Design Thinking Research: Innovation – Insight – Then and Now (Understanding Innovation) 3031361024, 9783031361029

Citation preview

Understanding Innovation

Christoph Meinel Larry Leifer  Editors

Design Thinking Research Innovation – Insight – Then and Now

Understanding Innovation Series Editors Christoph Meinel, Potsdam, Germany Larry Leifer, Stanford, USA

“Everyone loves an innovation, an idea that sells.” Few definitions of innovation are more succinct. It cuts to the core. Yet in doing so, it lays bare the reality that selling depends on factors outside the innovation envelope. The “let’s get creative” imperative does not control its own destiny. Expressed another way, in how many ways can we define innovation? A corollary lies in asking, in how many ways can the innovative enterprise be organized? For a third iteration, in how many ways can the innovation process be structured? Now we have a question worth addressing. “Understanding Innovation” is a book series designed to expose the reader to the breadth and depth of design thinking modalities in pursuit of innovations that sell. It is not our intent to give the reader a definitive protocol or paradigm. In fact, the very expectation of “one right answer” would be misguided. Instead we offer a journey of discovery, one that is radical, relevant, and rigorous.

Christoph Meinel • Larry Leifer Editors

Design Thinking Research Innovation – Insight – Then and Now

Editors Christoph Meinel Hasso-Plattner-Institut University of Potsdam Potsdam, Germany

Larry Leifer Stanford University Stanford, CA, USA

ISSN 2197-5752 ISSN 2197-5760 (electronic) Understanding Innovation ISBN 978-3-031-36102-9 ISBN 978-3-031-36103-6 (eBook) https://doi.org/10.1007/978-3-031-36103-6 © The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed 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

In 2023, design thinking will be implemented by many and recognized by most. The trajectory of design thinking cannot be slowed. Change is inevitable, and it is prudent to employ the tools that enable strategic leaders to transform, scientists to build prototypes, and the average person to do both. Design thinking is accessible to everyone. What is design thinking? Design thinking is a mindset that enables you to potentially improve any innovation process. The simple steps yield human-centered solutions that are both dynamic and reflective. Best results are possible when an interdisciplinary and diverse team collaborates to obtain a 360-degree view. The Design Thinking Research volumes of the series Understanding Innovation published by Springer present research findings of top scholars at Hasso Plattner Institute in Potsdam, Germany, and Stanford University in California, who were selected to participate in the transatlantic Hasso Plattner Design Thinking Research Program. Research teams have developed inspiring new tools and state-of-the-art approaches. In an accessible manner, this volume summarizes the research results of our research teams in the program year 2021/2022. Fourteen years ago, I was curious what could be achieved and so decided to initiate this program. Over the last decade and a half, researchers have indeed made ground-breaking discoveries. We have come a long way in terms of clearly defining problems and understanding the underlying principles at work in project teams that employ design thinking. Why is design thinking important? It provides guidance and inspiration and ultimately has societal impact. Practitioners use the design thinking skillsets to advance and expand their transformation initiatives. Goals vary, but invariably the results foster innovation in multiple realms: social, economic, technological, and so forth. Challenge yourself and those around you, iterate, keep an open mind, learn from others, communicate effectively, and consider yourself part of the design thinking community. Adopt this method and work with us to understand its nuances and

v

vi

Foreword

broad realms of application. This is an interesting time for the field of design thinking, with individuals of all ages enabled to take valuable learnings from the design thinking process. Enjoy the cutting-edge findings presented in this publication. Palo Alto, CA, USA

Hasso Plattner

Contents

Introduction/Roadmap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christoph Meinel and Larry Leifer Decades of Alumni: Perspectives on the Impact of Project-Based Learning on Career Pathways and Implications for Design Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sheri D. Sheppard, Helen L. Chen, George Toye, Aya Mouallem, Micah Lande, Lauren Shluzas, Timo Bunk, Nada Elfiki, Johannes J. L. Lamprecht, and Katharina Prantl Part 1

25

Application of Design Thinking to Governance and Social Causes

Predicting Creativity and Innovation in Society: The Importance of Places, the Importance of Governance . . . . . . . . . . . . . . . . . . . . . . . . Julia von Thienen, Kim-Pascal Borchart, Detlef Bartsch, Lars Walsleben, and Christoph Meinel An Exploration of Agile Governance in Rwandan Public Service Delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reem Abou Refaie, Lena Mayer, Karen von Schmieden, Hanadi Traifeh, and Christoph Meinel Voices from the Field: Exploring Connections Between Design Thinking Approaches and Sustainability Challenges . . . . . . . . . . . . . . . Nicole M. Ardoin, Alison W. Bowers, and Daniella Lumkong Part 2

1

47

77

91

Prototyping

User Perceptions of Privacy Interfaces in the Workplace . . . . . . . . . . . . 109 Michelle S. Lam, Matthew Jörke, Jennifer King, Nava Haghighi, and James A. Landay vii

viii

Contents

Assisting Learning and Insight in Design Using Embodied Conversational Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 Rebecca Currano and David Sirkin How to Tame an Unpredictable Emergence? Design Strategies for a Live-Programming System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Marcel Taeumel, Patrick Rein, Jens Lincke, and Robert Hirschfeld Part 3

Enhancement through Design Thinking

What Is Design Thinking? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Jan Auernhammer and Bernard Roth NeuroDesign: Greater than the Sum of Its Parts . . . . . . . . . . . . . . . . . . 197 Jan Auernhammer, Jennifer Bruno, Alexa Booras, Claire McIntyre, Daniel Hasegan, and Manish Saggar A Neuroscience Approach to Women Entrepreneurs’ Pitch Performance: Impact of Inter-Brain Synchrony on Investment Decisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 Stephanie Balters, Sohvi Heaton, and Allan L. Reiss Priming Activity to Increase Interpersonal Closeness, Inter-Brain Coherence, and Team Creativity Outcome . . . . . . . . . . . . . . . . . . . . . . . 227 Stephanie Balters, Grace Hawthorne, and Allan L. Reiss Design the Future with Emotion: Crucial Cultural Perspectives . . . . . . . 243 Chunchen Xu, Xiao Ge, Nanami Furue, Daigo Misaki, Hazel Markus, and Jeanne Tsai Part 4

Design Thinking Best Practices and Strategy

Opportunities and Limitations of Design Thinking as Strategic Approach for Navigating Digital Transformation in Organizations . . . . 271 Annie Kerguenne, Mara Meisel, and Christoph Meinel Designing Innovation in the Digital Age: How to Maneuver around Digital Transformation Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323 Carolin Marx, Thomas Haskamp, and Falk Uebernickel Facets of Hybrid Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347 Selina Mayer, Martin Schwemmle, Claudia Nicolai, and Ulrich Weinberg Design Thinking Transfer Gap: Differences Between Knowledge and Application of Design Thinking in the Organizational Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 Lena Mayer, Selina Mayer, Katharina Hölzle, Nikolaus Bönke, and Christoph Meinel A Genealogy of Designing as Performance . . . . . . . . . . . . . . . . . . . . . . . 383 Jonathan A. Edelman, Joaquin Santuber, and Babajide Owoyele

Introduction/Roadmap Christoph Meinel and Larry Leifer

1 Fourteen Years of the HPI-Stanford Design Thinking Research Program After 14 years of strong collaboration and exchange, it is with fond memories and a long list of community successes that we prepare this volume for the series Design Thinking Research. The Hasso Plattner Design Thinking Research Program, which was led by Larry Leifer, professor of mechanical engineering at Stanford University, and Christoph Meinel, professor of Internet and Web technologies and former director and CEO of the Hasso Plattner Institute (HPI), began with 13 projects in its inaugural year of 2008. Through the generous funding of the Hasso Plattner Foundation, approximately 12 projects were funded annually between 2008 and 2022—each year six at Stanford University and another six at Hasso Plattner Institute. This is the 14th comprehensive volume that we have released together covering the research studies carried out by our affiliated researchers at Stanford University and Hasso Plattner Institute. Over this period, approximately 200 projects received funding and contributed to the shared knowledge that developed in our network. At the start of our cooperation, design thinking was a relatively young concept and understood to be a method of achieving a human-centered innovation culture in multidisciplinary teams. HPI offered the truly unique opportunity to students finishing their graduate or doctoral studies to spend one or two semesters at HPI to train in the Design Thinking Basic Track or Advanced Track program. A maximum C. Meinel (✉) Hasso Plattner Institute for Digital Engineering, Potsdam, Germany e-mail: offi[email protected] L. Leifer Stanford Center for Design Research, Stanford University, Stanford, CA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_1

