesigned icon for the “Post in Moments”, different from the Post in Moments icon that is a camera icon in version 6.6.1. To make the image of the “Post in Moments” icon task-relevant, we designed a camera shutter icon with a pencil integrated into the camera shutter icon to indicate the nature and function of posting in Moments— that is, uploading photos and text. Principles of icon design such as using a grid system, key-shapes, stroke consistency, corner style, and weight were observed to enhance the icon’s aesthetic quality and visual unity. A label (in Chinese version) was also added to the icon to allow improved recognition for older adult users in China (See Fig. 4 above).

Optimizing WeChat User Experience for Older Adults in China

217

7 Conclusion As more young people have moved to China’s urban areas for better opportunities, older adults in China have increasingly faced loneliness and felt a lack of family and social support. Exacerbating that trend is the common decline in cognition and physical performance in the older adult population. Given China’s growing population of older adults, supporting such individuals and communities has been on the agenda, and significant public attention and support have been given to infrastructure, health care, and design services for older adults to improve their quality of life and well-being. Our literature review allowed us to paint an in-depth picture of our study’s target audience—that is, older adult WeChat users in China who represent the country’s aging population. We explored their UX with WeChat by conducting a survey of more than 100 older adults. Our findings regarding their social, cognitive, and physical conditions served as the driving force behind our identification of UX shortcomings, the guidelines that we have provided, and the design process that we have described in this paper. We hope that our findings through a rigorous data analysis and research process can contribute to better UX, visual graphic, and service-based design for older adults as well as direct more attention to their communities from researchers.

8 Limitation and Future Work Our research was potentially affected by a few factors that might limit the use of its findings. For one, Tencent.com updates the WeChat app frequently, and because our study concerned only WeChat version 6.6.1, it did not consider some new features and changes in more recent versions of WeChat. For another, regarding the diversity of the sample, although more than 100 participants responded to the survey, most were recruited from cities in China, whereas few came from suburban (5.7%) or rural (4.8%) areas. Researchers such as Fisk et al. (2004) and Hawthorn (2003) have suggested that any product targeting older adults should be evaluated by older adults. Thus, future work should involve usability testing and invite older adults to test and evaluate the prototype, which should afford researchers and designers the feedback needed for a more pervasive, practical design.

References Beaver, H.: Refractive error in the geriatric population. In: Beaver, H., Lee, A. (eds.) Geriatric Ophthalmology, pp. 7–14. Springer, Cham (2010). https://doi.org/10.1007/978-3-03004019-2_2 Beek, N., et al.: Single finger movements in the aging hand: changes in finger independence, muscle activation patterns and tendon displacement in older adults. Exp. Brain Res. 237, 1141–1154 (2019) Blazer, D., Yaffe, K., Liverman, C.: Cognitive aging: progress in understanding and opportunities for action. Mil. Med. 180(11), 1111–1113 (2015) Brabyn, J.A., et al.: Visual impairments in elderly people under everyday viewing conditions. Am. Found. Blind 94(12), 741–755 (2000)

218

N. Hu

Burke, D., Light, L.: Memory and aging: the role of retrieval processes. Psychol. Bull. 90(3), 513–546 (1981) Chen, F.: Re: why voice messaging is not often favored by users with higher education background. Discussion post. https://zhuanlan.zhihu.com/p/43119457. Accessed 2020 Chisnell D.: Designing web sites for older adults: expert review of usability for older adults at 50 websites (2005). http://www.aarp.org. Accessed 12 Dec 2022 Crampton, T.: Social media in China: the same, but different. Bus. Rev. https://www.chinabusines sreview.com/social-media-in-china-the-same-but-different/. Accessed 08 Nov 2022 Creswell, J., Clark, V.: Designing and Conducting Mixed Methods Research, 3rd edn. Sage, California (2018) Demiris, G., Finkelstein, S., Speedie, S.: Considerations for the design of a web-based clinical monitoring and educational system for elderly patients. J. Am. Med. Inform. Assoc. 8(5), 468–472 (2001) Dodd, C., Adam, M., Athauda, R.: Designing user interface for the elderly: a systematic literature review. In: Proceedings of Australasian Conference on Information Systems, Hobart, Australia (2017) Feng, T.: Report examines how internet use affects senior life. Chin. Soc. Sci. Today Rev. Anal. 6 (2018) Fisk, A., Rogers, W., Charness, N., Czaja, J., Sharit, J.: Designing for Older Adults: Principles and Creative Human Factors Approaches, 3rd edn. CRC Press, Boca Raton (2004) Graziani, T.: WeChat impact report 2018: all the latest WeChat data. https://walkthechat.com/wec hat-impact-report-2016. Accessed Dec 2019 Hawthorn, D.: How universal is good design for older users?. In: ACM SIGCAPH Computers and the Physically Handicapped, no. 73–74, pp. 38–45. Association for Computing Machinery, New York, NY, USA (2003) Hutchison, D., Eastman, C., Tirrito, T.: Designing user interface for oder adults. Educ. Gerontol. 23(6), 497–513 (1997) Iwase, H., Murata, A.: Design proposals to assist older adults in using a computer mouse. IEICE Trans. Inf. Syst. (E86-D:1), 141–145 (2003) Ke, Y.: Ageing on WeChat: the impact of social media on elders in urban China. J. Vis. Media Anthropol. 1(1), 8–21 (2015) Keller, K., Engelhardt, M.: Strength and muscle mass loss with aging process. age and strength loss. Muscles Ligaments Tensions 3(4), 346–350 (2013) Kessel, J., Mozur, P.: How China is changing your internet. The New York Times. https:// www.nytimes.com/2016/08/10/technology/china-homegrown-internet-companiesrest-of-theworld.html. Accessed 2022 Kiat, B., Chen, W.: Mobile instant messaging for the elderly. Procedia Comput. Sci. 67, 28–37 (2015) Lepicard, G., Vigouroux, N.: Touch screen user interfaces for older subjects. In: Proceedings of 12th International Conference on Computers for Handicapped Persons, pp. 592–599, Vienna, Austria (2010) Liu, Z.: An empirical analysis of the community life of new urban migrants. Chin. Soc. Anthropol. 43(3), 5–37 (2011) McDougall, P., Curry, M., Bruijn, O.: Measuring symbol and icon characteristics: norms for concreteness, complexity, meaningfulness, familiarity, and semantic distance for 239 symbols. In: Behav. Res. Methods Instrum. Comput. 31(3), 487–519 (1999) Moscovitch, M.: A neuropsychological approach to perception and memory in normal and pathological aging. In: Craik, F.I.M., Trehub, S. (eds.) Aging and Cognitive Processes, p. 56. Plenum Press, New York (1982) Park, Y., et al.: Aging differentially affects online control and offline control in finger force production. PloS ONE (2018)

Optimizing WeChat User Experience for Older Adults in China

219

Power, G., Dalton, B., Rice, C.: Human neuromuscular structure and function in old age: a brief review. J. Sport Health Sci. 2(4), 215–226 (2013) Price, D.L., et al.: Alzheimer’s disease: generic studies and transgenic model. Annu. Rev. Generics 32, 461–493 (1998) Rot, A., Kutera, R., Gryncewicz, W.: Design and assessment of user interface optimized for elderly people: a case study of ActGo-gate platform. In: ICT4AWE, pp. 157–163, Porto, Portugal (2017) Salthouse, T.: Adult Cognition: An Experimental Psychology of Human Aging, pp. 133–154. Springer, New York (1982). https://doi.org/10.1007/978-1-4613-9484-6 Song, L., Ge, Y., Zhang, X.: The relationship between WeChat use by Chinese urban older adults living alone and their subjective well-being: the mediation role of intergenerational support and social activity. Psychol. Res. Behav. Manag. 14, 1543–1554 (2021) Sorond, F., et. al.: Aging, the central nervous system, and mobility in older adults: neural mechanisms of mobility impairment. J. Gerontol. Ser. A 70(12), 1526–1532 (2015) Stern, P., Carstension, L.: The Aging Mind: Opportunities in Cognitive Research. National Academies Press, Washington (DC) (2000) Tseng, S., Marsh, R., Hamilton, T., Booth, W.: Strength and aerobic training attenuate muscle wasting and improve resistance to the development of disability with aging. J. Gerontol. Biol. Sci. Med. Sci. 50, 113–119 (1995) Vijay, V., et al.: Aging, the central nervous system, and mobility in older adults: interventions. J. Gerontol. Biol. Sci. Med. Sci. 71(11), 1451–1458 (2016) Watkins, I., Kules, B., Yuan, X., Xie, B.: Heuristic evaluation of healthy eating apps for older adults. J. Consum. Health Internet 18(2),105–127 (2014) Wei, L., Gao, F.: Social media, social integration and subjective well-being among new urban migrants in China. Telematics Inform. 34, 786–796 (2017) Wright, R.: Aging, divided attention, and processing capacity. J. Gerontol. 36(5), 605–614 (1981)

I Know How but I Do not Want to Discern Falsehoods: Older Adults’ Self-Reported Inference Process to Identify and Share Short-Form Videos Wei Hu1

, Honglian Xiang2

, Jia Zhou2(B)

, Wenjun Sun3 , and Jinjun Xia1

1 School of Arts, Chongqing University, Chongqing, People’s Republic of China 2 School of Management Science and Real Estate, Chongqing University, Chongqing,

People’s Republic of China [email protected] 3 China Resources Mixc Lifestyle Services Limited, Shenzhen, People’s Republic of China

Abstract. Short-form videos are penetrating the daily lives of older adults, but falsehoods among them are a major threat to older adults. To understand how older adults perceive and discern fake short-form videos, semi-structured interviews were conducted among 18 older adults. Their motives, sharing behaviors, and strategies to identify the authenticity of short-form videos were analyzed. Three major findings were derived. First, older participants discern falsehoods when they intended to share short-form videos among family members. Specially, older participants with the motive of hedonism did not care about the authenticity of short-form videos. Second, older participants reported four strategies to discern falsehoods. Third, almost all older participants reported that they knew how to discern the falsehoods of short-form videos, but in practice, none of them spent mental efforts to do it. Based on these findings, we propose suggestions for intervention to fight against fake short-form videos. Keywords: Older Participant · Short-form Video · Falsehood

1 Introduction The popularity of short-form videos brings challenges and threats to older adults. The China Silver Economy Insight Report 2021 showed that the usage rate of short-form videos among 62.204 million older internet users was as high as 80.2% [1]. Compared to text, images can depict visual content, occupy more of people’s sensory perceptions, and grab attention and cognitive resources [2]. Thus, images and videos are more attractive and salient [3]. Furthermore, a general tendency among individuals to suspend efforts at cognitive processing in favor of relying on intuition. And that will be exacerbated when facing false audiovisual content [4]. All of these make it difficult to discern the falsehoods of short-form videos. And, fake audiovisual content is more likely to be widely distributed [5]. However, people are relatively less able to recognize this visual © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 220–232, 2023. https://doi.org/10.1007/978-3-031-34866-2_17

I Know How but I Do not Want to Discern Falsehoods

221

deception [6–8]. Especially older adults [9]. This means older adults are more likely to be deceived by visual content. For example, 61-year-old Ms. Huang was deceived by a short-form video in Tik Tok. This video used simple editing techniques to fake celebrities. And Ms. Huang was induced to consume and even ran away from home [10]. The news generated a lot of discussions when it was exposed. It was then that people realized there were many older adults like Ms. Huang who had been cheated. Using short-form videos to publish false advertisements and fraud is increasing. And most of the victims are older adults [11]. Subsequent paragraphs, however, are indented. Some researchers attribute recognition difficulties in older adults to declining visual function [9] or the digital divide [12]. Older adults can be tricked by poor video manipulative techniques, as in Ms. Huang’s case. Moore and Hancock [13] conducted a one-hour digital literacy education for older adults. The results showed the accuracy of educated older adults increased from 64% to 85%. In the traditional belief that seeing is believing [14], older adults may not be aware that videos can be manipulated. And it is not certain whether older adults will discern the falsehoods of short-form videos. Therefore, this study mainly explores two questions through semi-structured interviews: (1) Whether old adults will discern falsehoods of short-form videos. (2) If they will, what are the motives and strategies? Short-form videos were measured in seconds, ranging in length from a few seconds to 600 s. This study mainly refers to videos on Tik Tok.

2 Materials and Methods 2.1 Research Methodology This study only refers to the method of three-level coding used by the Programming Grounded Theory [15]. Analyzing the content of the interview by using NVivo 12 software. First, researchers divided the process of watching short-form videos by older participants into three stages. Second, analyzing the main actions of older participants generated at each stage. Finally, the main behaviors of the older participants were specified. 2.2 Participants In total, 18 elderly people (mean age = 67, SD = 4.7) living in the main city of Chongqing were invited. Researcher numbered the participants. For example, in P15F66, the first letter indicates the participant; the second and third number indicates the serial number of the participant; the third letter indicates gender, M is male and F is female; and the last number indicates the age of the participant. The gender distribution included a total of 11 females and 7 males. According to the self-reported data of older participants, they used multiple apps to watch short-form videos. Of these, 12 older adults use Tik Tok; 8 older adults use Tou Tiao; 7 older adults use Kwai, these are some of the apps that appeared more popular in the interview. Older adults also use WeChat, browsers, and other apps to watch short-form videos. And the demographic characteristics of the participants are shown in Table 1.

222

W. Hu et al. Table 1. Demographical statistics of participants (N = 18). N

Percentage (%)

Mean (SD)

Gender Male

7

38.9

11

61.6

60–64

4

22.3

65–69

9

50.0

70–74

4

22.1

75 or above

1

5.6

Primary school

2

11.1

Middle school

7

38.9

High school

8

44.4

College degree or above

1

5.6

Female Age

67 (4.7)

Education Background

Whether they live with their children Yes

4

22.2

No

14

77.8

2.3 Interview Process Each older adult participated in the semi-structured interview lasting 30 to 40 min. The full process was shown in Fig. 1. Before the experiment started, researchers inform participants of the following information. First, researcher introduced the purpose of this study and the principle of data confidentiality to older participants. Second, asked older participants whether the interview could be recorded, and explained older participants’ questions in detail. Finally, older participants signed an informed consent form for the experiment. Each participant was paid at the end of the interview. The specific interview process was as follows: The formal interview began with basic information about the participants. Then, the older participants were asked to answer the interview questions and self-reporting. Next, researcher conducted a misleading shortform video test on older participants. To avoid spreading falsehoods during the interview, the researcher would explain the content of the test video at the end of interview. The specific content of the interview consisted of two parts: a question-and-answer session and a short-form video test. In the first part: learn basic information about older participants, such as age, education level, what platforms they watch short-form videos on, and how much time they spend each day. Second, researchers wanted to understand the situation of older participants watching short-form videos. For example, what types of short-form videos they watched, and whether they would question the authenticity of the content. Thirdly, understand older participants’ clicking the “like” button, commenting,

I Know How but I Do not Want to Discern Falsehoods

223

and sharing behaviors on short-form videos. And researcher would ask them whether they would question the authenticity of short-form videos before making these actions. Fourthly, when older participants indicated the behavior of discerning the authenticity of short-form videos, researcher would make a detailed inquiry about the strategies for identification. Fifthly, researcher would ask the older participants to self-report how they discern the reality of the event in real life.

Fig. 1. The full process of the experiment.

In the second part: the older participant was asked to watch a 14-s long short-form video test. The video shows a person using a syringe to inject some orange liquid into a crab after taking it out of a bowl on the side. In addition to this, the video also includes two sentences of text content, “Why to inject crabs?”, and “netizens: Seek experts for the popularization of science”. Most users in the comment section believed that merchant was injecting fake crab yolk to sell them for a higher price, as in Fig. 2(a). However, some users clarified that the video was a rumor, as shown in Fig. 2(b), the content of comment in red box. A fact-checking platform, China Internet Network Information Center, once verified this video1 . It is a typical behavior of spreading rumors. Misleading people to think that the normal food handling process is spraying medicine. When older participants watched the video, researchers would observe their behaviors, such as going to the publisher’s home page or clicking on the comments section. After viewing, researchers would interview them again. Finally, researcher would produce printed fact-checking evidence and explain the rumor to the older adults. 1 https://www.piyao.org.cn/2021-04/22/c_1211122866.htm.

224

W. Hu et al.

Fig. 2. Short-form video’s comment interface (https://v.douyin.com/BWVSdEV/)

Only one participant was interviewed at a time. At the end of the day’s interview, based on the specific interview, the interview process was reflected on and some details of the interview script were adjusted before selecting the participants for the interview.