1

2

C. Meinel and L. Leifer

of 80 students were accepted, and on 2 days per week for an entire semester, these students worked in groups of four to six students with professors and lecturers from many diverse backgrounds—not just digital engineering or computer science. This training at HPI still exists. What has changed? In 2022, interested students can pursue coursework in design thinking at many institutions. In this way, we have achieved something wonderful. Over the last decade, design thinking has become a reality on campuses around the globe and in most workplaces. HPI and Stanford have been and remain key multipliers in achieving growth and the associated professionalization of the design thinking field. Researchers who were selected as program participants were welcomed into the community. The strategic project management techniques and deliverables have been shared at regularly offered community building workshops in Stanford and in Potsdam. This brought about a multiplier effect. Researchers presented ideas, prototypes, and results that found resonance with colleagues from other fields. These gatherings brought us all together, and we became a team. Each team member had the clear goal of advancing design thinking. Additionally, many transatlantic and collaborative research designs and numerous doctoral candidates were funded. The program without question has enhanced research quality and exchange while simultaneously creating new opportunities for talented young researchers in the field.

2 HPDTRP: First Year 2008/2009 In our first volume, Hasso Plattner shared a definition of design thinking that has been iterated over the years. In 2012, we looked to “Design Thinking as a framework to understand the issues people are experiencing in their daily lives and to generate accordingly helpful innovations for them. In Design Thinking, interdisciplinary teams set off to learn about people’s concerns and the obstacles they are facing. By means of Design Thinking, the teams look for solutions regarding the identified problems, which should be genuinely new as well as extensively useful. Thus, Design Thinking teams work toward products or services that are technically feasible, economically viable and, in addition, truly desirable for people” (Plattner, 2012). We would like to focus for a moment on one phrase: “the teams head for solutions regarding the identified problems” (Plattner, 2012). What has changed over time is that more than ever before, teams are unlearning and relearning. We tackle the design thinking process in the belief that each step must be performed with great creativity, precision, and humility. Defining the problem is crucial. In business and in academia, metrics are important, and results are meticulously measured. If the parameters of the identified problem are refined and nuanced, there can be more or better results to report because the question was properly defined. This is just one example. Today, we are essentially transmitting a return to the basics: building Design Thinking into the thought process for interdisciplinary and multidisciplinary work.

Introduction/Roadmap

3

The first volume was entitled Design Thinking: Understand – Improve – Apply. Christoph Meinel with his coauthors Tilmann Lindberg and Ralf Wagner prepared chapter one of this first volume, “Design Thinking: A Fruitful Concept for IT Development?” (Lindberg et al., 2011). The authors stated that IT experts, especially, need to consider the user-perspective because in fact it is the users that ultimately decide the success of products and services. In 2011, this perspective was further developed, as the technical perspective dominated the field at that time. There was a rich discussion on the challenges of translating design thinking into action. The conclusion was reached that “an effective design thinking strategy needs well-trained people as well as an organization that supports them instead of slowing them down” (Lindberg et al., 2011). In comments on the challenges to implementing Design Thinking, the authors wrote, “as it stands (Design Thinking exists) in contradiction to common management techniques such as stage gate innovation management that rely strongly on predefined workflows and standardized quality gates, anticipating and selecting solution paths in a restrictive manner so that explorative and creative solution paths become rather constrained” (Lindberg et al., 2011). Many of the chapters in this volume share innovative value creation. One such example is the chapter by Mathias Weske und Alexander Luebbe. The authors demonstrate how the design thinking process could optimize business processes by bringing employees into the discussion and creating prototypes that make sense to management and the core service team (Luebbe & Weske, 2011). Do not underestimate the importance of any single step. Time invested in fully developing each step has tremendous rewards. Cutting corners to save time will not provide the desired results. Time is a critical variable. The visualization that occurs in the prototyping phase can be perfected over time. In fact, the research team including Steven Dow and Scott Klemmer, at that time from Stanford University, in their chapter “The Efficacy of Prototyping under Time Constraints” focused on the fact that research teams developing multiple prototypes have a distinct advantage over research teams that work with only one prototype (Dow & Klemmer, 2011). This is one more example of why design thinking shouldn’t be rushed. Design teams often work with tight timelines, but if design teams plan accordingly—allocating sufficient time to phases—ideas can evolve and support in the development process be gained from stakeholders. A retrospective look at all that has been achieved in the Design Thinking Research Program could fill many volumes. Our most recent and final community building workshop in Potsdam provided us with the opportunity to reflect upon our experiences, our learnings, and our future trajectories in design thinking. What have we achieved? Design thinking has established itself, and our research volumes accompany this sustained shift. The types of questions our researchers ask have evolved as the social schema similarly evolves. The issues that were once painstakingly argued in 2011 can now be addressed in a footnote because of the substantial body of reference literature we now have. Our HPDTRP steering committee faced difficult decisions. Over the years, applications from many outstanding academicians have been received. At the

4

C. Meinel and L. Leifer

same time, the interdisciplinarity of funded projects has increased dramatically. This made it possible for our group to include natural sciences and social sciences and to reach conclusions that change how academia and practice view established beliefs. Is a new paradigm coming? Do current observations help us find a new paradigm that is more appropriate to explain existing phenomena? To better answer these questions, it may help to list some keywords that point to the areas where real innovation has been brought to light by our researchers on the path to a new paradigm: co-creation, innovation ecosystems, collaboration, prototyping, measurement, and neurodesign. Of course, each of these words tells a story and contains the connotation of an accretion of value creation over time. A close examination of volumes in this series show that since the outset of the program, definitions were broadened and benchmarking became more systematic; this occurred as more academic institutions, professional organizations, and industry took note of design thinking as a field and as a vehicle for mobilizing efficiency and the growth of real innovation. The overall goal of the program was (1) to discover metrics that describe innovations and creativity, predict design thinking team performance, and tell more about the value of design thinking and (2) to design tools supporting design thinking processes and teams. The Hasso Plattner-Stanford Design Thinking Research Program has achieved so much more. Our affiliated faculty guided and staked out, with their academic publications, important facets of the relatively rapid spread of design thinking around the globe. Thousands of students have completed academic coursework at HPI and at Stanford University in the area of design thinking. We have trained today’s leaders. The trajectory of design thinking should not be underestimated. Already in 2015, an article in the Harvard Business Review stated that “Design thinking is an essential tool for simplifying and humanizing. IBM and GE are hardly alone. Every established company that has moved from products to services, from hardware to software, or from physical to digital products needs to focus anew on user experience. Every established company that intends to globalize its business must invent processes that can adjust to different cultural contexts. And every established company that chooses to compete on innovation rather than efficiency must be able to define problems artfully and experiment its way to solutions” (Kolko, 2015). There are established steps of design thinking, and it is a given that innovative measures can fail at any point. Companies, governments, institutions, and all manner of stakeholders must be aware of the risk and yet still embrace this chance to move forward on the trajectory for innovation. The human-centric approach that defines design thinking requires that each design team collaborate with an open mind and progress through all phases of the process.

Introduction/Roadmap

5

3 HPDTRP: Second Year 2009/2010 In the second year of the program, research teams broadened the area of inquiry to also encompass digital tools in creative collaborative projects, ideally for teams with members located around the globe. There were many important results, which were published in our second volume, Studying Co-Creation in Practice. One tool that emerged from the research team of Christoph Meinel in year two of the program was Tele-Board, which combined white boards with videoconferencing functionalities (Gericke et al., 2012). It was a groundbreaking innovation at that time. A second team of researchers looked at the importance of the concept of “space” in the design thinking process. The study showed that design thinking teams are more creative when they are in a design thinking space than in a traditional classroom space (von Thienen et al., 2012). Much research has been done on this topic since 2012, and universities around the world now outfit spaces suitably for design thinking coursework. Some schools have even designated particular spaces for each phase in the design thinking process. Another article in the second volume examined the role of diversity in design teams in their pursuit to find out more about “team cognitive diversity;” the study sample included masters-level engineering students from nine universities in eight countries (Kress & Schar, 2012). The topic of another research team explored the importance of nonverbal information transfer as seen through the use of telepresence robots, and how this input increases design team engagement (Sirkin et al., 2012). The final chapter in the volume offered support suggestions for software developers, which build upon the basics of design thinking (Steinert & Hirschfeld, 2012).

4 HPDTRP: Third Year 2010/2011 The third volume in the series, Measuring Performance in Context focused on design thinking research topics in three different areas: design thinking research in the context of colocated teams, design thinking research in the context of distributed teams, and design thinking research in the context of embedded business teams. Of the many fruitful investigations, we will highlight three here. The team of Scott Klemmer examined prototyping dynamics to determine that it was better to Share Multiple than to Share One; study results showed that when designers Share Multiple, there is improved outcome, exploration, sharing, and group rapport (Dow et al., 2012). A second research team actually examined paradigm shift and the increasingly important role of design thinking. There was a far-reaching call for particular cognitive skills to be taught to students but no toolbox for teachers. The researchers first defined the term “twenty-first century skills,” which explained why design thinking is an appropriate enabler for teachers to prepare students for the twentyfirst-century challenges and shared an empirical study to substantiate claims

6

C. Meinel and L. Leifer

(Noweski et al., 2012). In a similar vein, the team of Bernard Roth looked at success trajectories of alumni who participated in graduate level study in design thinking. To assess alumni outcomes, the team looked particularly to “creative confidence, comfort with risk and failure, and building creative environments” (Royalty et al., 2012). Both of these projects show the importance of design thinking curricula in building human capacities for future success in tackling challenges to innovation in various settings.

5 HPDTRP: Fourth Year 2011/2012 In volume four of the Design Thinking Research, Building Innovation Eco-Systems, the work of research teams could be grouped into four general categories: the social nature of design, the need for a preservation of ambiguity, all design as redesign, and the tangibility of ideas. The interdisciplinary nature of the program grew, welcoming among other disciplines, scholars from the Department of Psychiatry at Stanford University. Allan Reiss’ team “proposed a unique experimental design to test whether creativity can be acquired or learned by an individual over time and how this relates to cognition, behavior, and the brain” (Hawthorne et al., 2014). The team practiced intervention research relating to creative capacity building, asking first if a person can improve their creative capacity and then whether it is retained using various metrics. In Riitta Katila’s team, researchers explored the relationship between design and the development of complex products and systems questioning what role user involvement (novice/average and expert/lead users) had in new product development in its early phases (Shluzas et al., 2014). Many factors were considered, including, but not limited to, perceived usability, functionality, efficiency, adaptability, and cost-benefit of product concepts (Shluzas et al., 2014). Mathias Weske and his team introduced the concept of need-finding iterations. This entails starting agile software development projects with a kickoff workshop that enables a shared understanding among stakeholders based on early-phase story prototypes. These feed into later software prototypes, which streamlines the entire process (Guentert et al., 2014). Individual stakeholders, therefore, have a concept of the big picture rather than simply one piece of the larger undertaking. Stakeholders can better project trajectories and become more efficient in each project phase.

6 HPDTRP: Fifth Year 2012/2013 One year later, the volume Building Innovators was published, with project reports from the fifth year of the program. Three overarching categories were defined for the projects in this year: assessing influential factors in design thinking, empowering team collaboration, and supporting information transfer. In TeamSense, Larry Leifer

Introduction/Roadmap

7

and his team proposed “a future sensor platform that explores how modularity and mutability affect electronics prototyping with sensors” (Sadler & Leifer, 2015). What was truly unique about this toolkit was its role in “reducing barriers to entry for rapid prototyping with sensors” and the toolkit’s relevance for many other areas of design thinking. “The need for more (1) modular and (2) mutable electronics and software components were discovered to be a limiting factor in allowing more experimentation in the early stages of sensor system prototyping. Modularity enables fixed functional blocks to be swapped in and out of a system (enabling combinations), and mutability allows modification of blocks to change their function (enabling mutation)” (Sadler & Leifer, 2015). In year one of the Design Thinking Research Program, the Tele-Board system was created. In year five of the program, the development of Tele-Board MED focused on applications in medicine to assist physicians and medical personal in meeting documentation needs and at the same time empowering the patient. “TeleBoard MED is a medical documentation system designed to support patient-doctor cooperation at eye level . . . and tackles the challenge of turning medical documentation from a necessity, which disturbs the treatment flow, into a curative process by itself” (von Thienen et al., 2015). In a third project, researchers create scenarios that blend existing models for generating innovation. “We present DT@Scrum, a process model that uses the Scrum framework to integrate Design Thinking into software development. Design Thinking activities formed the foundation of the approach. Development teams choose their respective operation mode after each sprint based on how well the requirements of the product are understood” (Häger et al., 2015). Here, researchers used design thinking to supplement scrum. In later years, other HPDTRP researchers returned to this question and also looked more carefully at whether it is possible to combine various frameworks for innovation.

7 HPDTRP: Sixth Year 2013/2014 The sixth volume of Design Thinking Research, Making Design Thinking Foundational, began with a manifesto. In this manifesto, we staked out how we are to address and measure the needs of society. “While engineering has been described as the application of science and mathematics to the needs of society, up until now we have known and taught our students very little about finding and understanding the needs of society. Thus, we have indirectly turned out half-built engineers. These were engineers who could only serve the explicit needs of others without a direct understanding of who had the need and why. With the advent of the design thinking paradigm, this situation has changed. Engineers are now taught how to engage with society through empathy training, coping with multiple points of view, actively managing teamwork, and realizing the full potential of product and service prototyping (experimenting). In effect, a critical link has been made between engineering analysis, the

8

C. Meinel and L. Leifer

science-focused part of the discipline, which is context independent, and engineering design, the human society-focused aspect of the discipline, which is context dependent” (Leifer & Meinel, 2016). With this manifesto, we succinctly stated that humancentered focus is critical to creating the full potential of engineering projects. Furthermore, this human-centered focus has been adopted by many disciplines. At this juncture, we had a toolbox with which it was possible to create metrics, meaning that results could be quantified. In addition, important steps were taken to bring about a professionalization of the field. It was in this sixth year of the program that we felt a sureness that design thinking was foundational. Our researchers’ contributions for this year fell into four basic categories. In “Tools and Techniques for Improved Team Interaction,” one topic was remote collaboration, which has become even more relevant in the last pandemic years (Wenzel et al., 2016; Shluzas et al., 2016b). A second area, “Creativity and Creative Confidence” was addressed in team spaces, in prototyping and in creating metrics (Hawthorne et al., 2016; Nicolai et al., 2016). The former team developed a novel Design Thinking Creativity Test (DTCT) in order to assess a person’s creative capacity in actual situations (Hawthorne et al., 2016). The latter team created a new qualitative method that examined the relationship between creative environments on team well-being and performance, including a careful examination of perceptions, feelings, and interactions of each team member with their respective environment (Nicolai et al., 2016). Part three looked more specifically at “Measuring Design Thinking.” One chapter in this section delved into design thinking metrics as a driver of creative innovation. The authors argued that the traditional metrics used by companies are not the best metrics to assess how well individuals and teams learn and apply design thinking (Royalty & Roth, 2016). The final section of the sixth volume covered “Documentation and Information Transfer in Design Thinking Processes.” Here, Holger Giese’s research team, in year three of funding on “Connecting Designing and Engineering Activities III,” presented a recovery approach that allowed for recovery modules to be created at particular junctures so that teams consisting of two groups of stakeholders, in this case, engineers and designers, could function more seamlessly together (Beyhl & Giese, 2016; Beyhl & Giese, 2016). The recovery modules used extraction procedures and recovery algorithms at regular intervals.