3 Result 3.1 Data Coding Researcher converted the interview recordings into text and imported them into NVivo12 software for coding and descriptive statistics. There are two goals in mind throughout the coding process. The first goal was to understand older participants’ motives for discerning falsehoods. It was extracted from the types of short-form videos watched and viewing behavior by older participants. Excavating their strategies of discerning falsehoods was the second goal. It was extracted from the self-report of older participants. Table 2 shows the three-level coding. Researcher used the structure of coding to provide descriptive statistics on the content of the interviews. The overall situation of older participants in discerning the falsehoods of short-form videos after coding is shown in Fig. 3. 3.2 Exposure to Short-Form Video Older participants said they were exposed to short-form videos for entertainment reasons, or because they had smartphones with more entertainment functions. For example, a 66year-old female participant said, “My family bought a smartphone for me, so I started to

I Know How but I Do not Want to Discern Falsehoods

225

Table 2. Three-level coding table for the content of the interview. Open coding

Axial coding

Selective coding

Exposure to short-form video

Motive or purpose of watching the short-form video

Reason for contact

Behavior related to watching the short-form video (multiple choice)

Give a like

In what situation

Comment Repost and share None of the above

Identify content authenticity

Whether older participants will discern falsehoods of short-form video

Yes

Motives to discern the authenticity of short-form video

Related to self-interest

No Doesn’t matter Related to sharing

Strategies for discerning the Personal or life experience authenticity of short-form Publishing channels video Reasoning by external clues of short-form videos Through the help of a third party Real-life experience

Motives to discern the authenticity of events in real life

Personal or life experience

Attention to fact-checking platforms or information

Will actively follow

Whether official or not Judging from the clues Seen but will not actively follow No attention at all

learn how to watch it (P5F66)”. Another said, “I just think this is fun, and we all started to play it (P8F66) “. Among them, two participants said that they were influenced by their friends. For example, “Friends recommended them, like Kwai or something, we recommended them to each other (P18M62)”, and “when I see people playing, I also want to watch (P1M63)”. The older participants said they usually watch short-form videos when they were leisure or bored. Only one older participant reported that he had fixed time to watch them usually before bed. 55.6% of older participants spent half an hour to three hours a day watching short-form videos. 11.1% reported that they spend five to six hours a day. Two older participants reported that they scheduled their time of watching short-form videos regularly to do incentive tasks. For example, “I will get up in the morning, I will finish that task at five o’clock, and I will watch it when I have time (P8F66)”, “I woke up

226

W. Hu et al.

at five o’clock in the morning, first finished that task, then washed up. At seven o’clock there were so many arrangements, probably watching short-form videos again (P7F67)”. When watching short-form videos, 61.1% of the older participants indicated that they would click the “like” button. There were 27.8% of the older participants would comment. And 77.8% of the older participants would share short-form videos with their families or friends. Older participants have their screening criteria when it comes to sharing short-form videos. For example, they were keen on sharing health-related content: “If it is good for health or good for life, I will share it with my friends (P8F66)”. Or they will share short-form videos that they think are meaningful to others, like “If it is particularly meaningful and helpful to everyone, I will share it, but generally they will not share it (P4M65)”.

Fig. 3. The overall situation of identifying behaviors.

3.3 Strategies and Motives to Discern Falsehoods Researcher counted the older participants’ discernment behaviors first. Then, researcher analyzed motives that trigger such behavior. Finally, four identification strategies were extracted from the self-reports of older participants. First, researcher counted the discernment behaviors of older participants for two dimensions: the content of short-form videos and the sharing behaviors of older participants. Sixteen older participants indicated that they would discern falsehoods. Only participants numbered P9F65 and P14F73 had no identifying behavior in either dimension. Participant numbered P9F65 said she did not care about the authenticity of the content of the short-form videos and never shared them. Participants numbered P14F73 also expressed no care about the authenticity of content. She often shared videos for entertainment purposes. Therefore she did not care about the falsehoods of short-form videos.

I Know How but I Do not Want to Discern Falsehoods

227

Table 3. Example of respondent’s motives to discern the authenticity of content. Motives to discern authenticity

Number of references (N = 13)

Examples

Related to self-interest

7 (53.8%)

“If it is related to their interests, regularly, like this aspect can be concerned about. Doesn’t involve their interests, generally don’t bother to ask these (P1M63).” “What they say in front is still correct, but what they sell at the back, they say they can control the disease, but whether they can control the disease at the back, still need to be carefully identified (P18M62)”

Related to sharing

7 (53.8%)

“I rarely, sometimes I feel something meaningful that I have to send to others to read (P12F65).” “Sometimes, depending on what is said, I think it is reliable I will turn, not reliable I will not turn (P5F66)”

Note: Older participants who mentioned both motives were counted repeatedly

Second, the interview results showed that 72.2% of older participants’ motives could be mainly divided into two categories, as shown in Table 3. Older participants followed short-form videos in news, entertainment, and health categories. News included international news, entertainment news, and local life news. Entertainment included singing and dancing, musical instrument or folk songs, funny videos, and film clips. Health included health knowledge, regimen, exercise, and disease-related knowledge. Older participants were more concerned about the authenticity of news and health short-form videos. But they did not care about the authenticity of entertaining short-form videos. Whether the content of short-form videos were related to personal interests is one of the motives for older participants to discern falsehoods. For example, when a short-form video recommends a health product, older participants will become cautious because it involved their health and money expenditure. An Older participant said, “Sometimes I wonder if that is what the video says, especially in the health industry, like to buy some medicine and health products (P18M62)”. Another motive is when they have the intention to share short-form videos, they will discern falsehoods. Older participants said they would only share short-form video “useful”, “meaningful” or “reliable” shortform videos. These descriptions of short-form videos need to be based on authenticity. For example, “Sometimes, when I feel something meaningful, I will share it with other people(P12F65)”.

228

W. Hu et al.

Table 4. Example of rules for participants to discern the authenticity of short-form video content. Rules

Number of references (N = 16)

Examples

Personal or life experience

15 (93.8%)

“I take it personally, what you think is right and what you think is wrong (P11F62).” “The basis is probably based on the previous knowledge or my understanding of the judgment (P1M63).”

Publishing channels

11 (68.8%)

“Those things on TV are public, sometimes randomly fabricated in WeChat (P12F65).” “Generally Tou Tiao is not noticed because they are a national network and are generally trusted(P3M70).”

Reasoning by external clues of short-form videos

3 (18.8%)

“Sometimes his analysis may not be correct, so I pay more attention to the time when the news appears, and the comments about the news (P16M80).” “It is indeed right is right, if it is not right, well, you can turn the comments (P7F67).”

Through the help of a third party

4 (25.0%)

“We’ll take a look at these on Baidu (P10F67).” “I have to go inside the Tou Tiao to look for the news…… If I find it I will think this thing is true, I generally do not believe if I don’t find it. (P15F66).”

Note: Older participants who used several strategies simultaneously were counted repeatedly

Then, researcher defined four strategies to discern falsehoods based on self-reports of older participants, as shown in Table 4. Personal or Life Experience. Older participants discern falsehoods based on their knowledge or life experience. Publishing Channels. This category refers to the playing platform of short-form videos or who published them. Some older participants believe Tik Tok and Kwai are untrustworthy. But hold a more trusting attitude toward Tou Tiao app. Reasoning by External Clues of Short-Form Videos. External clues refer to information such as publishing time, publisher or comments on short-form videos. For example,

I Know How but I Do not Want to Discern Falsehoods

229

one older participant said, when short-form videos from the news category are published two months ago, he is suspicious of their authenticity. Through the Help of a Third Party. This category represents that older participants will search third-party for help, such as families and friends. 3.4 Short-Form Video Test Among older adults who took the video test, researchers did not observe any older participants using strategies they reported to discern falsehoods. For example, an older participant said he would check the comment section to discern. But in the video test, the participant did not click on the comment section. During the clarification of the video content, researchers understood older participants’ attention to fact-checking platforms. The results are shown in Table 5. Almost all older participants did not follow them actively. Table 5. Older participants’ concerns about fact-checking information and platforms. Attention level

Number of references (N = 17)

Examples

Active attention

1(5.9%)

“Sometimes look, my daughter helped me download an app like this (P12F65).” “Because I sent her that information all day, she got me a fact-checking app (P12F65).”

No attention at all

7 (41.2%)

“I generally don’t pay attention to this aspect, because the concern isn’t useful (P1M63).” “No, I’ve never seen it before (P17M74).”

Have read the content but not actively follow

9(53.0%)

“No attention, I just usually flip to short-form videos, then will look (P6F65).” “I don’t deliberately see, but if I browse it I will read (P9F65).”

4 Discussion This study used semi-structured interviews to understand older adults’ attitudes toward the authenticity of short-form videos. Attempted to uncover directions of interventions from older participants’ existing identification situations. In this section, first, we

230

W. Hu et al.

describe a model of the relationship between motives, behaviors, and strategies to discern falsehoods of short-form videos. Second, we discuss intervention on fake short-form videos. Finally, we discuss several limitations of the study. First, we sorted out the storyline of older adults to discern falsehoods of short-form videos. Described why older adults discern falsehoods of short-form videos and how they do so. As shown in Fig. 4. Older participants do not pay attention to the authenticity of each short-form video in their daily watching. That means the identifying act needs a motive to be triggered. On the one hand, motive comes from the participation of older participants in current short-form videos. And users’ thematic interest in news content affects engagement with that news [16]. Therefore, the content of short-form videos can affect older participants’ behavior to identify. Another motive is sharing behavior of older participants. In the interview, 77.8% of old participants would share short-form videos with family or friends. They will screen out fake short-form videos. That may be because the people they were sharing with are familiar, which lead older participants to believe that they should be responsible for the information. Therefore, in the interviews, they constantly emphasized that they would only share short-form videos which are “meaningful”, “reliable”, and “no problem”.

Fig. 4. The relationship among the motives, behaviors, and strategies of the older participants to discern falsehoods of short-form videos.

Second, interfering with falsehoods in short-form videos can be focused on technology [17, 18] or person [14, 19]. Based on the content of the interview, this study puts forward two intervention measures from the interaction process between person and short-form videos. Timing of Interventions. Not necessarily covering all content areas, but only intervening in content areas that older adults are interested in. A large reason why some users will share false content is that they are not aware the information may be false. Therefore, making older adults aware of the falsehood of short-form videos is the first step. It has been shown that trust in false news can be reduced if people are informed that the information was maybe inaccurate in advance [20]. Therefore, on the one hand, it can be combined with the content recommendation system to prompt content that older adults

I Know How but I Do not Want to Discern Falsehoods

231

recently paid attention to. On the other hand, prompts can be provided on the sharing interface of older adults. The Prompt Content. Some evidence can be included when alerting older adults to the authenticity of short-form videos. According to the identification strategies of older participants in interview, it can be prompted “The short-form video is published by an unofficial account, please pay attention to screening falsehoods”. Or prompt based on external clues from short-form videos. For example, “The short-form video was published in March 2019, please pay attention to screening falsehoods”. Similarly, giving older adults cues based on their discrimination strategies to get their attention. Third, although older participants knew how to discern falsehoods of short-form videos. But in practice, none of them used the strategies they report to discern. For example, the act of clicking on the comment section or entering the publisher’s homepage. And older participants did not discern the falsehoods in the short-form video. There may be two reasons for this phenomenon. First, the content of the short-form video is not what they are currently concerned about. Because in the results section, we mentioned that one of older participants’ discriminative motives was “related to selfinterest”. Second, older participants used the researcher’s phone to watch, not their own when tested. Therefore, it was difficult to produce an intention to share at the moment of watching. The intention to share is another motive for older participants to discern falsehoods of short-form videos. It may be lack of motive caused older adults not to discern the falsehoods. This research is limited in several ways. First, the sample size of the interview is relatively small. Further experiments are needed to explore the findings of interviews. Second, all the old participants are from the urban areas of Chongqing, China. Lack of knowledge about the situation of older people in rural areas.

5 Conclusion This study investigated 18 older participants’ behaviors to discern the falsehoods of shortform videos. In terms of motive, this study found that short-form videos are related to their interests or involve the behavior of sharing, which would encourage them to discern the falsehoods actively. And research found four strategies for identifying fake information of short-form videos: personal or life experience, publishing channels, reasoning by external clues of short-form videos, and through the help of a third party. Furthermore, in the short-form video test section, researcher found that older participants did not discern the falsehoods as they reported. Acknowledgment. The authors would like to acknowledge the support from National Natural Science Foundation of China (72171030).

References 1. MobTech: 2021 China Silver Hair Economy Insight Report. https://www.mob.com/mobdata/ report/142. Accessed 12 Dec 2022

232

W. Hu et al.

2. Xiang, H., Zhou, J., Wang, Z.: Reducing Younger and older adults’ engagement with COVID19 misinformation: the effects of accuracy nudge and exogenous cues. Int. J. Hum.–Comput. Interact. 1–16 (2023) 3. Cao, J., Qi, P., Sheng, Q., Yang, T., Guo, J., Li, J.: Exploring the role of visual content in fake news detection. In: Shu, K., Wang, S., Lee, D., Liu, H. (eds.) Disinformation, Misinformation, and Fake News in Social Media. LNSN, pp. 141–161. Springer, Cham (2020). https://doi.org/ 10.1007/978-3-030-42699-6_8 4. Appel, M., Prietzel, F.: The detection of political deepfakes. J. Comput.-Mediat. Commun. 27(4), zmac008 (2022) 5. Khodabakhsh, A., Ramachandra, R., Busch, C.: Subjective evaluation of media consumer vulnerability to fake audiovisual content. In: 2019 Eleventh International Conference on Quality of Multimedia Experience (QoMEX), Berlin, Germany, pp. 1–6. IEEE (2019) 6. Newman, E.J., Garry, M., Unkelbach, C., Bernstein, D.M., Lindsay, D.S., Nash, R.A.: Truthiness and falsiness of trivia claims depend on judgmental contexts. J. Exp. Psychol. Learn. Mem. Cogn. 41(5), 1337–1348 (2015) 7. Nightingale, S.J., Wade, K.A., Watson, D.G.: Can people identify original and manipulated photos of real-world scenes? Cogn. Res. Principles Implications 2(1), 1–21 (2017). https:// doi.org/10.1186/s41235-017-0067-2 8. Stenberg, G.: Conceptual and perceptual factors in the picture superiority effect. Eur. J. Cogn. Psychol. 18(6), 813–847 (2006) 9. Nightingale, S.J., Wade, K.A., Watson, D.G.: Investigating age-related differences in ability to distinguish between original and manipulated images. Psychol. Aging 37(3), 326–337 (2022) 10. Chinanews: After the exposure of “fake Jin Dong” cheating middle-aged and older fans, they haven’t stopped yet ...... (2020). https://www.chinanews.com.cn/yl/2020/10-15/9313057. shtml. Accessed 12 Jan 2023 11. Pan, Y.: Accurately regulate the use of short-form video platforms to carry out fraudulent behavior. Procuratorate Daily 03 (2022). http://newspaper.jcrb.com/2022/20221020/202 21020_003/20221020_003_3.htm. Accessed 30 Jan 2023 12. Brashier, N.M., Schacter, D.L.: Aging in an era of fake news. Curr. Dir. Psychol. Sci. 29(3), 316–323 (2020) 13. Moore, R.C., Hancock, J.T.: A digital media literacy intervention for older adults improves resilience to fake news. Sci. Rep. 12(1), 1–9 (2022) 14. Tahir, R., et al.: Seeing is believing: exploring perceptual differences in deepfake videos. In: Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (CHI ‘21), Article 174, pp. 1–16. Association for Computing Machinery, New York, USA (2021) 15. Corbin, J., Strauss, A.: Basics of Qualitative Research, 4th edn. SAGE Publications Inc., Upper Saddle River (2015) 16. Karnowski, V., Kümpel, A.S., Leonhard, L., Leiner, D.J.: From incidental news exposure to news engagement. How perceptions of the news post and news usage patterns influence engagement with news articles encountered on Facebook. Comput. Hum. Behav. 76, 42–50 (2017) 17. Khodabakhsh, A., Busch, C., Ramachandra, R.: A taxonomy of audiovisual fake multimedia content creation technology. In: 2018 IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), pp. 372–377. IEEE (2018) 18. Tolosana, R., Vera-Rodriguez, R., Fierrez, J., Morales, A., Ortega-Garcia, J.: Deepfakes and beyond: a survey of face manipulation and fake detection. Inf. Fusion 64, 131–148 (2020) 19. Li, X., Li, S., Li, J., Yao, J., Xiao, X.: Detection of fake-video uploaders on social media using Naive Bayesian model with social cues. Sci. Rep. 11(1), 1–11 (2021) 20. Clayton, K., et al.: Real solutions for fake news? Measuring the effectiveness of general warnings and fact-check tags in reducing belief in false stories on social media. Polit. Behav. 42(4), 1073–1095 (2020)