8 HPDTRP: Seventh Year 2014/2015 Volume 7, entitled Taking Breakthrough Research Home, summarized many important research undertakings for the academic year 2014 to 2015. The volume was organized in four thematic categories: design thinking in practice, exploring humantechnology interaction, prototyping and developing DT teaching, and coaching tools and approaches. In section one, the team headed by Claudia Nicolai and Ulrich Weinberg examined the process of developing design thinking workspaces. They specifically worked with one nonprofit for social entrepreneurs. Here, it should be

Introduction/Roadmap

9

noted that the workspaces that were developed were outfitted for best-use scenarios for multiple stakeholders: that is fulfilling the needs of the management while also accommodating and listening to users (Klooker et al., 2016). It should be mentioned that in this year, several projects explored humantechnology interaction in the medical field. The development of wearables allowed our design team to reach three objectives: introduce mobile wearable computing technology for distributed care, achieve increased knowledge of benefits through enhanced visualization capabilities with human augmentation, and equip clinicians with the tools needed to inform decision-making and assess behavioral changes in this group (Shluzas et al., 2016a). In our second project, patient-provider interaction is analyzed to assess how this relationship can be enhanced by means of the medical documentation system TeleBoard MED. Tele-Board MED is a collaborative eHealth application designed to support the interactions between patient and provider in clinical encounters (Perlich et al., 2016). First applications were in the field of psychotherapy. Researchers found that using Tele-Board MED increased the acceptance of diagnoses and contributed to a good team dynamic between therapist and patient (Perlich et al., 2016). In a different application area, we looked at autonomous vehicles. Using an experimental design approach, this chapter details three studies of autonomous vehicle interfaces and behaviors (Sirkin et al., 2016). In phase one, a “trustworthy” interface was developed; in phase two, haptic precues were recorded as human reactions to prototypes were charted; and phase three analyzed public reactions to the deployment of a prototype on public streets (Sirkin et al., 2016). Here, it is noteworthy that the public’s reactions to autonomous vehicles without a visible driver need to be taken into account in further prototype development. In the final section of volume seven, toolkits for teaching and coaching were developed. In the design thinking at scale project, our researchers looked at whether and how design thinking can be taught in Massive Open Online Courses (MOOCS) (Taheri et al., 2016). Creating MOOCS for scalable and sustainable peer interactions was the topic addressed by Michael Bernstein’s team. The team developed PeerStudio, which is an assessment platform that enabled rapid feedback from peers within well-attended MOOCs. In practice, students could resubmit in-progress work for peer review multiple times. Online peer interactions could, therefore, be maximized and course outcomes improved. With TalkAbout, the research team showed that rapid feedback and revision can make peer interactions sustainable at scale (Kulkarni et al., 2016).

9 HPDTRP: Eighth Year 2015/2016 Collaboration versus cooperation was a fundamental question that our projects tackled in the eighth program year. Included in this volume was a homage to John E. Arnold’s theories on creative thinking—the theoretical foundations of design thinking (von Thienen et al., 2018). Following this interesting perspective, the

10

C. Meinel and L. Leifer

volume was organized into four sections: (1) modelling and mapping teamwork; (2) tools and techniques for productive collaboration; (3) teaching, training, priming: approaches to teaching and enabling creative skills; and (4) design thinking in practice. In the first section on modelling and mapping teamwork, the authors address team interaction and performance. Scholars reflect on role distribution in design teams on multiple levels. In one project, breaks were examined and the team observed during or after breaks. Analysis showed that breaks could be grouped into one of three dimensions: activity level (active or passive), social aspect (group or individual), and distance to the project (related or unrelated to the project); the effect of each dimension on the team took a distinctive form (Dobrigkeit et al., 2018). The commonality for the contributions in Part 2 is how productive collaboration comes about. Here, we highlight the work of Michael Bernstein’s team. With the example of fiction writing, the researchers examine crowdsourcing as a technique for achieving interdependent complex goals (Kim et al., 2018). In field experiments, Mechanical Novel yielded higher-quality stories showing that complex work could be coordinated through interactive crowdsourcing workflow (Kim et al., 2018). In the third section of this volume, the research team of Erin MacDonald concluded that both implicit and explicit priming are promising techniques that can be used to enhance design skills (She et al., 2018). Researchers present two design methods that prime designers use to place more emphasis on certain stills in the conceptual design, here, for example, high immersion versus low immersion. In both cases, they use primes that require participation from the subject and sensory engagement (She et al., 2018). In the final section of this volume, we highlight a project focusing on the optimization of human-technology teamwork. Our researchers explored how design thinking could be applied to health IT systems engineering to improve the communication of levels of pain between patients and providers at Stanford Health Care, and they created early conceptual prototypes (Shluzas & Pickham, 2018).

10

HPDTRP: Ninth Year 2016/2017

Our ninth volume, published in 2019, looks at design thinking beyond solutionfixation. What does this mean exactly? In our society, most designers and managers are praised for their ability to quickly and effectively bring about a suitable resolution that essentially solves the problem. But this is not the only way or the best way. What we focus on in the Hasso Plattner Design Thinking Research Program is “the development and nurturing of a problem-oriented mindset” (Leifer & Meinel, 2019). Our natural tendency is to identify one possible solution and act on it. However, it is better to set aside time to better interrogate and reframe problems with the end user and end-user gains in mind. The goal of design thinking work is to generate meaningful and worthwhile design solutions that are sustainable (Leifer & Meinel, 2019).

Introduction/Roadmap

11

In Volume 8, we shared information on John E. Arnold and his “creative engineering theory.” Following up in Volume 9 with the second installment of the theoretical foundations of design thinking, we examined the work of one of Arnold’s successors at Stanford, Robert H. McKim. McKim first published his theory on need-based design in 1959. He shared human-centered design conceptions, assessed the roles of designers in the evolution of culture development and introduced guidelines for increasing design value (von Thienen et al., 2019). The volume is divided into three sections. In Part 1 on understanding success factors, the team led by Sheri Sheppard and Bernard Roth looked to challenge a common design thinking misconception: design thinking principles are one size fits all and thus should be implemented uniformly. This chapter shows this idea as a misconception in providing a description of design thinking that is dependent upon context. Revealed are a series of measures that can highlight both different aspects of design thinking and how these are variable in nature based on context (Royalty et al., 2019). Authors in this section also contribute to a deeper understanding of how to use space as a strategic tool. In Part 2, the authors address the digital potential of teaching, research, and organizational approaches. Here, we highlight the chapter on redesigning social organization in order to allow for fast-paced innovation. The authors started with the given that hierarchy in organizations is slowly being replaced by new, more flexible structures. By exploring the shift, they were able to redesign social organizations and the means to accomplish the associated sociological and psychological transformations (Sonalkar et al., 2019). Section three looks at design thinking in practice. Matthias Uflacker’s team, for example, explored if and how parts of the design thinking process can be combined with Lean Startup and Scrum (Dobrigkeit et al., 2019). Their team developed Innodev, which combines elements of all three of these approaches, to create an agile software development process (Dobrigkeit et al., 2019). Another noteworthy chapter prepared by Kesler Tanner and James Landay details their project, in which a meaningful design scale was created (Tanner & Landay, 2019). What makes this design scale especially attractive is that it can be created from novice comparisons (Tanner & Landay, 2019).

11

HPDTRP: Tenth Year 2017/2018

The diverse contributions to Volume 10 all address design team performance and range from approaches to interaction styles and from tools to applications. In Part 1, new approaches to design thinking are expounded upon. One team explored the use of immersive virtual reality (VR) in design thinking learning scenarios (Sonalkar et al., 2020). In this chapter, immersive VR is presented as an accompaniment to action-reflection pedagogy and yields augmented design teamwork learning and practice. The clear implication is that VR is a medium for design teamwork with potential to implement it in design education courses.

12

C. Meinel and L. Leifer

The second section of Volume 10 dedicates itself to exploring effective team interaction. A chapter to highlight here examines team creativity and looks at differences between average conversations and creative conversations. Using the Coherence Style Framework, the research team maps divergent and convergent thinking on the conversational level and sees highly creative teamwork as an alternation between local disruption (local low coherence) and global integration (global high coherence) (Menning et al., 2020). Volume 10 covers various tools to support design thinking practices; Part 3 and each chapter in this section illustrate how digital technologies are leveraged. Prototyper, for example, is a web browser-based collaborative virtual environment that supports the joint real-time creation of three-dimensional low-fidelity prototypes by 3D printing (Wenzel & Meinel, 2020). Another new tool introduced is Poirot, a web inspection tool for designers that enables them to make style edits to websites using a graphical interface (Tanner et al., 2020). The final section of the volume deals with the applications of design thinking practices. One chapter showcases the software system Tele-Board MED. The first implementation experiences of this technology in psychotherapy consultation session are captured here (Perlich et al., 2020).

12

HPDTRP: Eleventh Year 2018/2019

Volume 11, Interrogating the Doing, begins with a tribute to Robert H. McKim. In examining the theoretical foundations of design thinking, the chapter takes a closer look at McKim’s “Experiences in Visual Thinking,” which was published in 1972 and in which McKim develops a comprehensive framework of creativity as embodied and embedded cognition (von Thienen & Meinel, 2021). The volume is divided into four sections, each chapter in its own way questions how design research is best done. In section one, entitled “Effective Design Thinking Training and Practice,” chapters examine differences between theory and practice and how to bridge the gap. One chapter, which grapples with the concept of “space,” lets readers better understand, reflect, and teach space in a design thinking context and offers insights into how readers can develop individual interventions of presented tools (Schwemmle et al., 2021). The other contributions in section one enable readers to use and reuse a video-based design archiving system and also explore warm-up games in MOOC contexts, also showing how to set these games up. Section two, entitled “Understanding Design Thinking Team Dynamics,” examines the functioning of design teams in several contexts. The second chapter offers a scientific approach for the application of fNIRS hyperscanning to uncover specific qualities of team interaction. The research team assesses states of interbrain synchrony that could correlate with the behavioral states of cooperation and collaboration (Balters et al., 2021). Part 3 of the volume includes contributions that look to new approaches and new application fields. We underscore here two contributions. The first is concerned with the potential factors that influence text comprehension of

Introduction/Roadmap

13

code documents (Rein et al., 2021). In the final chapter on this section, researchers examine the tradeoffs of bodystorming techniques within human-robot interaction design (Abtahi et al., 2021). The volume concludes with two chapters in the new area of neurodesign. HPDTRP researchers, in cooperation with neuroscientists, develop new ideas and research in this up and coming field. Neurodesign is a field of study at the intersection of neuroscience research and design thinking practice, which should in fact bridge this gap. The chapters entailing neurodesign here, and the earlier chapter in this volume allow readers to understand neurocognitive processes among individual designers and teams (Auernhammer et al., 2021; von Thienen et al., 2021).

13

HPDTRP: Twelfth Year 2019/2020

Translation, Prototyping, and Measurement is the title of the twelfth volume, which is divided into three parts. In section one, research teams, each in their own way, explore the concept of translation as an analytical category: translation of design thinking to the digital sphere, the translation of concepts across cultures, an examination of interbrain synchrony, and how certain activities (physical movement and nonverbal actions) translate to action in both the in-person and virtual realms. One chapter explores the history, application, and understanding of design thinking in the Arab world, mapping out the spread of design thinking in the Arab region in several sectors (Traifeh et al., 2021). Volume 12’s second section highlights the work of design teams in creating models for prototyping. Prototyping models outlined in these chapters help the reader visualize how prototyping can meet design challenges. One research team developed PantoGuide, which is a low-cost system that provides audio and haptic guidance, via skin-stretch feedback to a user’s hand while the user explores a tactile graphic on a touchscreen. PantoGuide was developed for use in the remote classroom. Developing such prototypes for blind and visually impaired users allows these student users to learn both remotely and independently (Siu et al., 2021). Another research team designed a prototype for photographic guidance, designing new interfaces that provide contextual in-camera feedback to aid users in learning visual elements of photography with guided photography interfaces (Jane & Landay, 2021). These interfaces aid in iterating through the exploration of lighting, composition, and decluttering. Measurement in design thinking is the overarching theme for Part 3 of Volume 12. One contribution examines the relationship between the ME310 curriculum at Stanford and alumni engagement in entrepreneurship and innovation and, in so doing, attempts to understand how course-based training in design thinking can translate to professional endeavors and entrepreneurial outcomes of alumni (Sheppard et al., 2021). Another set of researchers apply design thinking to scrum. Their findings reveal that agile practitioners changed scrum to improve collaboration by incorporating teamwork, innovation, and design activities (Dobrigkeit et al.,

14

C. Meinel and L. Leifer

2021). Design thinking techniques prove to be a fruitful addition to the scrum meeting toolkit.

14

HPDTRP: Thirteenth Year 2020/2021

In Achieving Real Innovation, which is Volume 13 of the Understanding Innovation Series, readers are drawn into the thematic area with a humanistic and creative philosophy of design. In short, authors elaborate on qualities that support the conception of creative design—with the intention of developing innovators (Auernhammer et al., 2022a). Three thematic areas are defined for contributions in this volume. The first area presents value creation in virtual innovation spaces. The research team around Michael Shanks questions why ambiguity is so common in online interactions and emphasizes the aim of implementations to improve clarity and remove ambiguity in online actions (Park et al., 2022). The team addresses specific design implications that refer to technical functioning and context. In Part 2 of Achieving Real Innovation, each of the listed authors spoke to the topic of fostering innovation behavior and coevolution to determine how such behavior can be cultivated and augmented. One example of augmentation can be seen in the work surrounding the constantly developing area of program practice. The research team of Robert Hirschfeld began with the assumption that novices to programming have a high learning curve in understanding mindsets and tools (Tauemel et al., 2022). Using the idea of patterns, it was possible to stake out traditional and modern practices of exploratory programming and create a novel pattern language that focused on the typical question/response cycle that programmers follow. The problematization of design thinking as a concept is the subject of the remaining contributions in area three of the volume. In other words, this third part is concerned with building comparative frameworks. The chapters in this section help the reader visualize existing models and stake out relationships between design thinking education and societal change as well as understand the evolution of design thinking over the last 50 years. One example is the chapter dealing with the concepts of human needs. The author team sees “human needs” as an important component of design thinking and grapples with how this variable “human needs” relates to the variable (radical) innovation (von Thienen et al., 2022). By reviewing the need theories of John Arnold, Abraham Maslow, and Robert McKim, pivotal questions are raised and appraised including a questioning of the future role to be played by design thinking. Concepts such as incremental innovation (improvement within existing solutions) are discussed on the way to developing a comprehensive framework of human needs, which can be used to analyze both the risks and benefits of new products in a systematic manner (von Thienen et al., 2022). Another chapter of note concerns the evolution of design, with the research team staking out the evolution of multiple human-centered design approaches over the last 50 years (Auernhammer et al., 2022b).