Using Media Literacy to Fight Digital Fake News in Later Life: A Mission Impossible? Eugène Loos1(B) and Loredana Ivan2 1 Utrecht University School of Governance, Utrecht University, Utrecht, The Netherlands

[email protected]

2 Communication Department, National University of Political Studies and Public

Administration, Bucharest, Romania [email protected]

Abstract. Fake news is a threat to the trustworthiness of digital information sources. Media literacy training that can be used to empower people to fight fake news - understood to refer to any kind of misleading information that could mistakenly be considered accurate, regardless of the mechanisms that led to its propagation [1] - is mostly oriented toward younger people in an educational institutional context. Yet in later life, older people, too, may benefit from such training. Are there appropriate institutions to provide this? In this paper we explore the possibility of enhancing older people’s digital resilience [2] using insights from the field of media literacy to answer the following research questions: 1. To what extent are older people vulnerable to fake news? 2. To what extent are older people able to learn to become digitally resilient? 3. What institutions could play a role in providing media literacy training specifically tailored for older people? Finally, we will present some implications for future research in this field. Keywords: fake news · digital resilience · media literacy training · generations · younger people · older people

1 Introduction In our society, access to digital information is of prime importance. De Jong and Rizvi [3], in The State of Access: Success and Failure of Democracies to Create Equal Opportunities, argue that democracies are judged by whether citizens have equal access to public services (a primary good [4]), economic opportunities, justice and participation in the democratic process. Anno 2023, we can add citizens’ access to reliable digital information (https://education.ec.europa.eu/focus-topics/digital-education/action-plan/ action-7). Bovens (2002) [5] and Bovens and Loos (2002) [6] even suggest that the equal right of access to information should be considered a basic right of all citizens, on a par with the classic (human) rights. In this media landscape, fake news (“any kind of misleading information that could mistakenly be considered accurate, regardless of the mechanisms that led to its propagation” [1, p. 204] plays an increasingly important role, particularly with the rise of © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 233–247, 2023. https://doi.org/10.1007/978-3-031-34866-2_18

234

E. Loos and L. Ivan

social media. “Fake news is not a new phenomenon ([7, 8] because the partisan press has always peddled biased opinions and stories lacking factual basis [8]. New technologies, from the telegraph in the 19th century to contemporary social media algorithms, have led to the proliferation of fake news [7].” [9, p. 147]. In their empirical study ‘The spread of true and false news online!’, Vosoughi et al. (2018) [10] analyzed the differential diffusion of all of the verified true and false news stories distributed on Twitter from 2006 to 2017 (~126,000 stories tweeted by ~3 million people more than 4.5 million times). They classified news as true or false using information from six independent fact-checking organizations that exhibited 95 to 98% agreement on the classifications. They concluded: “Falsehood diffused significantly farther, faster, deeper, and more broadly than the truth in all categories of information, and the effects were more pronounced for false political news than for false news about terrorism, natural disasters, science, urban legends, or financial information.” [10, p. 1146]. Hence, enabling citizens to build digital resilience [2] is of the utmost importance. To achieve this, a media literacy approach can be applied: “Media literacy should not only focus on people’s ability to use certain devices and technologies, but also on promoting a deep understanding of modern forms of media, how these work and how they produce and use news items, all of which may be attained through systematic media education programs [11]. It is of course important to investigate the feasibility of interventions at an early age to empower young citizens such that they are able to establish the trustworthiness of news.” [1, p. 293] The European Commission therefore “encourages fact-checkers and civil society organizations to provide educational material to schools and educators and include targeted initiatives on disinformation online in the #SaferInternet4EU Campaign” (https://ec.europa.eu/digital-single-market/en/tackling-onlinedisinformation). It is also essential, though, not to focus solely on young citizens; we should also involve older generations as “due to the paucity of studies in this field, it would be naive to assume that they are not vulnerable to fake news” [1, p. 293]. We therefore advocate the implementation of educational measures to tackle the consumption of fake news, such as promoted by the European Commission’s flagship Digital Education Action Plan (2021–2027) (https://education.ec.europa.eu/focus-top ics/digital-education/action-plan/action-7) that presents guidelines providing support for teachers and educators in addressing the pressing topics of disinformation and digital literacy in the classroom, see also [1, 12]. Primary school children have difficulties assessing the reliability of digital information [12–19]. For this reason, media literacy programs in primary schools offer guidance on how to deal with fake news (see for example: https://education.ec.europa.eu/news/ guidelines-for-teachers-and-educators-on-tackling-disinformation-and-promotingdigital-literacy-through-education-and-training). Secondary school pupils also need and get - training in this field (e.g., [20–25]). And the following studies clearly show that university students also have difficulties assessing the reliability of digital information [26–34], but extensive media literacy programs to fight fake news of the kind developed for primary and secondary school pupils, are not available for this group. In Sect. 2.1, the results of the above-mentioned empirical studies related to younger people’s vulnerability to fake news will be discussed, and in Sect. 2.2 the results of empirical studies related to older people and fake news will be presented.

Using Media Literacy to Fight Digital Fake News in Later Life

235

Building digital resilience [2] is not only crucial for younger, but probably also for older people. In this paper, we therefore explore the following research questions: 1. To what extent are older people vulnerable to fake news? 2. To what extent are older people able to learn to become digitally resilient? 3. What institutions can play a role in providing media literacy training specifically tailored for older people?

2 Fake News: A Generational Approach 2.1 Younger People’s Vulnerability to Fake News Empirical studies such as [12–34] clearly show that younger people are vulnerable to fake news. We will briefly discuss the results of several studies that looked at various different age groups. These show that younger people’s abilities to detect and avoid being exposed to fake news, from primary school children to more cognitive complex individuals such as university students, were rather poor. Primary School Children In 2017, inspired by a study conducted 10 years previously in the U.S. [13], we ran a study in the Netherlands to examine whether school children, 11 to 12 years of age, were able to recognize a website (Save The Pacific Northwest Tree Octopus, http://zap atopi.net/treeoctopus/) as a fake [12]. The results of this Dutch study were intriguing, as they showed an even higher vulnerability compared to the U.S.-study among 13year-old school pupils conducted 10 years before. Only 2 out of the 27 school children participating in the Dutch study identified the website as being a hoax, compared to 6 out of 53 in the US study. Emotional involvement may have played a role (the topic was an animal in danger), making it more difficult for them to perceive the information as fake. Possibly, too, the fact that the information had been endorsed by their teacher made them less critical about the trustworthiness of the information. The results were also disturbing because it might have been expected that in the ten-year period between the two studies, people (including primary school children) would have become more and more familiar with the idea that some of the information obtained through the use of digital technologies might well be fake. A recent study conducted in Brazil [14] with two groups of seventh-year children at a private bilingual school used the research design developed by [12] revealed that none of the 40 school pupils could identify the site as being a hoax. Along those same lines, in 2020, Dumitru [15] conducted a study among Romanian primary school children (26 girls and 28 boys) in the local language featuring a website about a fake animal (https://salvamjacalopul.wordpress.com/). In this study, a group of primary school children (aged 10–11) was compared with a similar group of high school pupils (aged 18–19). The results in both groups were similar, with neither group demonstrating good capabilities of identifying the website as a hoax. Although the adolescent group was more inquisitive, showing a more complex way of treating information compared to the primary school children, this group, too, was found to lack the digital resistance needed to recognize fake information. Disturbingly, when confronted with a fake website, even the school pupils exhibiting more critical thinking styles in

236

E. Loos and L. Ivan

both the primary school group and the secondary school proved to be susceptible to fake news. Similar results were found in another recent study [16], conducted in the U.S., in which 86 primary school children were respectively asked to judge the veracity of ten news stories, five fake and five true: not only were they unable to identify the fake information, their performance did not vary by age or by their scores on a cognitive reflective text. The study by Dumitru [15] enriched the findings by adding group discussion to understand how the school pupils approached the information from another hoax website. The results showed that some, mainly those in the primary school group compared to the high-school pupils, regarded the information as being trustworthy and did not bother to check, even though they had Internet available. Others doubted the existence of the animal, but were prepared to share the information, as it was “cute”. For some school pupils, the fact that “an expert” – in this case the researcher – had introduced the website counted as a guarantee of the accuracy of the information. Another study [17] conducted in the U.S. also used the fake website of Save The Pacific Northwest Tree Octopus to test the reactions of primary school children. The study included 354 primary school children, ranging from first graders to fifth graders. This study found that the school pupils in the 5th grade were indeed more critical about the website and were less prone to trust the information (42%), but no linear effect of age could be established; for example, 80% of those in the 4th grade considered the website to be trustworthy, compared to 50% of those in the 2nd grade). Instead of age, other contextual aspects seemed more important in determining whether these primary school children judged the information to be reliable or not, namely, the strategies they used to determine the credibility of the website, such as prior knowledge of the subject, a check of the text features and general knowledge (factual knowledge). In that same vein, study [18] underlined the role of educators and media literacy in allowing children from primary school to question fake information and reduce their vulnerability. Study [19] recently explored media literacy strategies using a meta-analysis of sixteen research studies. The goal was to develop news literacy or media education in primary schools. The analysis showed that a strategy where primary school children are taught how to create their own news using text messages as well as videos, audios, pictures and animations was the most effective in equipping them with the proper skills to recognize fake news. To conclude, primary school children are vulnerable to fake news, unless they are familiar with the content (because they have studied a particular subject or have a wealth of general knowledge). Even if they doubt the trustworthiness of the information, they will still tend to share or act upon it, especially if the information is “cute” [15] and worth sharing with peers; or an authority like their teacher or a researcher has endorsed it. Emotional involvement also plays a role; for example, the fact that the information concerned an animal in danger made it more difficult for them to perceive the information as fake [12]. Their ability to fight fake news does not necessarily increase with age or with the development of reflexive skills. In primary school children, therefore, an effective strategy to promote media literacy and to address fake news vulnerability could be to

Using Media Literacy to Fight Digital Fake News in Later Life

237

have them practice by creating news items of their own, using multi-modal elements such as text messages, as well as video, audio, pictures and animations [19]. Secondary School Pupils Other research also looked at the vulnerability of adolescents between the ages of 14 and 18 to fake news. A report released in 2016 by the Stanford History Education Group presented the results of a study [20, 21] examining the capabilities of pupils in middle school through college of judging the credibility of information. Several online tasks were administered to 7,804 school pupils, asking them to evaluate information found on the internet, and particularly on social media sites. The main conclusion regarding their social media use was: “Our “digital natives” may be able to flit between Facebook and Twitter while simultaneously uploading a selfie to Instagram and texting a friend. But when it comes to evaluating information that flows to social media channels, they are easily duped.” [20, p. 4]. Dumitru [15], in the more recent study referred to above, found that secondary school pupils (last years of high school, 18–19 years of age) reacted to a fake website in almost the same way as primary school children. Although teenagers were more skeptical about the fake information, their skepticism did not help them in “taking action” and they did not check the veracity of the information. Marchi (2012) [22] used individual interviews and focus groups, recruiting 61 participants from U.S. high schools between the ages of 14 and 19, to explore how teenagers viewed news, and found that “teens gravitate toward fake news, “snarky” talk radio, and opinionated current events shows more than official news, and do so not because they are disinterested in news, but because these kinds of sites often offer more substantive discussions of the news and its implications.” (p. 257). In the case of secondary school pupils, interactive strategies, such as the use of games, proved to be more effective in increasing their abilities to distinguish between real and fake information [23, 24] (see also https://inoculation.science/inoculation-games/). Another study [25] focused on more sophisticated strategies of fact checking, such as footing, taking bearings and lateral reading for high school pupils. Given by experts or by their teachers, this type of training proved to be effective, leading researchers to advocate their inclusion in the curriculum. Although pre-test and post-test experiments conducted in the studies described above demonstrated the efficiency of teaching secondary school pupils fact-checking techniques, it is unclear whether adolescents will implement these strategies in their online interactions in the long run after the literacy training. University Students Studies involving adult students (mainly undergraduates) have included a number of variables to understand their vulnerability to fake news, starting from the assumption that there is a combination of task characteristics, individual aspects and media literacy differences that could be considered. For example, a study on undergraduate students’ reactions to information about climate change [26] showed that when students were asked to summarize the information and construct arguments starting from the information, they were more able to distinguish between the reliable and less reliable pieces of information. Also, the effect was moderated by the reader’s prior beliefs, proving

238

E. Loos and L. Ivan

that individual aspects (attitudes, beliefs) always play a role in the success of particular cognitive strategies. This particular study showed that having an overview of a particular subject without allowing people to summarize or formulate their own arguments, made them particularly vulnerable to fake news. Academic education should therefore stimulate students not only to access overviews on different topics, but also to use their own mental capacities to react (with arguments) to different issues. One important study conducted by Hargittai et al. (2010) [27] clarified the process by which people invest trust in online sources. It involved a relatively large number of students (N = 102) and deployed both quantitative and qualitative inquiry. Researchers asked students to rate different criteria of trust and gave them the opportunity to talk openly about how they made their decision to trust certain online sources and distrust others. Briefly, the results showed that some search engines (for example, google.com) are automatically invested with trust by students based on their familiarity with these; furthermore, the brand names of different online sources are automatically associated with specific positive or negative perceptions, while asking others also appeared to be a common strategy for students when deciding what information to trust. In addition, students’ success in investing trust in reliable online sources also depended on their digital skills (digital experience and expertise). This particular study points out the process by which younger people might become vulnerable to fake online information and indicates the fact that people’s perceptions of different online sources (and brands), as well as trusting others, are important factors in the equation, often ignored in previous studies. Many studies exploring students’ vulnerability to fake news have actually used a deterministic approach: tasks were assigned to students, and experts (researchers, teachers, trainers) then evaluated their (in) ability to distinguish between what is accurate and what is fake; what is credible and what lacks credibility. In some of these studies students proved to be as vulnerable as the other (younger) age groups. For example, a recent study [28] using the same Save The Pacific Northwest Tree Octopus fake website, explored first year students’ ability to question the existence of the Fake Tree Octopus, as a part of a biology class. Similar reactions to those in the studies on that same website among primary school children (as described above) were seen, with many students not questioning the existence of the species described there. Also, students failed to recognize a real species when it was presented in a satirical way (using a YouTube video), when this was part of a biology lesson. A Spanish study [29, p. 407] focusing on university students in Spain (Andalusia) concluded the following: “In order to ascertain the degree of credibility that young users in Andalucía give to information, this study presents the results of the evaluation of online news by university students pursuing degrees in communication and education (N = 188), using the CRAAP test. The data reveal differences in gender and degree program in the credibility assigned to the news. The conclusion is that university students have difficulty differentiating the veracity of the sources, in line with previous studies, with fake news earning higher ratings than real news.” Such surprising results [see also 30–34] could indicate a high vulnerability to fake information in relation to how the information is “framed”: while a website presenting information in a structured way (similar to what students are used to being presented with

Using Media Literacy to Fight Digital Fake News in Later Life

239

in classes) might a priori be perceived as “credible”, presenting valuable information in a humoristic/satirical way could create the opposite effect, discrediting the information. To explore such an approach, a more phenomenological view, such as that deployed in the study conducted by Hargittai et al. (2010) [27], is needed, starting from the process by which people navigate through online content and invest trust in different online sources. However, strategies to reduce university students’ vulnerability to fake news follow the same determinist path as described above in the case of primary and secondary school pupils: interventions in which students are exposed to “what needs to be done” when confronted with the need to assess the credibility of an online information source (e.g., website). In such situations, the efficacy of the training/intervention is judged by a pretest-and-post-test experimental design in which participants’ abilities to judge the online information are evaluated prior and post intervention [31, 32]. To assess the longtime effect of the intervention, students were tested some weeks later [32]. The results were encouraging, showing that the effects of the intervention remained stable over time. Besides different fact-checking techniques [31] - of the kind used to train high school pupils - observational learning and feedback have also been tested in university students, with positive results [33]. Literacy Training for Younger People This section has shown that younger people, in primary, secondary and university education are vulnerable to fake news, regardless of their age, and that media literacy training tools, including the use of games and fact-checking techniques, could certainly play a role in enhancing their capabilities in this field. Though the short term effect looks promising in reducing school pupils’ vulnerability to fake news, the studies we discussed did not pay attention to the long term impact of media literacy interventions on the capabilities to deal with fake news Also, the role of cultural differences is still under-researched in this age group. It is also noteworthy that there are no studies that compared the effectiveness of different media literacy strategies in fighting school pupils’ vulnerability to fake news. 2.2 To What Extent are Older People Vulnerable to Fake News? It may be observed that the European Commission’s flagship Digital Education Action Plan (2021–2027), to which we referred in our Introduction, presents guidelines for teachers and educators on tackling disinformation and promoting digital literacy through education and training, with the aim of ensuring that younger people (our emphasis) are equipped with the skills and competences to live and thrive in the digital age. This observation is confirmed by Dumitru et al. (2022) [1] in their overview of publications related to media literacy training and interventions and the extent to which these are evidence based from a generational perspective, who concluded: “We found that students and educators were the main target groups, almost wholly to the exclusion of other groups; that they took place mainly in educational settings; and that, at least in the case of the training sessions, they were not evidence based, which meant that neither the long-term nor short-term efficacy could be tested.” (p. 291). As far as we know, there are only a few studies focusing on the extent to which specifically older people are vulnerable to fake news. Though the special issue “Fighting