Introduction/Roadmap

15

The last chapter in the volume argues that design thinking is a useful, appropriate, and necessary approach to support collective action in sustainability issues (Ardoin et al., 2022). The research team, headed by Nicole Ardoin, seeks to understand the relationship between design thinking and sustainability. Specifically, the team grapples with whether and how sustainability issues can be reframed as broader societal considerations (Ardoin et al., pp. 325–340). Included in this brief review of the Understanding Innovation Series, the reader is not only introduced to the ongoing research undertaken at HPI and Stanford University in the Hasso Plattner Design Thinking Research Program but also the research reports for groundbreaking research that is still in progress. The research is intended as inspiration for all readers, regardless of professional or academic background.

15

HPDTRP: Fourteenth Year 2021/2022—The Recent Volume

In this last year of the Hasso Plattner Design Thinking Research Program, the chapters prepared by the research teams summarized the current state of their projects. Although the formal program is ending, the relationships that developed are not and neither are the research trajectories. Over the years of the program, we have seen that projects funded have become more diverse including many different faculty departments at Stanford University. Design thinking has also grown exponentially from where we started many years ago. Design thinking education is now common on many college campuses. We begin this volume with a contribution from the team of Sheri Sheppard that summarizes findings from interview-based studies that track career pathways of female graduates. Her research group analyzed a cohort of students at Stanford University, who participated in the foundational course ME310 many years ago to determine the types of careers the group pursued and collected feedback on what role the design thinking principles imparted in ME310 have played in career path trajectories of this group. Her team assessed what an education in design thinking really means on the individual level and developed an Academic Workplace Relational (AWR) Model to explain bidirectional relationships and to improve project-based learning. This chapter helps us visualize the longterm implications of a design thinking mindset. Contributions to the volume were divided into four parts. Part I is concerned with the application of design thinking to governance and social causes. Within this issue area, the demands for transformation, digitization, and agility have increased dramatically over the last few years. In this part, HPDTRP researchers examine design work in a number of contexts, including governance and sustainability. Teams dissect governmental practice and discern how it is possible to leverage success in outcomes, examine existing barriers, and stake out programmatic changes for facilitating (virtual) design thinking. The first contribution from the team of

16

C. Meinel and L. Leifer

Christoph Meinel and Julia von Thienen looked at governance approaches. Their understanding of governance includes the encouragement or discouragement of projects and how this impacts creativity. They introduce a four-leaf clover model in design thinking and are thereby able to quantify the impact of governance. They argue that environment is an important factor impacting how people feel and behave. The second chapter from the project team of Christoph Meinel is concerned with agile governance in Rwanda and seeks to answer the question as to what agile means in the public sector. Much is still left to be learned about agile governance or agility in the governmental context. Using the example of the one-stop shop in Rwanda, the team brings in three variables to provide a deeper understanding of agile governance: technology, innovation, and organization. The final chapter in this section is from the team of Nicole Ardoin. The team conducted 20 interviews with participants from nine countries to gather support for their supposition that design thinking is good for generating sustainability solutions. By identifying and exploring five characteristics of design thinking, they gather support for using design thinking as a tool in sustainability work. The five characteristics identified were inspiring creativity; participatory and people-focused; diversity in thought and action; holistic, systems thinking mindset; and streamlined, action-oriented approach. In fact, many professionals in the field already employ design thinking. Part II two covers new prototypes that were developed by three HPDTRP teams. The wide-ranging goals included enhancing prototypes with a conversational agent, understanding privacy design frameworks, and designing strategies for live programming. Perceived problems were clearly explained, and the designed prototypes bridge gaps in existing practice. The Stanford team headed by James Landay examined user perceptions of privacy interfaces in the workplace. The chapter covers implications for the design of privacy interfaces and informs large-scale studies on privacy attitudes in the workplace. Of course data privacy is an important concern. The team established that there is a negative privacy attitude when deployment scenarios were company-wide and when there was a negative relationship with the manager. James Landay and Matthew Joerke explore two technical privacy approaches: differential privacy and K-anonymization. In light of their research questions regarding owner and valence, they prototyped four user interfaces for different privacy design frameworks. The second chapter prepared by Rebecca Currano and David Sirkin explores how design learning can be improved through reflective hands-on engagement and through spoken conversation. Their two-part hypothesis describes the possibility of improving design when there is conversation between the student designer and prototyping materials and, therefore, that it is possible to enhance their prototype with a conversational agent. Robert Hirschfeld’s research group tackled design strategies for a liveprogramming system. The system should architect communication patterns that reward changes with immediate effects. They introduce the “Squeak/Smalltalk” prototype that enables programmers in a live system to see the system adapt

Introduction/Roadmap

17

immediately. This is a user-centered innovation and the programmer sees immediate “output.” In Part III, Enhancement Through Design Thinking, contributions tackle the question of how to foster and augment innovation behavior. The researchers identify perceived problems or target the creation of novel approaches to augment existing practice. We begin this section with a chapter from Bernie Roth and Jan Auernhammer that answers the question: what is design thinking from different perspectives? The comprehensive chapter covers methodology, designer thinking, and design practice. The second chapter from the team of Manish Saggar explores the intersection of neuroscience and design. The authors outline opportunities and challenges and map the importance for positioning this intersection in the development of a new formal area of study. The following chapter, from the team of Allan Reiss, looks at women entrepreneurs and their pitch performance to examine the impact of interbrain synchrony on investment decisions. Their study, which included 40 entrepreneur-investors, assessed which interbrain signatures corresponded to successful versus unsuccessful pitches. In an effort to catalyze the success of women entrepreneurs, the team developed a neuroscience approach that could help match startups with investors. This would be achieved through leveraging fNIRS hyperscanning to measure interbrain dynamics of an entrepreneur-investor dyad. The team studied how and when neural processes become synchronized and how this relates to behavioral measures of interaction. In a second chapter from the team of Allan Reiss, the relationship between interpersonal closeness, team performance, and creativity outcomes is examined. The researchers conclude that there is increased interpersonal closeness in tests performed with near-infrared spectroscopy neuroimaging on teams that worked with design thinking activities as compared to the control groups. Furthermore, they determined that there is a distinct interbrain signature in the right frontocortical region linked to design thinking, which is different for in-person and virtual groups. In this section’s final chapter, the team of Hazel Rose Markus explores how culture shapes designers’ emotion in creative problem-solving activities. Two survey studies reveal the interplay of affect, culture, and idea generation. This chapter represents an important contribution to the emerging field of emotion research. A stated goal is to enhance creative experience and enable cross-cultural collaboration in creative work. The fourth part of this volume covers best practices and strategy in design thinking. The contributions of five author teams are concerned with creating best practices and toolkits for effective implementation of design thinking. Also included in this study are elements of forward-thinking strategy. The team of Christoph Meinel and Annie Kerguenne examined how design thinking informs digital transformation activities. A tangible output was the development of an integrated digital transformation process to create an adaptive digital transformation strategy kit. This strategy kit would ideally manage digital transformation endeavors.

18

C. Meinel and L. Leifer

Falk Uebernickel’s group concluded that companies using human-centered digital innovation are better at digital transformation. In the contribution by Carolin Marx, a three-layered framework was developed to describe how companies could save an initiative from failure; in essence, how companies can come from behind out of a trap. This selection describes how to implement mitigation strategies. These mitigation strategies most likely represent human-centered digital innovation. The output is an integrative trap framework for digitally transforming organizations. The team of Ulrich Weinberg and Claudia Nicolai shared facets of hybrid education. The chapter covers a taxonomy of time- and place-based hybridity as well as the opportunities and challenges of teaching settings. It also imparts awareness and inspiration for teaching in hybrid settings. The chapter “Design Thinking Transfer Gap: Differences between Knowledge and Application of Design Thinking in the Organizational Environment” in this part seeks a solution to the design thinking transfer gap in its exploratory research to determine the role of individual factors in design thinking knowledge and application in a company. While it is noteworthy that employee knowledge of design thinking is high, this is not reflected in the application of design thinking in the workplace. This chapter explores the organizational climate for innovation and employee capabilities. Methods to bridge the gap are introduced. The final chapter of the volume from the team of Jonathan Edelman, Joaquin Santuber, and Babajide Owoyele breaks down the “designing as performance” approach. The stated goal is to bridge design research and practice. This chapter provides a genealogy of designing as performance, which consists of five steps: accessing highly effective performative patterns; designing as performance for bridging the gap between research and practice in design thinking education; designing as performance in digital product development settings; designing for value creation: principles, methods, and case insights; and beyond brainstorming: introducing MEDGI, an effective, research-based method for structured concept development. We hope that the contributions to this volume are valuable to your design thinking undertakings. As previously stated, interest in design thinking has grown dramatically year-by-year. With many new initiatives launched by Stanford, HPI, and the Hasso Plattner Foundation, we look forward to what is still ahead.

16

Design Thinking Today Around the World

October 2022 marked the official opening of the Hasso Plattner School of Design Thinking in Africa at the University of Cape Town. The six-star green-rated academic building is the first building in Africa architecturally created to be a design school and is already the center of excellence for design thinking on the continent (https://dschoolafrika.org/our-new-home/). The opening took place as part of the d. confestival, hosted for the first time in Africa with international participants. The d. confestival is a combination of academic conference and a festival due to its creative

Introduction/Roadmap

19

and interactive aspects. The regular gatherings of design practitioners and their academic counterparts at d.confestivals and annual conferences of the Global Design Thinking Alliance (GDTA) help us understand what a thriving community exists in the field. It is also important to underline the role of the GDTA in fostering dialogue among academic institutions teaching design thinking around the globe. Ulrich Weinberg, Director and Professor of the HPI D-School, has served as founding President of the GDTA and built a strong international network. Starting with ten institutions in 2017, the initial goals were identified, and faculty came together regularly to discuss joint research projects, plan conferences, and develop joint educational opportunities. By 2023, there are now approximately 30 members who meet monthly to participate in the GDTA Spotlight sessions. Representatives from each member institution take turns in updating the community on what is happening at their home institution in the area of design thinking and hold an open discussion. Sharing best practices is just one way that these design thinking thought leaders can stay abreast of new directions in the field and hone individual campus implementation strategies. There has been interaction between the group of HPDTRP researchers and the GDTA community. Ulrich Weinberg and Claudia Nicolai, Academic Director and co-head of the HPI School of Design Thinking, have been selected for funding in the HPDTRP several years running and providing a valuable link to the GDTA community. The regularly offered GDTA Spotlight sessions yield additional synergies between these groups. There is also a vibrant web forum for contributions from the HPDTRP community and international academics—The Electronic Colloquium on Design Thinking Research: https://ecdtr.hpi-web.de/. This forum is a space for uploading papers, short notes, and surveys that are of interest to the design thinking research community. Science communication is promoted, and this forum is a meeting ground for researchers with innovative ideas to solve design thinking research challenges. Individuals bring their own unique creativity to the table in design thinking. Over the years, design thinking has been primarily offered at academic institutions for college students and adult learners. By 2022, we have realized that design thinking could be and should be introduced at a much earlier age. The HPI d-school is currently working on initiatives in this area. It should also be noted that in the Berlin-Brandenburg region, there have been initiatives to implement design thinking in the school system (https://berliner-ideenlabor.de/kidskreatethefuture). Why start so early? Just imagine what will be possible when children and youth are able to help design their own schools and become part of the feedback loops that bring innovation and change. As design thinking gains traction in primary and secondary educational institutions, a generation will be encouraged to question, relearn, and unlearn as a matter of course. The overall goal of the HPDTR program was to discover metrics that predict design thinking team performance. Impressive results have been achieved. There are two special themes that remained important over the years of our program. First is the exploration of design thinking in the areas of Information

20

C. Meinel and L. Leifer

Technology and IT Systems Engineering. Second is to explore and enhance the neurocognitive foundations of design thinking. Our researchers grappled with many questions, for example: • What is design? How do practitioners think and create in design innovation? How do we create new frameworks, tools, toolkits, and entire systems that can improve old practices, encapsulate current practices, and reuse successful practices? • Human, business, and technology performance are central to design thinking, but how do we understand their interrelationship? It is essential for the frameworks, tools, toolkits, and systems that are created to be applied at the right moment. Nevertheless, sometimes there is failure. How do these design instances fail? The conclusion of the 14-year research program is not the end. Alumni of the program remain networked. The many joint documents and publications that resulted during the academic exchange stake out the important territory that has been covered. Numerous prototypes have been built and gained exposure, which in turn encourages other researchers in the field to expand upon these dynamic project deliverables. The research endeavors that have developed over the years build new networks of exchange between our institutions. As program alumni move on to take up faculty positions at academic institutions around the world and embark upon new collaborations, our network continues to grow. We hope that this volume inspires you to evaluate design thinking and your role in multiple professional and personal contexts. Responses to the texts may be addressed directly to the authors. The scholars who have contributed to this volume each represent excellence in their individual research areas. Design thinking will continue to grow in importance, and the breadth and depth of ideas shared in the Understanding Innovation series will help us all to navigate the future. Acknowledgments We thank all authors for sharing their research results in this publication. Our special thanks go to Dr. Sharon Therese Nemeth for her constant support in reviewing the contributions and to Jill Grinager for the publication’s project management.

References Abtahi, P., Sharma, N., Landay, J. A., & Follmer, S. (2021). Presenting and exploring challenges in human-robot interaction design through bodystorming. In Interrogating the doing (pp. 327–345). Ardoin, N. M., Bowers, A. W., Lin, V., & Phukan, I. (2022). Design thinking as a catalyst and support for sustainability solutions. In Achieving real innovation (pp. 325–340). Auernhammer, J., Sonalkar, N., & Saggar, M. (2021). NeuroDesign: From neuroscience research to design thinking practice. In Interrogating the doing (pp. 347–356). Auernhammer, J., Meinel, C., Leifer, L., & Roth, B. (2022a). Humanistic and creative philosophy. In Achieving real innovation (pp. 1–17). Auernhammer, J., Zallio, M., Domingo, L., & Leifer, L. (2022b). Facets of human-centered design: The evolution of designing, by, with and for people. In Achieving real innovation (pp. 227–246).