240

E. Loos and L. Ivan

Fake News: A Generational Approach”, in 2022 published by the journal Societies (https://www.mdpi.com/journal/societies/special_issues/fake_news) [35] examines the role of age in fake news consumption, the focus is on the young to adolescent age group. Empirical studies comparing younger and older people’s vulnerability to fake news are rare. Loos and Nijenhuis (2020) [36, p. 69] examined the way older people, compared to younger age groups, are at risk of consuming fake news: “Social media are increasingly being used by young and old as a source of information.” They carried out an empirical study between the beginning of February 2018 and the end of June of 2018, disseminating 14 political fake news articles (e.g., relating to Brexit and Donald Trump) in the form of advertisements on Facebook and tracking user interaction with the fake content to analyze the number of users in the age groups 13–17, 18–24, 25–34, 35–44, 45–54, 55–64, 65+. Their results show that “the articles had a higher reach amongst the older age groups, as well as that many people likely took the headlines at face value without clicking on the link. The number of emotional responses posted by the pro-Brexit and pro-Trump groups was greater than those posted by the pro-remain and anti-Trump groups.” Michael and Sanson (2021 [37]) conducted two preregistered experiments in the U.S. (with data for the first collected between 31 January - 21 April 2018, and for the second on 2 May 2019) and found that “adults across a range of ages rely on information other than news content—such as how they feel about its source—when judging whether news is real or fake. Moreover, our findings help explain how people experiencing the same news content can arrive at vastly different conclusions.” In other words, they argue that vulnerability to the consumption of fake news occurs in all age groups. Moore and Hancock (2022) [38] state: “Recent research has identified older adults as a demographic group especially susceptible to fake news online. For example, during the 2016 U.S. presidential campaign, people 65 and older were twice as likely to be exposed to fake news on Twitter and seven times more likely to share fake news on Facebook than 18–29-year-olds [39, 40]”. In their analysis of a nationally representative sample of mobile, desktop, and television media consumption over a period of nearly three years, Allen et al. (2020) found that older individuals were substantially greater consumers of fake news than younger people [41]. Oxford Analytica (2020) [42] found that “The growing numbers of senior citizens in the US, their rapidly increasing adoption of social media and their high levels of voter turnout make their vulnerability to disinformation a matter of special concern. Other advanced democracies likely mirror the US experience.” We can therefore conclude that there are only a limited number of empirical studies examining older people’s vulnerability to fake news [35–43]. The studies that are available do indeed indicate that older people consume fake news. Although we do not know if they are more vulnerable to fake news than younger people, at least a certain degree of susceptibility may be assumed. To gain insight into ways to address older people’s fake news consumption, we will now explore how a media literacy approach can be used to reach this goal in Sects. 3 and 4.

Using Media Literacy to Fight Digital Fake News in Later Life

241

3 To What Extent are Older People Able to Learn to Become Digitally Resilient? 3.1 Aged Heterogeneity Moore and Hancock (2022) [38] state: “Scholars have postulated that older adults’ limited digital literacy may explain their heightened susceptibility to fake news online [40].” We take issue with this generalization, as it does not consider the diversity occurring within the group of older adults. Before discussing in Sect. 3.2 the extent to which older people are able to become digitally resilient [2], we must first take a closer look at the phenomenon of aged heterogeneity. In a classic paper, Nelson and Dannefer (1992) [44] reviewed 185 gerontological studies to analyze the individual differences and empirical patterns of variability reported in these studies presenting measures of dispersion. They concluded that “Overall, a majority of all gerontological studies presenting data reported increases in variability with increasing age (65%). (…). The dominance of the pattern of increasing diversity does not appear to be domain-specific; the same general finding pattern emerged across physical, personality, and cognitive domains.” (p. 17). Twenty-four years later Stone et al. (2016) [45] analyzed 2,307 gerontological papers during a 6-year window (2005– 2010) and concluded: “Turning to the question of patterns of variability observed in those studies that report measures of variability, we have seen that the great majority of studies report either stability or increasing variability with age. However, the pattern varies substantially among outcome types. Half of the biological studies reported increasing variability, but only about a third of the psychological studies did so.” (p. 4) See also [46–50] that confirm the occurrence of the aged heterogeneity phenomenon related to internet use by older people. 3.2 Using Tools to Fight Fake News in Later Life: Never Too Old to Learn It is a myth that older people are not able to learn how to use digital information in their everyday life. Experience [46, 47, 49] and motivation [46] enhance the digital capabilities of older people considerably. And applied to the field of literacy, Grace and Hone (2019) [51] present the following interesting example of a game designed to serve as new literacy education tool: “The game underwent two primary designer iterations. As a result of design changes and renewed political chatter about fake news, the game’s second iteration gathered more than 500,000 plays. The data collected reveals useful patterns in understanding news literacy and the perception of play experiences. These data, of more than 45,000 players, indicates that the older the persons the better they are at identifying fake news, until the approximate age of 70. It also indicates that higher education correlates to better performance at identifying real news from fake, although the time it takes to do so varies. This case study demonstrates the potential for such game designs to collect data useful to non-game contexts.” (p. 8). To enhance the digital resilience [2] of older adults, it is important to take into account aged heterogeneity as discussed above [44–50] as well as to consider their digital capabilities on a digital spectrum (Lenhart and Horrigan [52]). See also Van

242

E. Loos and L. Ivan

Kampen et al. (2023) [53] who state that it is never too late to learn for the group they call “third-agers” as they experience little (self-) ageism or barriers to learning. We can conclude that as older people are capable of learning digital competences, media literacy training and interventions can be used to enhance their digital resilience [2] and ability to fight fake news. This view is also supported by Rasi et al. (2021) [54, p. 37] who conducted a systematic review of 40 empirical studies published between January 2005 and April 2019, focusing on the promotion of media literacy among older people. However, they also noted that “interventions aimed at fostering media literacy in older people need further development and creative enrichment in terms of aims, content, providers, recipients, and pedagogical approaches.” (p. 1).

4 What Institutions Could Play a Role in Providing Media Literacy Training Specifically Tailored for Older People? In the above, we pointed out that the European Commission’s flagship Digital Education Action Plan (2021–2027) presents guidelines for teachers and educators, with the aim of ensuring that younger people (our emphasis) are equipped with the skills and competences they need to live and thrive in the digital age. This observation is further borne out by the overview of publications related to media literacy training and interventions from a generational perspective, and the extent to which they are evidence based, presented by Dumitru et al. (2022) [1]. On the one hand, it would seem natural to start teaching media literacy to fight fake news at a young age, enabling the benefits. From acquiring these competencies early on to be reaped throughout an entire life span. As Loos et al. (2018) [12, p. 524] state: “As even college-age students have limited new literacies capabilities, it is important to start early and to educate young children on how to critically evaluate online information.” On the other hand, it would not be fair to focus solely on the young, leaving older people without media literacy capabilities vulnerable to fake news. Training opportunities must be created for the older population as well. It is a human right [55] to improve their digital and online abilities and boost their digital resilience [2] by enabling them to acquire media literacy. We argued in Sect. 2 that, although older people are indeed vulnerable to fake news, they are also well able to learn how to become digitally resilient, as was shown in Sect. 3. The question which then arises is how to enable them to acquire the media literacy capabilities needed to identify and establish whether or not information can be trusted. As older people have long since left the school system, it is important to evaluate which institutions would be suitable for teaching this group the media literacy skills needed in order to deal with fake news in their everyday life. The University of the Third Age (U3A) (https://www.u3a.org.uk/) in the UK, the Lifelong Learning Institutes in the US (https://en.wikipedia.org/wiki/Lifelong_learning_institutes) and the Université du Troisième Age in French speaking countries, are good examples of such institutions. The same goes for Community centers, and Organizations for senior citizens (e.g., AARP, https://www.aarp.org/, media literacy institutions (e.g., https://www.med iawijsheid.nl/).

Using Media Literacy to Fight Digital Fake News in Later Life

243

Then there are the institutions offering older people help with developing their digital skills. Examples include Senior web in Switzerland and the Netherlands (e.g., https://seniorweb.ch/; https://www.seniorweb.nl/). Instead of solely focusing on technical skills and safe internet use for older people, such an institution could additionally offer a media literacy course. A good example of an online course, specifically developed for older people (related to the 2020 U.S. election, https://www.poynter.org/shop/fact-checking/mediawise-for-seniors-hands-onlessons-on-separating-fact-and-fiction-online/ and https://www.poynter.org/shop/factchecking/how-to-spot-misinformation-online-july-2021/) is Media Wise for Seniors in the US. Moore and Hancock (2022) [38] evaluated this online course by conducting a 1-h intervention, composed of self-directed series of interactive modules designed to teach concepts and skills for identifying misinformation online, and concluded that “consistent with our pre-registered hypothesis, older adults (M age = 67) in the treatment condition (N = 143) significantly improved their likelihood of accurately discerning fake from true news from 64% pre-intervention to 85% post-intervention. In contrast, older adults in the control condition (N = 238) did not significantly improve (from 55% to 57%). The treated older adults were also more likely to employ strategies for identifying misinformation online compared to pre-intervention and the control group.” Libraries can play an important role in teaching media literacy to older people by giving in-person and online courses (e.g., using an interactive approach or video tutorials), see also [56, 57]. Finally, intergenerational knowledge transfer from children who have received media literacy training at school to their (grand) parents could also be an interesting option. In that way the children would act as so-called warm experts [58, 59].

5 Conclusions 1. To what extent are older people vulnerable to fake news? A range of empirical studies show that younger people, regardless of their age, in primary, secondary and university education are vulnerable to fake news, and media literacy training tools, including the use of games and fact-checking techniques, could certainly play a role in enhancing their capabilities in this field. But there is only a limited number of empirical studies on older people’s vulnerability to fake news [35–43]. The studies that are available show that older people do, indeed, consume fake news. Although we do not know if they are more vulnerable to fake news than younger people, they may be assumed to be vulnerable, at least to a certain extent. 2. To what extent are older people able to learn to become digitally resilient? Older people are capable of learning digital competences. Media literacy training and interventions can be used to enhance their digital resilience [2] and enable them to become more adept at detecting fake news [1, 38]. To bolster the digital resilience of older people, it is important to take into account aged heterogeneity [44–50] and to evaluate their digital capabilities on a digital spectrum (Lenhart and Horrigan [52]. See

244

E. Loos and L. Ivan

also Van Kampen et al. (2023) [53], who stated that it is never too late for “third-agers” to learn, as they experience little (self-) ageism or barriers to learning. 3. What institutions could play a role in providing media literacy training specifically tailored for older people? As older people have long since left the school system, it is important to consider other institutions that would be suitable for teaching them the media literacy skills needed in order to deal with fake news in their everyday life. The following institutions could play a role: • The University of the Third Age (U3A) (https://www.u3a.org.uk/) in the UK. • The Lifelong Learning Institutes in the US (https://en.wikipedia.org/wiki/Lifelong_ learning_institutes). • The Université du Troisième Age in French speaking countries • Community centers, Organizations for senior citizens (e.g., AARP, https://www.aarp. org/). • Media literacy institutions (e.g., https://www.mediawijsheid.nl/). • Organizations offering older people help in developing their digital skills, such as senior web in Switzerland and the Netherlands (e.g., https://seniorweb.ch/; https:// www.seniorweb.nl/). • Libraries can play an important role in teaching media literacy to older people by giving in-person and online courses (e.g., using an interactive approach or video tutorials), see [56, 57]. • Intergenerational knowledge transfer from children, who have received media literacy training at school, to their (grand) parents could also be an interesting option. In that way, the children would act as so-called warm experts [58, 59].

6 Implications for Future Research To gain more insight into the dynamics underlying the effectiveness of media literacy tools designed to bolster resilience [2] to fake news in later life, the following four points should be taken into consideration. (1) It would be interesting for future empirical studies to adopt an intersectional approach, linking the role of age to gender, educational level. (2) Such studies should also pay attention to longitudinal impact. (3) Country differences must also be considered. (4) We strongly recommend using an experimental research design that compares the degree to which different media literacy tools are evidence based. Finally, it is important to note that there are no studies comparing the effectiveness of different media literacy strategies in addressing older people’s vulnerability to fake news.

References 1. Dumitru, E.-A., Ivan, L., Loos, E.: A generational approach to fight fake news: in search of effective media literacy training and interventions. In: Gao, Q., J. Zhou, J. (eds.) HCII 2022, LNCS, vol. 13330, pp. 291–310. Springer, Cham (2022). https://doi.org/10.1007/978-3-03105581-2_22

Using Media Literacy to Fight Digital Fake News in Later Life

245

2. Bjola, C., Papadakis, K.: Digital propaganda, counter publics and the disruption of the public sphere: the Finnish approach to building digital resilience. Camb. Rev. Int. Aff. 33(5), 638–666 (2020) 3. De Jong, J., Rizvi, G.: The State of Access: Success and Failure of Democracies to Create Equal Opportunities. Brookings Institution Press, Washington, D.C., USA (2008) 4. Rawls, J.: A Theory of Justice. Belknap Press and Harvard University Press, Cambridge, USA (1971) 5. Bovens, M.A.P.: Information rights. Citizenship in the information society. J. Polit. Philos. 10(3), 317–341 (2002) 6. Bovens, M.A.P., Loos, E.F.: The digital constitutional state: democracy and law in the information society. Inf. Polity. 7(4), 185–197 (2002) 7. Gelfert, A.: Fake news: a definition. Informal Log. 38(1), 84–117 (2018) 8. McGonagle, T.: Fake news: false fears or real concerns? Neth. Q. Hum. Rights 35(4), 203–209 (2017) 9. Obad˘a, R.: Sharing fake news about brands on social media: a new conceptual model based on flow theory argumentum. J. Semin. Discursive Log. Argum. Theory Rhetor. 17(2), 144–166 (2019) 10. Vosoughi, S., Roy, D., Aral, S.: The spread of true and false news online. Science 359(6380), 1146–1151 (2018) 11. Buckingham, D.: The Media Education Manifesto. Polity Press, Cambridge, UK (2019) 12. Loos, E., Ivan, L., Leu, D.: “Save the Pacific Northwest tree octopus”: a hoax revisited. Or: how vulnerable are school children to fake news? Inf. Learn. Sci. 119(9–10), 514–528 (2018) 13. Leu, D.J., et al.: Defining online reading comprehension: using think aloud verbal protocols to refine a preliminary model of Internet reading comprehension processes. In: Alvermann, D. (Chair) 21st Century Literacy: What is It, How do Students Get It, and How do We Know if They have It (2007) 14. Meyrer, K.P.: “Help save the endangered pacific northwest tree octopus from extinction”: are 7th year students susceptible to fake news? Mater thesis. Universidade do Vale do Rio dos Sinos, São Leopoldo, Brazil (2021) 15. Dumitru, E.A.: Testing children and adolescents’ ability to identify fake news: a combined design of quasi-experiment and group discussions. Societies 10(3), 71 (2020) 16. Xu, S., Shtulman, A., Young, A.G.: Can children detect fake news? In: Proceedings of the Annual Meeting of the Cognitive Science Society, vol. 4, no. 44 (2018) 17. Pilgrim, J., Vasinda, S., Bledsoe, C., Martinez, E.: Critical thinking is critical: octopuses, online sources, and reliability reasoning. Read. Teach. 73(1), 85–93 (2019) 18. Housand, B.C.: Fighting Fake News! Grades 4–6: Teaching Critical Thinking and Media Literacy in a Digital Age. Routledge, Milton Park, Abingdon-on-Thames, Oxfordshire, UK (2021) 19. van Helvoort, J., Hermans, M.: Effectiveness of educational approaches to elementary school pupils (11 Or 12 years old) to combat fake news. Media Lit. Acad. Res. 3(2), 38–47 (2020). https://bit.ly/3m2Eo6m 20. SHEG (Stanford History Education Group): Evaluating information: the cornerstone of civil online reasoning. Research Report (2016). www.sheg.tandford.edu 21. Wineburg, S., McGrew, S.: Why students can’t google their way to the truth. Educ. Week 36(11), 22–28 22. Marchi, R.: With Facebook, blogs, and fake news, teens reject journalistic “objectivity.” J. Commun. Inq. 36(3), 246–262 (2012) 23. Chang, Y.K., et al.: News literacy education in a polarized political climate: how games can teach youth to spot misinformation. HKS Misinf. Rev. (2020) 24. Roozenbeek, J., Van Der Linden, S.: The fake news game: actively inoculating against the risk of misinformation. J. Risk Res. 22(5), 570–580 (2019)

246

E. Loos and L. Ivan

25. Wineburg, S., Breakstone, J., McGrew, S., Smith, M.D., Ortega, T.: Lateral reading on the open Internet: a district-wide field study in high school government classes. J. Educ. Psychol. (2022) 26. Bråten, I., Strømsø, H.I.: Effects of task instruction and personal epistemology on the understanding of multiple texts about climate change. Discourse Process. 47(1), 1–31 (2010) 27. Hargittai, E., Fullerton, L., Menchen-Trevino, E., Thomas, K.Y. :Trust online: young adults’ evaluation of web content. Int. J. Commun. 4(27) (2010) 28. Unger, S., Rollins, M.: Do not believe everything about science online: revisiting the fake pacific northwest tree octopus in an introductory biology college course. Sci. Educ. 32(2), 159–163 (2021) 29. Herrero-Diz, P., Conde-Jiménez, J., Tapia-Frade, A., Varona-Aramburu, D.: The credibility of online news: an evaluation of the information by university students/La credibilidad de las noticias en Internet: una evaluación de la información por estudiantes universitarios. Cultura y Educación 31(2), 407–435 (2019) 30. Wineburg, S., McGrew, S.: Evaluating information: the cornerstone of civic online reasoning (2016). https://apo.org.au/node/70888 31. McGrew, S., Ortega, T., Breakstone, J., Wineburg, S.: The challenge that’s bigger than fake news: civic reasoning in a social media environment. Am. Educ. 41(3), 4–9 (2017). https:// files.eric.ed.gov/fulltext/EJ1156387.pdf 32. McGrew, S., Smith, M., Breakstone, J., Ortega, T., Wineburg, S.: Improving university students’ web savvy: an intervention study. Br. J. Educ. Psychol. 89(3), 485–500 (2019) 33. Axelsson, C.-A.W., Guath, M., Nygren, T.: Learning how to separate fake from real news: scalable digital tutorials promoting students’ civic online reasoning. Future Internet 13, 60 (2021) 34. Huy, D.T.N., Nhung, P.T.H., Hue, L.T., Lan, L.: Teaching social sciences students with a case study of publishing fake news and analyzing negative aspects of Vietnam education, State Professor Council and Banking University HCM city Vietnam (2022) 35. Loos, E., Ivan, L.: Special Issue “Fighting fake news: a generational approach”. Societies 12(2), 57 (2022). https://www.mdpi.com/journal/societies/special_issues/fake_news 36. Loos, E., Nijenhuis, J.: Consuming fake news: a matter of age? The perception of political fake news stories in Facebook ads. In: Gao, Q., Zhou, J. (eds.) HCII 2020. LNCS, vol. 12209, pp. 69–88. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-50232-4_6 37. Michael, B., Sanson, M.: Source information affects interpretations of the news across multiple age groups in the United States. Societies 11, 119 (2022) 38. Moore, R.C., Hancock, J.T.: A digital media literacy intervention for older adults improves resilience to fake news. Sci. Rep. 12(1), 1–9 (2022) 39. Grinberg, N., Joseph, K., Friedland, L., Swire-Tompson, B., Lazer, D.: Fake news on Twitter during the 2016 U.S. presidential election. Science 363, 374–378 (2019) 40. Guess, A., Nagler, J., Tucker, J.: Less than you think: prevalence and predictors of fake news dissemination on Facebook. Sci. Adv. 5, eaau4586 (2019) 41. Allen, J., Howland, B., Mobius, M., Rothschild, D., Watts, D.J.: Evaluating the fake news problem at the scale of the information ecosystem. Sci. Adv. 6, eaay3539 (2020) 42. Oxford Analytica. Online disinformation is greater risk for older users. Emerald Expert Briefings, oxan-db (2021). https://www.emerald.com/insight/content/doi/10.1108/OXANDB263856/full/html 43. Brashier, N.M., Schacter, D.L.: Aging in an era of fake news. Curr. Dir. Psychol. Sci. 29, 316–323 (2020) 44. Nelson, E.A., Dannefer, D.: Aged heterogeneity: fact or fiction? The fate of diversity in gerontological research. Gerontologist 32(1), 17–23 (1992) 45. Stone, M.E., Lin, J., Dannefer, D., Kelley-Moore, J.A.: The continued eclipse of heterogeneity in gerontological research. J. Gerontol.: Ser. B 72(1), 162–167 (2016)

Using Media Literacy to Fight Digital Fake News in Later Life

247

46. Loos, E.F.: Senior citizens: digital immigrants in their own country? Observatorio (OBS*) J. 6(1), 1–23 (2012) 47. Loos, E.: In search of information on websites: a question of age? In: Stephanidis, C. (ed.) UAHCI 2011. LNCS, vol. 6766, pp. 196–204. Springer, Heidelberg (2011). https://doi.org/ 10.1007/978-3-642-21663-3_21 48. Hargittai, E., Dobransky, K.: Old dogs, new clicks: digital inequality in skills and uses among older adults. Can. J. Commun. 42(2), 195–212 (2017) 49. Hill, R.L., Dickinson, A., Arnott, J.L., Gregor, P., McIver, L.: Older web users’ eye movements: experience counts. In: Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, pp. 1151–1160 (2011) 50. Hänninen, R., Taipale, S., Luostari, R.: Exploring heterogeneous ICT use among older adults: the warm experts’ perspective. New Media Soc. 23(6), 1584–1601 (2021) 51. Grace, L., Hone, B.: Factitious: large scale computer game to fight fake news and improve news literacy. In: Extended Abstracts of the 2019 CHI Conference on Human Factors in Computing Systems, Glasgow, UK, 4–9 May 2019, pp. 1–8. Association for Computing Machinery, New York, USA (2019) 52. Lenhart, A., Horrigan, J.B.: Re-visualizing the digital divide as a digital spectrum. IT Soc. 1(5), 23–39 (2003) 53. Van Kampen, E., van den Berg, F., Brouwer, J., Keijzer, M.: Never too old to learn -third age adults experience little (self-)ageism or barriers to learning. Soc. Sci. Human. Open 7(1), 100384 (2023) 54. Rasi, P., Vuojärvi, H., Rivinen, S.: Promoting media literacy among older people: a systematic review. Adult Educ. Q. 71(1), 37–54 (2021) 55. Susło, R., Paplicki, M., Dopierała, K., Drobnik, J.: Fostering digital literacy in the elderly as a means to secure their health needs and human rights in the reality of the twenty-first century. Fam. Med. Prim. Care Rev. 3, 271–275 (2018) 56. Herrero-Diz, P., López-Rufino, C.: Libraries fight disinformation: an analysis of online practices to help users’ generations in spotting fake news. Societies 11, 133 (2022) 57. Bonnet, J.L., Herakova, L., Deng, R.: Setting the stage for civic-minded education: casting new roles for librarians in critical information literacy instruction. ACRL. Recasting the narrative. Cleveland, Ohio, USA (2019) 58. Bakardjieva, M.: The Internet in Everyday Life. Sage, Los Angeles, London, Delhi, Singapore, Washington DC (2005) 59. Olsson, T., Viscovi, D.: Warm experts for elderly users: who are they and what do they do? Hum. Technol. 14(3), 324 (2018)

Older Persons Media Usage During a One Month Election Campaign: A Bulgarian Case Study Lilia Raycheva(B)

, Mariyan Tomov , Neli Velinova , and Lora Metanova

The St. Kliment Ohridski Sofia University, Sofia, Bulgaria [email protected] Abstract. Modern families are characterized by heterogeneous media preferences, which are determined by age, gender, social status, cultural identity, etc. In order to achieve a more reliable information awareness of content, hybridity is often observed in media usage. This notion substantiated the aim of the undertaken study to analyze the daily use of media among different members of a Bulgarian family. The study was undertaken within the framework of the MEDIADELCOM research project of the Horizon 2020 program. The methodology is based on semistructured questionnaires in media diaries of 50 students about the media use on a particular day in the life of their families, including data for: age, gender, and educational level of the respondents; purpose, type, and temporality of media usage; media trust, particularly during election campaigns, etc. A one month period of data collection has been set up: October 14 – November 14, 2021. The period coincides with the one-month election campaign in media, related to the regular vote for president and vice president and the early one - for Parliament. In addition, cross-combining technique for data collection and analysis was applied in order to compensate some of the inherent limitations. The information obtained is summarized in order to be analytically compared with the results of the other 13 EU participating countries in MEDIADELCOM project. Keywords: Media Diary · Family · Media Usage · Trust in Media

1 Introduction Modern societies are characterized by hybrid media context [1] and ubiquity of the media in the social sphere [2]. The digital environment poses significant challenges to media studies. Researchers are looking for innovative ways to measure the role, which media are playing in people’s lives [3–8]. In analyses of the audience’s behavior certain difficulties arise, based on the variety of platforms that users can integrate in their media menu, as well as on the function attributed to the media content they consume. The limitations of the most common quantitative (measurable data) and qualitative (interpretation of specialized texts, interviews, and observations) methods can be supplemented by developing advanced approaches in order to improve in-depth the reasoning of the groundwork. The rapid development of information and communication technologies in recent years has created the need in audiences to acquire new skills and continuously to improve © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 248–260, 2023. https://doi.org/10.1007/978-3-031-34866-2_19

Older Persons Media Usage During a One Month Election Campaign

249

them. Media and digital literacy have grown from specific professional competencies to a mandatory condition for adaptation and orientation in the modern digital world. This situation creates a disturbing trend of permanent alienation of certain social groups. Elderly people are seriously at risk of “digital isolation”, especially those well beyond their active working age, who have not had to work in a high-tech environment and who have not acquired the habits of using new technologies and the Internet. The undertaken study targets the situation in Bulgaria. According to the National Statistical Institute data, the share of people between the ages of 65 and 74, who have regularly used the Internet in 2021, is 30.8 percent. In comparison, regular Internet users between the ages of 55 and 64 comprise 63.1 percent, and those between 45 and 54 - 82.1 percent. Among young people, the share rises to over 90 percent. The characteristics of Internet consumption between the different age groups are notable. Statistics show that the use of the Internet by Bulgarians over the age of 65 has increased significantly over the last ten years. Thus, in 2011, only 6.2 percent of elderly people in the country used the Internet regularly, and in 2004 they were only 0.6 percent. The share of Internet users in the third age group is gradually increasing, but the gap with younger age groups remains drastic [9]. The digital gap between the generations is also evident from the sources of information that the representatives of the different age groups use. Traditional media and especially - television is still the dominant source of information for Bulgarians over 50, and especially for those over 60, according to a survey conducted by the Open Society Institute. In all other age groups, the majority prefers mainly Internet-based information sites and social media. New technologies, albeit slowly, are already finding a place in the everyday life of people of the third age. Social networks, however, are not particularly popular among people over 65 in Bulgaria – 44 percent of respondents use them vs. 55.56 percent of non-users. The main issues that those who enter social networks are looking for are curious information (16.67 percent) and communication with friends (22 percent). Not a single respondent relies on social networks to get informed about news and current events [10]. Eurostat’s comparative data show that the digital activity of the elderly in Bulgaria is lower compared to the average levels in the EU, although a generational difference while accessing the modern information society is noticeable in the EU as well. Thus, in 2020 88 percent of people aged 16–74 in the European Union reported that they had used the internet in the last 3 months, compared to 61 percent of those aged 65–74. Although internet usage remains high among young people in the EU, it varies considerably among older people. In 2020, Denmark recorded the highest proportion (94 percent) of people aged 65–74 who had used the Internet in the last 3 months, followed by Luxembourg and Sweden (both 91 percent), the Netherlands (90 percent), as well as Finland (88 percent). In contrast, in 2020 only 25 percent of people aged 65–74 in Bulgaria used the Internet in the last 3 months, in Croatia – 28 percent, and in Greece – 33 percent. Almost 9 in 10 among younger people in the EU aged 16 to 24 have participated in social networks (87 percent). In comparison, among seniors aged 65 to 74, over a fifth (22 percent) have done so. This share varies from 10 percent in Croatia, 12 percent in Bulgaria to 60 percent in Denmark [11]. The world of media is dynamic and changing at an unprecedented pace. The lives of the new generations radically differ from the ones of their ancestors, and they perceive media from a different perspective. Indeed, technological progress facilitates access to

250

L. Raycheva et al.

information, but definitely, this vast flow hampers the publics to make reliable choices about various important issues on everyday life. On one side, there are more information options that are beneficial for those who are media literate inside the diverse group of older people [8, 12, 13], but on another, they further deepen the generation gap caused by the inertia of those older people who are not so media savvy and less able to adapt to the digital environment. In this regard, the aim of the present study is to collect data about the “usual” media day among different Bulgarian family milieus. The study focuses mainly on the role that media play in people’s lives in hybrid media contexts.

2 Research Methods The study is non-representative and is based on the methodological framework developed by the University of Tartu, Estonia: a written diary with semi-structured questionnaires for each member of the observed family (including the interviewer) on a particular day and an explanatory essay about the media preferences of Bulgarian family members. It has been anonymized and has been carried out on a voluntary basis by BA students during their first admission year at the Faculty of Journalism and Mass Communication of the St. Kliment Ohridski Sofia University, who have signed an informed consent to participate in the research. A one month period of data collection has been set up: October 14 – November 14, 2021. The period coincides with the one-month election campaign in media related to the regular vote for president and vice president and the early one for Parliament. This period has been chosen, because in less than eight months in 2021, the country has faced one regular election for MPs of April 4, and one early – of July 11. In addition, cross-combining technique for data collection and analysis was applied in order to compensate some of the inherent limitations. The study raises three main research questions: 1. For what purpose did people use media during the one month national election campaign? 2. Which media do people prefer most often as a source for information? 3. Why people prefer certain media for information during the election campaign? The voluntary interviewers have been asked to choose a particular day to spend with their family. Family is interpreted in a broad sense - it could be the interviewers’ own birth family, or the family of their friends, relatives, etc. Each interviewer has been required to observe for 24 h the normal life of the family, without disturbing it, to take notes on the media consumption of each family member, and on the next day to conduct a brief interview with the family members about their media preferences during the previous day. Media preferences refer to: listening to the radio; consumption of music streaming services; watching TV, films and videos (including on demand); reading newspaper and online news; reading magazines; reading books; usage of social media. Three main aspects have been described in the essays: access to media, characteristics of a particular day (media-related activities, their character and amount), and personal data about family members.

Older Persons Media Usage During a One Month Election Campaign

251

In structuring the essays, the following points have been taken in consideration: main socio-demographics of every family member (age, gender, and occupational status); description of the media environment and access to media (possession of media devices; newspaper and magazine subscriptions; individual or collective usage of media; media content serving as a basis for conversation between the family members; explanation of media preferences for each family member with regard to age, gender, education, role in family, personal hobbies, occupation, stage on the life-cycle topical interests, media habits, media trust, etc. The age group option was set to 18, 19–30, 31–50, 51–64, and 65 +. The short interviews were important to explain the reasons why a family member prefers particular channels/contents and is it rather usual or non-usual for him/her to do so. For the purpose of the study, 50 in-depth interviews have been conducted with students and their families, and the responses have been received and analyzed. The results are to be used for comparing with the results of the academic teams from the other 13 EU member states, participating in the MEDIADELCOM research project of Horizon 2020 program, initiated by the University of Tartu, Estonia.

3 Results The age groups have been relatively equally represented in the conducted survey. The fewest are persons under the age of 18 (10 percent), while the largest number of respondents is between the ages of 19 and 30 (32 percent). Persons aged between 31 and 50 years old are 29 percent. Respondents aged 51 to 65 (15 percent) and over 65 (14 percent) are relatively evenly distributed. Of the respondents, 1 percent are with primary education, 62 percent – with a secondary education, 35 percent possess a university degree, and 2 percent have a doctorate. By occupation, heterogeneity is observed, with 40 percent of students and 12 percent of pensioners predominating among the respondents (Fig. 1).

14%

10%

15%

32%

29%

up to 18

19-30

31-50

Fig. 1. Age groups

51-65

over 65

252

L. Raycheva et al.

When describing the media environment and access to media, it is noteworthy that, regardless of the age group, almost every home has a TV set (98 percent), but only 31 percent of the respondents possess radio receiver. The age group of 51–65 and over 65 have radio receiver (83 percent), while young people under 18 and between 19–30 listen to the radio mostly in their car when they are on their way to work or to university (only 19 percent of them claim that they have a radio receiver at home). The trend for personal computer ownership is reverse. 74 percent of the under-aged own a personal computer or have access to a shared one at home, and 95 of the respondents aged 19–30 possess such a device. For the age group 31–50, the percentage is 83, for 51–65 – 61, and for the oldest over 65 years it is only 6. Smartphone ownership is similar - up to 18 years – 78 percent, 19–30 years – 94 percent, 31–50 years – 59 percent, 51–65 years – 61 percent, and over 65 years – 5 percent. One of the respondents explained what devices are using the two grandparents in the family: “For many years, the two main sources of media content for the older family members was radio, followed by television, which was not watched for more than one hour a day, but since the beginning of 2021, television has started to be used more and more often and for a longer time. TV has become their only source of media information and radio is no longer turned on. They also have access to other devices, such as a personal computer, laptop and smartphone, but they do not show any interest in them.” Only 6 percent of all age groups subscribe to electronic publications, and only 4 percent subscribe to print publications (mostly monthlies such as the National Geographic magazine). Depending on the type of device (desktop computer, laptop, smartphone) is its usage – at home, outside or mixed. Another respondent shared: “We collectively use the TV set and sometimes we have a disagreement about which channel to choose to watch. This happens mostly in the evening. We spend most of the time accessing media when we are at home, less outside of it.” From the observations made, it seems that each member of the family has a different media consumption and to a large extent it depends on the age group to which he/she belongs. One of the respondents – a student, says: “While my grandfather listens to the radio and still prefers mostly print media, my parents have adapted to technologies and boldly use electronic media. My sister and I belong to Generation Z, and for us print media is rather old-fashioned, while communication on social networks, online learning and the Internet is completely normal”.

Older Persons Media Usage During a One Month Election Campaign

253

The survey asks whether news in the media serve as a topic of conversation among family members (which topics and how often). Among the most discussed topics are those about the political situation in the country and abroad (53 percent), the special military operation of Russia in Ukraine (24 percent), economics (10 percent), sports events (5 percent), as well as other topics (8 percent altogether), such as secular events, pandemics (COVID 19), lifestyle news, etc. One respondent shared: “News take up a large part of the family’s conversations, with political news at the fore, both nationally and globally. At the moment, the most attention is paid to the news related to the situation between Ukraine and Russia. Non-political news, such as accidents and robberies, for example, is less frequently commented on”. The frequency of the topics discussed varies from every day in 54 percent of the cases, to twice or thrice a week. 31 percent and 15 percent answered respectively “rarely” or “almost never”. The study focuses also on the most frequently used media (Fig. 2). Online media is the most preferred by 71 percent of the respondents. 54 percent prefer watching TV, while 34 percent use social media. The number favoring print publications and video platforms such as YouTube is negligible.

Video plaorms such as YouTube Social Media Newspapers TV Online 0

10

20

30

40

50

60

70

80

Fig. 2. What media do you use most often?

One of the respondents describes the media day in the family in this way: “My grandparents mostly watch TV because it’s what they have been used to and it’s easy for them to use, while I (a student) and my sister use online media because of the freedom it provides and the opportunity to tap into more diverse perspectives”.

254

L. Raycheva et al.

When asked why they use these media more often, 78 percent answered that they are looking for the information they need on certain topics that interest them. A significant percentage – 49, of people said that they use certain media for entertainment. Among the remaining responses (Fig. 3) were those for easy access/timely information, convenience and communication.

Information Entertainment Easy access/Instant information Communication Convеnience 0

20

40

60

80

100

Fig. 3. Why do you use these media more often?

One of the respondents shared observations about the grandfather’s (over 65) media preferences: “The oldest member of the family does not want to rely on modern methods of information such as websites and social networks. He has no desire to learn how to handle them and prefers to watch TV and to read newspapers. These media are used for both information and entertainment. The person finds everything he is interested in and does not need other media”. According to the observations of one of the respondents, the grandparents in the family mainly use two television channels, both of which are watched for entertainment and information. “On the day of my observation, BNT (the public service broadcaster) was the entertainment channel and bTV (largest commercial TV station) was the news channel. I (a young student) listen to the news of Darik radio ((largest commercial radio station) to get information, while for fun and socializing I mostly use Facebook. My brother often visits Youtube, both for information and for fun and entertainment. We both use Facebook and Instagram to communicate”.

Older Persons Media Usage During a One Month Election Campaign

255

The most used device is TV set (48 percent of respondents), followed by laptop (26 percent), smartphone - 21 percent, and a minor percentage is allocated to radio receiver (4 percent) and printed publications (1 percent). The answers regarding the frequency of use of these media are heterogeneous. They are accessed on a daily basis predominantly at home, but also between lectures, in free time, in cases when the respondents are purposefully looking for certain information or wish to learn more about a particular current topic of social life in the country and abroad. The mother of one of the respondents said: “Intensive periods of media use for my elderly parents over 65 years old are mostly during news broadcasts, while for the parents themselves, who belong to the age group of 31-50 years, any time when are free enough is an intense time to use media. Therefore, the busier they are, the less they access media content. The media-free moments for adults are when they are reading a book, while for us (31-50) there are almost no such moments. No matter the activity, there is always a chance to glance at the smartphone.” The largest share of respondents answered that they prefer to be informed by the programs of the commercial national TV channels bTV (38 percent) and NOVA TV (29 percent). The public service Bulgarian National Television (BNT) holds the third place with 14 percent. The share of the Dnevnik Daily is almost insignificant – 2 percent of the respondents indicated it as a source of information. Radio stations, including the public service Bulgarian national radio (BNR) have not been preferred. Surprisingly, social media or social networks have not been selected by the observed respondents as a reliable source of information (Fig. 4).

2% 29%

17%

14%

38%

BNT

bTV

NOVA TV

Daily Dnevnik (Diary)

Other Fig. 4. What is the preferred traditional media for information?

256

L. Raycheva et al.

The poll also pays serious attention to media preferences during election campaigns. The interviewees, when asked whether they receive the necessary information in media in order to make an informed choice about: the presidential candidates and the parliamentary candidates, 84 percent of them answered positively, while 16 percent gave a negative assessment. According to one of the respondent’s observations: “My grandparents prefer BNT and bTV, but they claim that they didn’t pay much attention this year to the election campaign because they feel that they are already too old to vote. For my brother, there is no preference or a channel, neither for the presidential nor for the parliamentary elections, because he has decided for a long time who he will vote for and does not intend to burden himself with additional information. I usually follow the daily newspaper "Dnevnik" for both the parliamentary and presidential elections, because it seems the most objective, but it may also be so because it leans more towards my political credo”. When asked why they prefer particular information channels, the respondents gave mixed answers. (Fig. 5). Most of the interviewees - 40 percent, answered “To choose who to vote for”, and 36 percent indicated that they trust the channel. Other responses included: “Convenience”, “More information and more detail”, “Synthesized information”, etc.

By habit Convenience No access outside the… To make informed choice… Trust in the media 0

10

20

30

40

50

Fig. 5. Why do you prefer this channel for information during election campaigns?

Significant questions regarding disinformation in modern communication have also been put by the observing pollsters. They are related to the credibility of the information and whether family members are able to distinguish between reliable facts and fake news. 60 percent of the respondents answered that they can filter out fake news, while 40 percent admitted that they cannot distinguish between fake and credible facts. It is noteworthy that according to young people, their parents and especially grandparents (over 65 years of age) are more vulnerable to fake news [14]. One of the interviewers noted:

Older Persons Media Usage During a One Month Election Campaign

257

“The oldest person in the sample I presented is radically different from the other participants. With him, the consumption of the Internet is reduced to zero and he trusted most the so-called "classical media" - newspaper, radio, television. For this generation it is more difficult to sift through fake news, since in the past years there was not such a large information influx, and even if there was forgery, it has not been commented extensively”. Another pollster shared that the lack of higher education and the reluctance to new technologically among older people has a negative impact on their formation of a personal opinion. “They feel that they have reached a period in their lives when they no longer need to technologically develop their knowledge and skills. They rely solely on other people’s opinions or on what they see on television, which is the main source of information for them. Computer illiteracy affects them as it restricts them from searching for additional information online. Reading newspapers and books has long ceased for them and they are completely not aware with the specifics of new media. With younger people, it is just the opposite - they feel that the information they are "flooded" with should be verified, as they are getting information mainly online. The digital environment they are in allows them to interact with more people, listen to different opinions and distinguish between fake news”. Of the elderly (over 65 years old), 100 percent of the students observing them – (mostly their grandchildren), are of the opinion that they do not have the skills and knowledge to distinguish fake news from reliable information.

4 Discussion In today’s world, information and entertainment dominate media choices. Contemporary computerized people are increasingly looking for the news that actually interest them with regard to the most convenient time and format. When the consumers are flooded with information and entertainment, their choice becomes difficult [15]. It is important to note the increasingly minor role of the print media, even during election campaigns. The audiences prefer TV channels or online platforms for information, looking for the most suitable channel to confirm their choice to match their personal political partialities. From a psychological point of view, the needs, because of which the respondents in the undertaken survey choose a certain type of media can be described as: cognitive - to acquire information and knowledge about the world; affective - to provoke emotions in themselves through the media content; personal-integrative, related to the social trust, stability and social status of the individual; socially integrative - for inclusion in conversations with family, friends, colleagues; escape from reality and from the burden of their own social roles [16]. To these needs could be added the need to search for information and arguments to cast the vote for certain candidates in parliamentary or in presidential elections, characterized by the desire for social responsibility and social integration of electors in their role as responsible citizens.

258

L. Raycheva et al.

Most theories of media preferences rely predominantly on psychological predispositions to explain preferences through individuals’ behavior and are consumer-focused. Structural characteristics of surrounding environment play a very important role in forming media choices. By expressing their preferences, people can change and shape the very structures of the environment which surrounds them [17].

5 Conclusion, Limitation, Implications for Future Research The survey exploring the media habits in a Bulgarian family on a particular day brought about to some interesting results. With regard to the first research question: “For what purpose did people use media during the one month national election campaign”, 78 percent of the respondents answered that they are looking for the information they need on certain topics that interest them. A significant percentage – 49, of people said that they use certain media for entertainment. Among the remaining responses were those for easy access/timely information, convenience and communication. Each member of the family has different media consumption and to a large extent it depends on the age group to which he/she belongs. The majority of the elderly prefer information. When answering the second research question “Which media do people prefer most often as a source for information?”, the opinion of the largest share of respondents (especially of the elderly) tend to national commercial and public service TV channels, followed by a tiny preference of dailies. Radio stations, social media or social networks have not been selected by the observed respondents. Interesting answers have been obtained when answering the third research question: “Why people prefer certain media for information during the election campaign?” Most of the interviewees - 40 percent, answer that they need to that in order to choose who to vote for, and 36 percent indicate that they trust the channel. It is noteworthy, however, that according to young people, their elderly parents and especially grandparents (over 65 years of age) are more vulnerable to fake news. Based on the conducted observations in the media diaries, some conclusions could be drawn. For the purpose of information and entertainment respondents mostly use online media, with television also being a preferred media. Young interviewees and middleaged ones access online media using personal computer and smartphone, while elderly mostly watch television and the trust in their preferred media, especially during election campaign, is high. Radio, as well as printed publications, is used less and less at home. On the other hand, an intergenerational gap exists where young people consider their ancestors to be people with outdated views of the world surrounding them, unprepared for modern technologies, unable to distinguish between fake and credible news, thus becoming subject to political manipulation. Young respondents are confident in the correctness of the information they have chosen, thus displaying the so-called “cognitive dissonance” – they are convinced that they can successfully distinguish fake news from correct data. It is noteworthy that the members of the observed families, although accessing different media, freely discuss important topical issues of the public agenda. The conclusions that emerge are that generational differences in usage of information sources, Internet consumption and social networks are serious. The rapid development

Older Persons Media Usage During a One Month Election Campaign

259

of digital technologies and the significant role of the Internet and social networks in disseminating information create a risk of social isolation for the elderly in an increasingly digitally oriented world. Bulgaria is among the EU countries where the digital activity of older people is low, but even in countries with a higher standard of living, where adults using the Internet and social networks have a higher share, there is a huge gap in media consumption between them and their young compatriots. This shows that the financial factor is not as important for digital activity as other reasons related to attitude and digital skills. Another trend, which is visible from the data, is that in recent years, elderly people in Bulgaria, as those throughout Europe, have become more active users of the Internet and social networks, although they do not reach the level of the younger generations. In this sense, it is necessary to pay more attention to training for the people of the third age in terms of digital and media literacy, which would help them to better integrate in the modern world. A limitation of the conducted study is that the sample is somewhat biased, since the students from the education-oriented family-backgrounds are overrepresented. Nevertheless, the experiences of the researchers from the Tartu University of Estonia, who initiated the frame of the research, indicate that the main concerns are mostly related to the variety of roles/habits and meanings media can have for the respondents. Future empirical studies may address this point (for example by conducting a controlled experiment) and enlarge the sample’s characteristics. The results of the conducted study are indicative to those interested in contemporary media developments and especially in the role that media play in people’s lives in hybrid media contexts. Acknowledgements. The text has been developed within the international academic research project MEDIADELCOM of the European Union’s Horizon 2020 research and innovation programme.

References 1. Chadwick, A.: The Hybrid Media System: Politics and Power. Oxford University Press, Oxford (2013) 2. Couldry, N., Hepp, A.: The Mediated Construction of Reality. Wiley, New York (2016) 3. De Vreese, C.H., Neijens, P.: Measuring media exposure in a changing communications environment. Commun. Methods Measures, 10:2–3, 69–80 (2016). https://doi.org/10.1080/ 19312458.2016.1150441 4. Hasebrink, U., Domeyer, H.: Media repertoires as patterns of behavior and as meaningful practices a multi method approach to media use in converging media environments. Participations: J. Audience Reception Stud 9(2), 757–779 (2012) 5. Kobbernagel, C., Schrøder, K.C.: From everyday communicative figurations to rigorous audience news repertoires: a mixed method approach to cross-media news consumption. MedieKultur: J Media Commun. Res. 32(60) (2016) 6. Adoni, H., Peruško, Z., Nossek, H., Schrøder, K.C.: Introduction: news consumption as a democratic resource – news media repertoires across Europe. Participations: J. Audience Reception Stud. 14(2), 226–252 (2017). http://www.participations.org/Volume%2014/ Issue%202/11.pdf

260

L. Raycheva et al.

7. Kõuts-Klem, R., Brites, M.J.: Comparison of news consumption on online among Estonian and Portuguese audience. Participations. J. Audience Reception Stud. 14(2), 464–483 (2017). http://www.participations.org/Volume%2014/Issue%202/23.pdf 8. Loos, E.F.: Senior citizens: digital immigrants in their own country? Observatorio (OBS*) Journal 6(1), 1–23 (2012) 9. National statistical institute households with internet access at home. https://www.nsi.bg/en/ content/2808/households-internet-access-home (2021) 10. Open society institute. Media trust and media usage in 2022. https://osis.bg/?p=4213&fbc lid=IwAR2zUple9SlABEQCIFWxnz4jZWgn_WALRMBxlPmvf8GrsS4B78fq1nXD15Q (2022) 11. Eurostat. How popular is internet use among older people? https://ec.europa.eu/eurostat/web/ products-eurostat-news/-/edn-20210517-1 (2021) 12. Ivan, L., Loos, E.F., Bird, I.: The impact of technology generations on older adults’ media use: review of previous empirical research and a seven country comparison. Gerontechnology 19(4) (2020) 13. Loos, E.; Ivan, L.: Not only people are getting old, new media are too: technology generations and the changes in new media use. New Media Soc. (2022) 14. Loos, E., Nijenhuis, J.: Consuming Fake News: a matter of age? The perception of political fake news stories in Facebook ads. In: Gao, Q., Zhou, J. (eds.) HCII 2020. LNCS, vol. 12209, pp. 69–88. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-50232-4_6 15. Mixalov, B. Dovepie na potpebitelite km cebe ci. Medii i obwectveni komynikacii. Cofi: Izd. UHCC/Alma komynikaci, № 8 (2011). www.media-jou rnal.info. [Mihailov, V. Doverie na potrebitelite k˘um sebe si. Medii i obshtestveni komunikatsii. Sofiya: Izd. UNSS/Alma komunikatsiya. № 8 (2011)]. (Mikhailov V. Consumers’ Confidence in Themselves. Media and public communications. Sofia: Ed. UNSS/Alma Communication, No. 8 (2011)) 16. Katz, E., Gurevitch, M., Haas, H.: On the use of the mass media for important things. Am. Sociol. Rev. 38, 164–181 (1973) 17. Webster, J.G.: The role of structure in media choice. In: Hartman, T. (Ed.), Media Choice. A Theoretical and Empirical Overview. New York Abington: Routledge, p. 222 (2009)

Experimental Survey on Bridging the Digital Divide Through Daily Text Chat Communication with Virtual Agents Taishi Sawabe(B) , Masayuki Kanbara , Yuichiro Fujimoto , and Hirokazu Kato Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma, Nara, Japan {t.sawabe,kanbara,yfujimoto,kato}@is.naist.jp Abstract. To bridge the digital divide, this study investigates the current state of smartphone use and work to improve smartphone skills by using both online support and on-site support approaches in collaboration with industry, academia, and government. The proposed method aims to improve the frequency of smartphone use and smartphone skills through daily interaction with a virtual agent “Goodchan” using the SNS (LINE) text chat function as online support, as well as to eliminate concerns about smartphone use and improve motivation by conducting on-site smartphone consultation as on-site support. In this effort for dialogue chat, we use the Wizard of Oz method, in which the operator acts as Goodchan, and conducts dialogue chats with the participants. In addition, we proposed a smartphone skill evaluation sheet called SmaSki 2022 to quantitatively evaluate smartphone skills and surveyed changes in the smartphone skills of the participants. As a result, we confirmed that approximately 30% of the 46 elderly participants improved their smartphone skills during the six-month demonstration experiment at Shij¯ onawate-city in Japan. Keywords: Bridge the digital divide · Virtual agents Smartphone skills evaluation sheets · Daily dialogue

1

· Text chat ·

Introduction

With the miniaturization of semiconductors, opportunities to use information and communication devices with multiple sensor modules and communication functions, such as smartphones and wearable devices, are increasing [1]. According to data from the 2022 edition of the Ministry of Internal Affairs and Communications’ White Paper on Information and Communications in Japan, the overall mobile household ownership rate of information and telecommunications devices, which are terminals used to connect to the Internet, etc., in year of 2021 was 97.3%. Of these, 88.6% are smartphones, indicating that the smartphone penetration rate is nearly 90%. The Internet usage rate for individuals is also 82.9%, with smartphones in particular on the rise at 68.5%. These new c The Author(s), under exclusive license to Springer Nature Switzerland AG 2023  Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 261–273, 2023. https://doi.org/10.1007/978-3-031-34866-2_20

262

T. Sawabe et al.

devices have made people’s lives more convenient and comfortable, and have become indispensable. On the other hand, when asked “How will Japanese society change in the future due to the promotion of digitalization?” many of those who answered that they think it will be “darker than now” cited the widening digital divide between those who can and cannot use information and communication devices as the main reason [2]. Three major telecommunication carriers in Japan have announced plans to terminate services using 3G lines, and as a result, many flip phone users who used to use conventional communication methods have switched to using smartphones [3]. However, the UI/UX of smartphones differs significantly from flip phones (e.g., input by touching the screen instead of using a physical keypad, many multiple application functions, etc.), and the various functions require time and effort to become accustomed to their operation. In addition, many of those who are already using these devices cite issues such as the existence of similar applications (hereinafter referred to as “app”), downloading methods, and security issues including account passwords [4,5]. Therefore, physical and mental support to increase the number of people who use and can use information and communication devices (especially smartphones) will become important in the future. In this study, we aim to improve the frequency of smartphone use and smartphone skills through daily interaction with a virtual agent using the SNS (LINE app) text chat function as online support, as well as to eliminate anxiety about smartphone use and improve motivation by conducting on-site smartphone consultation as on-site support. We also aim to create an evaluation sheet to quantitatively assess smartphone skills.

2 2.1

Related Study to Bridge the Digital Divide The Definition of Digital Divide

In 1995 the term ‘Digital divide’ was first used in a number of newspapers in the United States, published by the National Telecommunications and Information Administration [6]. The digital divide refers to “the economic disparity between those who can benefit from information and communications technology (ICT) and those who cannot” and can be broadly classified into the following three categories [7]: 1. Inter-regional digital divide: Regional disparities in the availability of the Internet, broadband, etc. within a country. 2. Digital divide between individuals and groups: ICT-related differences in physical and social conditions (gender, age, student status, etc.). 3. International digital divide: International disparities in the availability of Internet and broadband. The digital divide is an important issue to be solved in the future with the spread of smartphones.

Experimental Survey on Bridging the Digital Divide

2.2

263

Case Studies of a Municipality’s Efforts to Bridge the Digital Divide

This section introduces case study examples of municipal demonstration experiments aimed at bridging the digital divide. Various efforts are being made by local governments to eliminate the digital divide for the elderly. For example, on September 6, 2021, Shibuya Ward in Tokyo started a demonstration project to lend smartphones free of charge to approximately 1,700 ward residents whose age were 65 or older who do not own a smartphone. This project was to improve the quality of life by eliminating the digital divide for the elderly [8]. In addition, by holding study sessions to support the elderly when they use smartphones for the first time and providing physical support for promoting smartphone use based on the results of analysis of smartphone use by the experiment participants. The project will provide continuous support from the time elderly people start using smartphones until they become proficient in using them, thereby increasing the smartphone use rate among elderly people in Shibuya Ward and improving their quality of life (QOL) through bridging the digital divide. Next, Kaga City, Ishikawa Prefecture, implemented three attempt in 2021 to bridge the digital divide for the elderly [9]. First, the city subsidizes the purchase of smartphones compatible with the My Number Card (the individual number card in Japan), with an upper limit of 5,000 yen per person. Second, a smartphone class is held every Friday afternoon for about two hours on a specific theme. Third, a smartphone consultation center is opened at specific locations twice a week, where users can freely ask for advice on how to use smartphones. Similarly, Kyoto Prefecture is holding smartphone classes for beginners as a countermeasure against the digital divide to realize a symbiotic society where people can enjoy a rich life with diverse values and lifestyles through the use of digital technology [10]. In both of these demonstration experiments, smartphone classes are conducted with the support of companies such as NTT DoCoMo, Inc. and SoftBank Corp. 2.3

Bridging the Digital Divide Through ICT Technology

There are also many examples of efforts to bridge the digital divide that have been implemented using ICT technology. As part of a demonstration project to bridge the digital divide for the elderly in Shibuya Ward, Vespula Inc. is providing a family communication service “Family Site” using a health application “Brain-enhancing App” for 400 families from May 2022 [11]. The “Brainenhancing App” is an application for brain science and technology, and is unique in that it incorporates five elements that are said to be effective in preventing dementia in brain science: “exercise,” “diet,” “brain stimulation,” “stress relief,” and “social participation,” into its application functions as “notification,” “steps,” “diet,” “brain training,” and “evaluation,” making it a highly effective application for dementia prevention. AP TECH, Inc. is attempting to remotely bridge the digital divide by providing the “Mimamori Service” Hachi application that utilizes the Apple Watch and

264

T. Sawabe et al.

iPhone as a “family-connecting” monitoring service [12]. Hachi,” a dog character, plays a central role in the Hachi application service, and serves as a link between the person who is being watched over and the person who is being watched over. The app functions include health and physical condition check (Apple Watch), location information check (GPS function), video call, SOS sending/receiving, and notification when physical condition changes. 2.4

The Position of This Study

This study targets the ICT use disparity caused by differences in physical and social conditions, which is (2) the digital divide between individuals and groups among the digital divide and investigates the current status of smartphone use and smartphone skill improvement by using both online and on-site support approaches in collaboration with industry, academia, and government. The project will use both online and on-site support approaches in collaboration with industry, academia, and government to investigate the current status of smartphone use and improve smartphone skills. Unlike conventional methods, we aim to improve the frequency of smartphone use and smartphone skills through daily interaction with virtual agents using the LINE text chat function as online support and to improve anxiety and motivation toward smartphone use by conducting on-site smartphone consultation as on-site support.

3 3.1

The Interactive Support System for Bridging the Digital Divide Interactive Agent “Goodchan”

Online support uses an interactive chat system based on the humanoid agent “Goodchan”. Figure 1 shows the overall image of Goodchan. This agent was newly conceived and created by the authors for this project, and has a childlike and simple appearance, as well as a smooth face, which looks to give any user a sense of familiarity [13,14]. The name “Goodchan” comes from the meaning of “good” and its character of being lazy sometimes. Another research states that sometimes good and sometimes bad characteristics accelerate motivation of learning by the agent [15]. It is designed to make users who are learning new and complicated things, including how to use a smartphone, feel comfortable interacting with the agent and increase their motivation of learning new things. 3.2

Online Support via Goodchan LINE App

With the interactive agent “Goodchan” as the main character, we propose that online support provides a remote support method for when the user is unable to participate on-site or for daily communication by using LINE App provided by LINE Corporation, which is widely used by all ages from 10s to 70s in Japan among social networking service usage by age group [17,18]. Since many Japanese

Experimental Survey on Bridging the Digital Divide

265

Fig. 1. Interactive agent “Goodchan”.

smartphone users are already familiar with the UI/UX of LINE, we will use LINE apps that can utilize the same familiar UI/UX in this initiative. This dedicated application is divided into two parts: one for management (used by the operator) and one for users (used by the users). The operator can chat with users mainly through LINE as the Goodchan, and for the dialogue (or chat), the Wizard of Oz method is used, where the operator acts as Goodchan and chats with users [19]. This is because current chatbot-like technologies for automatic conversation alone cannot handle content if the context is difficult to understand or specialized content, and as a result, this may cause users to be discouraged from using the system and smartphones. In the experiment, the user is told that Goodchan is a semi-automatic agent, that it will respond automatically, and that a specialist may take charge of content if the content is too difficult or too complicated. Figure 2 shows the management screen of the operator side, where a list of users can be viewed. In this screen, users who have not yet replied are indicated in red text as “Unread” so that operator can see which users have already replied. Figure 3 of operators’ screen that appears when an operator selects the user from the list of management screens. The left side of the screen is a dialogue screen (or chat screen) with the user, and the right side displays a dialogue input screen for the operator to reply to and a stamp selection screen. The dialogue screen allows the operator to view part of the dialogue history with the user, and the stamp selection screen allows the operator to send a stamp-like image to the user by selecting an image that combines a specific word and Goodchan, which has been prepared in advance. Figure 3 of the user’s screen shows the screen of the user side, where the user can ask Goodchan to become a LINE friend, just as in a normal LINE chat session, and can ask Goodchan questions or concerns through chat messages.

266

T. Sawabe et al.

Fig. 2. The web screen of the users’ list for the operator.

4 4.1

Social Demonstration Experiment for Bridging the Digital Divide Experiment Overview

We conduct the demonstration experiment at three venues (municipal community center, Green Hall Tahara at Tahara branch office, and Nanpuso senior citizen welfare center) with the cooperation of Shijonawate City, Osaka Prefecture. We ask for the cooperation of the elderly those aged 65 or older who already have smartphones among the citizens who regularly use the venues as participants. The purpose of this experiment is to investigate the current status of smartphone use by the elderly, and from this, to investigate the current status of the digital divide and to verify the effectiveness of the proposed method as a solution to the digital divide. The experiment is conducted with 47 participants from the start of August 2022 to the end of January 2023. Online support will be available from the day of registration, allowing users to start a dialogue anytime, anywhere. On-site support will be provided at each location, with smartphone consultation sessions held once or twice a week. The smartphone consultation is divided into morning and afternoon sessions and is held at three locations on different days of the week. Depending on the day of the week, the morning and afternoon sessions may be held at different locations, and users are asked to participate at their preferred location, either in the morning or in the afternoon. The research will be conducted based on the evaluation through questionnaires and the analysis of actual data on smartphone usage through the interactive system.

Experimental Survey on Bridging the Digital Divide

267

Fig. 3. The operator’s web screen of chat reply on the left, and the user’s smartphone screen of LINE app on the right.

4.2

On-Site Support by Conducting Smartphone Consultations

On-site support consists of two main types of support. One is to hold on-site smartphone classes and the other is to hold smartphone consultation meetings. The smartphone class is mainly a one-to-many participants format, with explanations of the contents of the class based on the theme and overall explanations. While this is more efficient in terms of time, as the same content can be presented to the entire group at the same time, and the content can be decided by the theme, on the other hand, it makes it difficult to hold the class if there are differences in the level of smartphone use, as it is impossible to respond to each individual in detail. Unlike the smartphone class, the smartphone consultation session is based on a one-on-one response, and the content is not based on a theme, but rather on answering the questions and doubts that users have on a daily basis. One of the characteristics of this system is that it allows for one-on-one support, which can be tailored to the level of each participant, and also allows for sufficient response to questions and doubts that participants may have. Figure 4 shows on-site support at a smartphone consultation. 4.3

Results and Discussion of Overall Approach

Figure 5 of the right upper graph shows the results of the overall effort of the demonstration experiment. The horizontal axis of the graph is a 7-point Likert scale, with 1 representing very dissatisfied and 7 representing very satisfied, and the vertical axis representing the number of participant responses. The graph shows the responses of 44 participants in total (excluding 3 who did not respond), and the fact that no participant rated themselves as dissatisfied (those who chose 3 or less) and that more than 90% of the participants rated themselves at 5 or

268

T. Sawabe et al.

Fig. 4. On-site support scene at a smartphone consultation.

higher throughout the entire demonstration experiment indicates that they were satisfied or very satisfied with the content of this on-site and online support demonstration experiment. This indicates that the participants are satisfied or very satisfied with the content of the on-site and online support demonstrations. 4.4

Results and Discussion of Smartphone Consultations

The total number of users of the smartphone consultation was 496 from August 2022 to the end of January 2023. This number was calculated by measuring the number of participants at each of the three locations for each event and adding the sum of the three values for each month. Although the number of participants varied at each location, an average of about 20 people used the service each month. Figure 5 of the right upper graph shows the level of satisfaction with the on-site smartphone consultation service. In this graph, more than half (25 out of 44) of the participants rated themselves as very satisfied with the service rated as 7. Figure 5 left lower graph shows the results for the adequacy of the consultant’s support on a 7-point scale (7 being adequate), and the right lower graph shows the results for the ease of understanding of the consultant’s explanations on a 7point scale (7 being adequate). In both cases, more than half of the participants chose a rating of 7. From the above and the free comments sections, it can be seen that the smartphone consultation, which is on-site support, was a consultation that was satisfactory, and that the quality of the support provided by the on-site supporters had an impact on the reason for this. As a point for improvement, this year’s event was basically in the form of answering users’ problems and questions rather than a consultation, but some users commented that they would have liked more one-on-one lectures on certain themes, as in a classroom style. If there is an indicator that shows the current smartphone level of the user, it would be possible to understand the issues in advance and provide support suited to each user.

Experimental Survey on Bridging the Digital Divide

269

Fig. 5. Overall results about smartphone consultation.

4.5

Results and Discussion on the Interactive Chat System

Figure 6 shows the results of the interactive chat system with agent Goodchan. The left upper graph shows the results of the respondents’ evaluation of their enjoyment of LINE with Goodchan on a 7-point scale (7 being very enjoyable), and the right upper graph shows the results of their evaluation of their feeling of familiarity with Goodchan on a 7-point scale (7 being very familiar). The left lower graph shows the results of the evaluation of Goodchan’s appearance on a 7-point scale (7 being very good), and the right lower graph shows the results of the evaluation of whether the respondents wanted to use their smartphones more through Goodchan on a 7-point scale (7 being very much so). The results show that, overall, many users gave a rating of 5 or higher, indicating that Goodchan has an approachable appearance and provides an enjoyable opportunity for dialogue. However, there were a certain number of users who rated Goodchan low, and the reasons for this include the opinion that they did not need to ask the questions online through Goodchan because they could ask the questions directly to the consultants at the on-site consultation. Moreover, users also said that it was difficult to communicate issues and questions through chat and text-based communication. Figure 7 shows an example of dialogue between a user and Goodchan some users treat Goodchan as friends to tell their life by themselves. Figure 8 shows a histogram of the frequency of LINE conversations. This results show that a total of 25 respondents used LINE more than once a week, and 16 respondents used LINE more than once every three days, out of a total of 47 respondents.

270

T. Sawabe et al.

Fig. 6. The result on dialogue through LINE with Goodchan.

4.6

Results and Discussion on the Smartphone Skills Assessment Sheet

As an index for evaluating smartphone skills, we created a smartphone skill evaluation sheet “SmaSki 2022” based on the results of this study and information from commercially available textbooks [20–22]. A person’s current smartphone skills can be evaluated by answering the questionnaire items. The items consist of 10 major items and a total of 108 sub-items within each of the major items. As a result of using this evaluation sheet, we were able to confirm that the smartphone skills of the elderly improved by approximately 30% during the demonstration experiment. Figure 9 shows the Smartphone Skills Assessment Sheet 2022 Edition (SmaSki 2022). You can access the SmaSki 2022 through the QR code below or the URL in the appendix to check your smartphone skills.

5

Summary

In a summary, this study investigated the current state of smartphone use and worked to improve smartphone skills by conducting daily dialogues with virtual agents Goodchan using the LINE text chat function and on-site smartphone consultation sessions for the goal of bridging the digital divide. As an outcome, we proposed a smartphone skill evaluation sheet (SmaSki 2022) to quantitatively evaluate smartphone skills, and not only investigated changes in the smartphone skills of the participants but also confirmed that smartphone skills improved by approximately 30% for 46 elderly people during the six-month demonstration experiment. In the future study, based on the results and achievements of this study, we aim to construct and generalize a system that can look after the elderly daily by supporting the bridging of the digital divide through dialogue centered on dialogue agents.

Experimental Survey on Bridging the Digital Divide

Fig. 7. Examples of dialogue between users and Goodchan.

Fig. 8. Frequency of dialogue through LINE with the dialogue agent.

271

272

T. Sawabe et al.

Acknowledgement. This work was supported by the Grand Challenge, Initiative for Life Design Innovation (iLDi) MEXT Society 5.0 Realization Research Support Project, Osaka University.

Appendix

Fig. 9. The Smartphone Skills Assessment Sheet 2022 Edition (SmaSki2022) https:// ssq.ami-lab.online/review/questionnaire items.php.

References 1. The ministry of internal affairs and communications in everyday life, part 2: current status and issues in the information and communications sector section 8: trends in digital applications, 2022 White paper, pp. 1–13 (2022) 2. The ministry of internal affairs and communications in everyday life, introduction: history of digitalization in japan chapter 1: current status and challenges of digitalization, 2021 White paper, pp. 1–454 (2021) 3. The ministry of internal affairs and communications in everyday life, draft direction for consideration (matters related to trends in the telecommunications market), pp. 1–92 (2022) 4. AEON RETAIL Co., Ltd., Survey on Smartphone Use by Senior Citizens, AEON News Release, pp. 1–8 (2019) 5. Mohd Mohadis, H., Mohamad Ali, N.: A study of smartphone usage and barriers among the elderly. In: 2014 3rd International Conference on User Science and Engineering (i-USEr), pp. 109–114 (2014) 6. Van Dijk, Jan A.G.M.: The digital divide. Project: digital inequality, Polity Press (2019) 7. The ministry of internal affairs and communications, white paper on information and communication (2011) 8. Ward, S.: Launch of a demonstration project to bridge the digital divide for the elderly. https://www.city.shibuya.tokyo.jp/kusei/koho/hodo/20210906.html. Accessed 3 Feb 2023 9. Kaga City, measures to combat the digital divide, Kaga City’s efforts, pp. 1–23 (2021)

Experimental Survey on Bridging the Digital Divide

273

10. Kyoto prefecture, Measures to combat the digital divide kyoto prefecture’s efforts. https://www.pref.kyoto.jp/digital/digital-divide/index.html. Accessed 3 Feb 2023 11. BSPR Co., Ltd., brain health maintenance apps brain apps, brain apps site. https://www.braincure.jp/. Accessed 3 Feb 2023 12. AP Tech Inc., List of HACHI AP TECH services. https://aptechnology.co.jp/. Accessed 3 Feb 2023 13. Matsui, T., Yamada, S.: Robot’s impression of appearance and their trustworthy and emotion richness. In: 2018 27th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), pp. 88–93 (2018) 14. Nishimura, S., et al.: Positive emotion amplification by representing excitement scene with TV chat agents. Sensors 20(24), 7330 (2020) 15. Tainaka, K., et al.: TSUNDERE Interaction: behavior modification by the integrated interaction of cold and kind actions. In: International Conference on HumanRobot Interaction, ACM/IEEE, pp. 153–156 (2021) 16. NTT Docomo Inc, LINE usage rate exceeds 80%: 80–90% of 10 to 50 somethings use LINE survey and research: introduction of the themes of surveys and research conducted by the mobile society research institute. https://www.moba-ken.jp/index. html. Accessed 3 Feb 2023 17. LINE Corporation, LINE business guide (Summary) marketing solution company LINE corporation January-June 2021, V1.0, pp. 1–19 (2021) 18. LINE API LINE developers. https://developers.line.biz/en/services/messagingapi/. Accessed 3 Feb 2023 19. Kelley, J.F.: CAL - a natural language program developed with the OZ paradigm: implications for supercomputing systems. In: First International Conference on Supercomputing Systems, pp. 238–248 (1985) 20. Takarajimasha Inc., NHK Hobby Dokkit! introduction to smartphones 100 things you want to do with your smartphone, TJ MOOK, pp. 1–127 (2018) 21. NHK Publishing Inc, Hobby Dokki! Are you using it? Smart phone, NHK Textbook, pp. 1–127 (2022) 22. MIC Ministry of Internal Affairs and Communications, Troubleshooting book for active seniors to use smartphones, tablets, and other internet devices Safely, pp. 1–30 (2020)

Credibility Judgment Against Online Health Misinformation Among Older Adults: Integrated View of Psychological Distance and Health Literacy Shijie Song1,2 , Xinyue Li3 , Jianjun Sun3(B) , and Yuxiang Zhao4 1 School of Information Management, Wuhan University, Wuhan 430072, China 2 Business School, Hohai University, Nanjing 211100, China 3 School of Information Management, Nanjing University, Nanjing 210023, China

[email protected] 4 School of Economics and Management, Nanjing University of Science and Technology,

Nanjing 2110094, China

Abstract. Exposure to online health misinformation can be dangerous as it may lead to incorrect health decisions. This study aimed to investigate the predictors of credibility judgments against online health misinformation among older Chinese adults. We used an online experimental approach and adapted four stimuli containing misinformation from cancer and non-cancer articles. Sixty participants aged between 50 to 75 years completed the study. We measured objective health literacy (using the Newest Vital Sign) and subjective health literacy (using the eHEALS instrument) prior to the experiment. Participants then read the stimuli and rated each article’s credibility, as well as two types of psychological distances (i.e., social and hypothetical distance) to the health-related claims in the stimuli. Multiple linear regressions showed that older adults who had a closer psychological distance to the health issues had a decreased ability to judge the credibility of online health misinformation. Additionally, higher objective health literacy predicted more accuracy in credibility judgments, while higher subjective health literacy negatively influenced older adults’ credibility appraisal of misinformation. Keywords: Health misinformation · Credibility judgment · Older adults · Psychological distance · Health literacy

1 Introduction In recent times, people have increasingly turned to the internet to search for health-related issues. While the internet provides easy access to a wealth of health information, many researchers have raised concerns about the quality of online health information and have warned against online health misinformation. For example, during the COVID-19 pandemic, health misinformation spread rapidly through social media, leading to significant public health problems described as an “infodemic.” Online health misinformation can confuse individuals and mislead their health decisions. Hence, judging the credibility of online health misinformation is critical. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 Q. Gao and J. Zhou (Eds.): HCII 2023, LNCS 14042, pp. 274–283, 2023. https://doi.org/10.1007/978-3-031-34866-2_21

Credibility Judgment Against Online Health Misinformation

275

Credibility refers to the perceived information quality or believability by recipients [1]. Since the general quality of online health information is questionable, it is difficult for laypeople to distinguish between high-quality and low-quality health information sources. Previous studies have shown that misinformation is prevalent in many health topics, including cancer, vaccines, smoking, and drug-related issues [2]. For instance, Johnson et al. systematically assessed cancer information on social media and found that 32.5% of cancer-related articles contained misinformation [3]. Online health misinformation is often persuasive, and it elicits engagement from information recipients. A recent systematic review found that online health information is often framed using persuasive strategies such as fabricated narratives, emotional appeals, sensational language, biased reasoning, and exaggerated risks [4]. These persuasive contents or cues can distort individuals’ assessment of information credibility since judging credibility is a complex cognitive process that involves the evaluation of information content and heuristic cues [5]. Therefore, judging the credibility of misinformation is one of the most challenging tasks for users seeking online health information. Moreover, the capacity to judge the credibility of online health misinformation varies significantly across the population. Among various individual characteristics, age appears to be a significant predictor of users’ credibility judgments of online health information. For example, Liao and Fu found that older adults are less sensitive to credibility cues than younger adults [6]. Although online health misinformation threatens people across the lifespan, studies suggest that older adults are more vulnerable than other population groups, especially in judging the credibility of online health-related content [7]. In addition, health literacy is also conceived as a significant predictor of competencies in evaluating online health misinformation. For instance, Nan et al. suggested that an individual’s knowledge, literacy, scientific belief, and critical thinking skills help to judge health misinformation [8]. Despite the significant attention given to online health misinformation, at least two research gaps remain. First, although researchers have noted that older adults are vulnerable to online health misinformation, the underlying psychological mechanisms of health misinformation judgment among older adults are still understudied. Second, while the association between health literacy and credibility judgment is frequently mentioned in previous literature, the relationship is not statistically examined sufficiently. Therefore, this study aims to bridge these two gaps by combining psychological processing and health literacy views to understand the credibility judgment of online health misinformation among older adults.

2 Theoretical Background Despite common commentary in prior literature regarding older adults’ susceptibility to online health misinformation, few studies have uncovered the mechanisms of why older adults are vulnerable to such misinformation. Some studies have noted that older adults may have more health information needs than younger people due to the increased prevalence of health issues as one ages [9]. These varied information needs may trigger older adults’ health information seeking, during which they are likely to encounter misinformation. However, the view that information needs alone explain older adults’

276

S. Song et al.

difficulty with online health misinformation judgment may not be sufficient. We argue that older adults’ perceived psychological distance to health issues may be an appropriate perspective to understand their susceptibility to believing claims stated in online health misinformation. The concept of psychological distance was introduced in Trope and Liberman’s Construal Level Theory (CLT) [10]. Psychological distance is “the subjective distance between an actor and an event in the actor’s psychological space” ([11], p. 707). According to CLT, psychological distance can be categorized into four interrelated types of distances that separate an individual from external instances: social distance (between the self and instances), spatial distance (between here and instances), temporal distance (between now and instances), and hypothetical distance (between reality and instances). CLT posits that an individual’s psychological distance to an instance is associated with their information processing related to that instance. Specifically, reduced psychological distance leads to lower levels of construal, meaning that mental representations become more concrete rather than abstract [12]. Since people tend to view concreteness as more likely to be true [13], we suspect that psychological distance to health issues may influence older adults’ credibility judgments of online health misinformation. However, the concept of psychological distance has rarely been employed to understand credibility judgments. Therefore, we pose the following research question: RQ1: Does psychological distance to health issues influence older adults’ credibility judgments of online health misinformation? Besides, health literacy is often used to explain the effects of misinformation, as it is assumed to be a contributing factor [14]. The definition of health literacy has evolved over time, with Berkman et al. [15] defining it as the “degree to which individuals can obtain, process, understand, and communicate about health-related information needed to make informed health decisions” (p. 16). In online settings, the concept of health information is often referred to as eHealth literacy, which is defined as the “ability to seek, find, understand, and appraise health information from electronic sources and apply the knowledge gained to addressing or solving a health problem” (p. 1) [16]. Squiers suggested that health literacy can aid in consumers’ information seeking and assessment [17]. Thus, health literacy may predict older adults’ ability to assess the credibility of online health misinformation. However, studies have revealed inconsistent results regarding the explanatory power of health literacy in predicting older adults’ discrimination of online health misinformation. For example, Zhang and Song [18] found that older adults with lower health literacy have misconceptions about credibility judgments during online health information seeking. In contrast, Chang et al. [19] found that the effects of health literacy on predicting the number of credibility criteria were not statistically significant. Given the contradictory evidence, this paper will further explore the research question of whether health literacy predicts older adults’ credibility judgment on online health misinformation as the second research question: RQ2: Does health literacy predict older adults’ credibility judgment of online health misinformation?

Credibility Judgment Against Online Health Misinformation

277

3 Methods 3.1 Participants The study employed online experimental methods. We used the snowball sampling method to recruit potential participants from social media and offline communities. We conducted a pre-screening to select laypeople participants aged 50 or above with no medical background. Finally, the study obtained a sample of 60 participants. 3.2 Stimuli The preliminary stimuli were inspired by prevalent online health misinformation reported by a government-sponsored platform (http://www.piyao.org.cn/). We selected four messages containing online health misinformation covering cancer (i.e., cancer causes, cancer treatments) and non-cancer (i.e., diets, vitamin D) topics. The messages were manipulated to be of comparable length and readability and became the final stimuli. 3.3 Measures The survey collected basic information from participants, including age, sex, education, health status, and chronic conditions. In addition, we developed two items to measure social and hypothetical psychological distance. The social psychological distance was measured by asking participants if the message was relevant to them, while the hypothetical psychological distance was assessed by asking if the health issue was likely to happen in their life. Both items utilized 5-point Likert scales, with 1 indicating strong closeness and 5 indicating strong farness. We measured health literacy using two instruments: objective health literacy was measured using the Newest Vital Sign (NVS) [20], and subjective health literacy was measured using eHEALS [21]. The NVS scale consists of a nutrition label for ice cream and six questions, each worth one point, designed to assess whether an individual can read and understand the nutrition label accurately. The eHEALS scale consists of eight items that evaluate users’ self-reported comfort and skills in finding, assessing, and applying online health information to make health decisions. The six items in the NVS were summed to create a total score ranging from 1 to 6. The items in eHEALS used 5-point Likert scales (with 1 point indicating strongly disagree and 5 points indicating strongly agree), which were then used to calculate a mean score. Three items assessed the credibility perception of online health misinformation by asking how individuals felt about the message’s trustworthiness, expertise, and objectivity. The items utilized 5-point Likert scales, where a score of 1 meant “strongly disagree” and a score of 5 meant “strongly agree.” An average score was then calculated for further analysis. 3.4 Procedure The online survey link was sent to the participants by instant massager, who were required to first read and approve the informed consent before proceeding to the online experiment. Then, participants were asked to fill out the demographics and health literacy

278

S. Song et al.

questionnaire. They were then directed to read four stimuli and assess their credibility, as well as rate their perceived psychological distance from the articles after reading each one.

4 Results 4.1 Participant Characteristics The sample included sixty participants, of which 34 (56.7%) were female. Their ages ranged from 50 to 75 years (Median = 58, S.D. = 4.09). Twenty participants (33.3%) had bachelor’s degrees or above. More than half of the participants (37, 61.6%) reported feeling that they were in relatively good or very good health. Thirteen participants (21.7%) reported having one or more chronic conditions. Please refer to Table 1 for more details. Table 1. Participant characteristics (N = 60) Demographic variables Sex Age

Education

Health status

Chronic conditions

Frequency

Percentage

Female

34

56.7%

Male

26

43.4%

50–54

10

16.7%

55–59

37

61.7%

Over 60

13

21.7%

Primary school

1

Junior high school

3

Senior high school

32

1.7% 5% 53.3%

Associate degree

4

6.7%

Bachelor’s degree

10

16.7%

Master’s degree

10

16.7%

Relatively bad

4

6.7%

Normal

19

31.7%

Relatively good

26

43.3%

Very good

11

18.3%

Yes

13

21.7%

No

47

78.3%

4.2 Regression Results The data were analyzed using the ordinary least squares (OLS) regression method in Stata 15. Given the presence of heteroskedasticity in the sample, we utilized OLS regression with robust standard errors to estimate the effects of psychological distances and

Credibility Judgment Against Online Health Misinformation

279

health literacy on participants’ credibility judgments of online health misinformation. The findings of this analysis are presented in Table 2. Table 2. Results of OLS regression with robust standard errors Independent variables

Coefficient

Robust S.D.

t-statistic

Control variables

P-value

−.010

.070

−0.14

.892

Age

.001

.010

0.09

.931

Education

.034

.025

1.34

.181

Sex (Male)

−.009

.043

−0.21

.835

Chronic conditions (Yes)

.039

.081

0.47

.636

Psychological distances

Social distance

.264

.068

3.87