Introduction/Roadmap

21

Balters, S., Mayseless, N., Hawthorne, G., & Reiss, A. L. (2021). The neuroscience of team cooperation versus team collaboration. In Interrogating the doing (pp. 203–218). Beyhl, T., & Giese, H. (2016). Connecting designing and engineering activities III. In Making design thinking foundational (pp. 265–290). Dobrigkeit, F., de Paula, D., & Uflacker, M. (2018). Breaks with a purpose: A three-dimension framework to map break characteristics and their effects on design thinking teams. In Making distinctions: Collaboration versus cooperation (pp. 59–78). Dobrigkeit, F., de Paula, D., & Uflacker, M. (2019). InnoDev – A software development methodology integrating design thinking, scrum, and lean startup. In Looking further: Design thinking beyond solution-fixation (pp. 199–228). Dobrigkeit, F., Matthies, C., Teusner, R., & Perscheid, M. (2021). Joining forces: Applying design thinking techniques in scrum meetings. In Interrogating the doing (pp. 315–346). Dow, S. P., & Klemmer, S. R. (2011). The efficacy of prototyping under time constraints. In Design thinking: Understand – improve – apply (pp. 111–130). Dow, S., Fortuna, J., Schwartz, D., Altringer, B., Schwartz, D., & Klemmer, S. (2012). Prototyping dynamics: Sharing multiple designs improves exploration, group rapport, and results. In Measuring performance in context (pp. 47–70). Gericke, L., Gumienny, R., & Meinel, C. (2012). Tele-board: Follow the traces of your design process history. In Studying co-creation in practice (pp. 15–30). Guentert, M., Luebbe, A., & Weske, M. (2014). Sharing knowledge through tangible models: Designing kickoff workshops for agile software development projects. In Building innovation eco-systems (pp. 203–218). Häger, F., Koward, T., Krüger, J., Vetterli, C., Übernickel, F., & Uflacker, M. (2015). DT@scrum: Integrating design thinking with software development processes. In Building innovators (pp. 263–289). Hawthorne, G., Quintin, E. M., Saggar, M., Bott, N., Keinitz, E., Liu, N., Chien, H., Hong, D., Royalty, A., & Reiss, A. (2014). Impact and sustainability of creative capacity building: The cognitive, behavioral, and neural correlates of increasing creative capacity. In Building innovation eco-systems (pp. 65–78). Hawthorne, G., Saggar, M., Quintin, E.-M., Bott, N., Keinitz, E., Liu, N., Chien, Y.-H., Hong, D., Royalty, A., & Reiss, A. (2016). Designing a creativity assessment tool for the twenty-first century: Preliminary results and insights from developing a design-thinking based assessment of creative capacity. In Making design thinking foundational (pp. 111–124). Jane, L.E., & Landay, J. A. (2021). Designing photography guidance for rapid in-camera iteration. In Interrogating the doing (pp. 151–166). Kim, J., Sterman, S., Cohen, A. A. B., & Bernstein, M. S. (2018). Mechanical novel: Crowdsourcing complex work through reflection and revision. In Making distinctions: Collaboration versus cooperation (pp. 79–104). Klooker, M., Nicolai, C., Matzdorf, S., Trost, A., von Schmieden, K., & Böttcher, L. (2016). On creating workspaces for a team of teams: Learnings from a case study. In Taking breakthrough research home (pp. 67–86). Kolko, J. (2015). Design thinking comes of age. In Harvard Business Review (pp. 66–71). Accessed 12 October 2022, from https://hbr.org/2015/09/design-thinking-comes-of-age Kress, G., & Schar, M. (2012). Teamology – The art and science of design team formation. In Studying co-creation in practice (pp. 189–210). Kulkarni, C., Kotturi, Y., Bernstein, M., & Klemmer, S. (2016). Designing scalable and sustainable peer interactions online. In Taking breakthrough research home (pp. 237–273). Leifer, L., & Meinel, C. (2016). Manifesto: Design thinking becomes foundational. In Making design thinking foundational (pp. 1–4. Leifer, L., & Meinel, C. (2019). Looking further: Design thinking beyond solution-fixation. In Looking further: Design thinking beyond solution-fixation (pp. 1–12. Lindberg, T., Meinel, C., & Wagner, R. (2011). Design thinking: A fruitful concept for IT development? In Design thinking: understand – improve – apply (pp. 3–18).

22

C. Meinel and L. Leifer

Luebbe, A., & Weske, M. (2011). Bringing design thinking to business process modeling. In Design thinking: Understand – improve – apply (pp. 181–196). Menning, A., Ewald, B., Nicolai, C., & Weinberg, U. (2020). Team creativity between local disruption and global integration. In Investigating design team performance (pp. 133–142). Nicolai, C., Klooker, M., Panayotova, D., Hüsam, D., & Weinberg, U. (2016). Innovation in creative environments: Understanding and measuring the influence of spatial effects on design thinking-teams. In Making design thinking foundational (pp. 125–140). Noweski, C., Scheer, A., Büttner, N., von Thienen, J., Erdmann, J., & Meinel, C. (2012). Towards a paradigm shift in education practice: Developing twenty-first century skills with design thinking. In Measuring performance in context (pp. 71–94). Park, S. Y., Whiting, M. E., & Shanks, M. (2022). Dancing with ambiguity online: When our online actions cause confusion. In Achieving real innovation (pp. 37–56). Perlich, A., von Thienen, J., Wenzel, M., & Meinel, C. (2016). Redesigning medical encounters with tele-board MED. In Taking breakthrough research home (pp. 101–124). Perlich, A., Steckl, M., von Thienen, J., Wenzel, M., & Meinel, C. (2020). Getting hands-on with tele-board MED: Experiencing computer-supported teamwork in therapist-patient sessions. In Investigating design team performance (pp. 255–272). Plattner, H. (2012). Preface. In Studying co-creation in practice (pp. v-vi). Rein, P., Taeumel, M., & Hirschfeld, R. (2021). Towards a theory of factors that influence text comprehension of code documents. In Interrogating the doing (pp. 307–326). Retrieved October 13, 2022., from https://berliner-ideenlabor.de/kidskreatethefuture Royalty, A., & Roth, B. (2016). Developing design thinking metrics as a driver of creative innovation. In Making design thinking foundational (pp.171–186). Royalty, A., Oishi, L., & Roth, B. (2012). I use it every day: Pathways to adaptive innovation after graduate study in design thinking. In Measuring performance in context (pp. 95–109). Royalty, A., Chen, H., Roth, B., & Sheppard, S. (2019). Measuring design thinking practice in context. In Looking further: Design thinking beyond solution-fixation (pp. 61–74). Sadler, J., & Leifer, L. (2015). TeamSense: Prototyping modular electronics sensor systems for team biometrics. In Building innovators (pp. 87–100). Schwemmle, M., Nicolai, C., & Weinberg, U. (2021). Using ‘space’ in design thinking: Concepts, tools and insights for design thinking practitioners from research. In Interrogating the doing (pp. 123–146). She, J., Seepersad, C. C., Holtta-Otto, K., & MacDonald, E. F. (2018). Priming designers leads to prime designs. In Making distinctions: Collaboration versus cooperation (pp. 251–274). Sheppard, S. D., Chen, H. L., Toye, G., Kempf, F., & Elfiki, N. (2021). Measuring the impact of project-based design engineering courses on entrepreneurial interests and intentions of alumni. In Interrogating the doing (pp. 297–314). Shluzas, L. A., & Pickham, D. (2018). Human technology teamwork: Enhancing the communication of pain between patients and providers. In Making distinctions: Collaboration versus cooperation (pp. 313–326). Shluzas, L. A., Steinert, M., & Katila, R. (2014). User-centered innovation for the design and development of complex products and systems. In Building innovation eco-systems (pp. 135–152). Shluzas, L. A., Aldaz, G., Pickham, D., & Leifer, L. (2016a). Design thinking in health IT systems engineering: The role of wearable mobile computing for distributed care. In Taking breakthrough research home (pp. 87–100). Shluzas, L. A., Aldaz, G., & Leifer, L. (2016b). Design thinking health: Telepresence for remote teams with Mobile augmented reality. In Making design thinking foundational (pp. 53–66). Sirkin, D., Ju, W., & Cutkosky, M. (2012). Communicating meaning and role in distributed design collaboration: How crowdsourced users help inform the design of telepresence robotics. In Studying co-creation in practice (pp. 173–188).

Introduction/Roadmap

23

Sirkin, D., Baltodano, S., Mok, B., Rothenbücher, D., Gowda, N., Li, J., Martelaro, N., Miller, D., Sibi, S., & Ju, W. (2016). Embodied design improvisation for autonomous vehicles. In Taking breakthrough research home (pp. 125–145). Siu, A. F., Chase, E. D. Z., Kim, G. S-H, Boadi-Agyemang, A., Gonzalez, E. J., & Follmer, S. (2021). Haptic guidance to support design education and collaboration for blind and visually impaired people. In Interrogating the doing (pp. 167–180). Sonalkar, N., Mabogunje, A., & Leifer, L. (2019). Redesigning social organization for accelerated innovation in the new digital economy: A design thinking perspective. In Looking further: Design thinking beyond solution-fixation (pp. 143–157). Sonalkar, N., Mabogunje, A., Miller, M., Bailenson, J., & Leifer, L. (2020). Augmenting learning of design teamwork using immersive virtual reality. In Investigating design team performance (pp. 67–76). Steinert, B., & Hirschfeld, R. (2012). Applying design knowledge to programming. In Studying co-creation in practice (pp. 259–277). Taheri, M., Unterholzer, T., & Meinel, C. (2016). “Design thinking at scale: A report on best practices of online courses. In Taking breakthrough research home (pp. 217–236). Tanner, K., & Landay, J. (2019). “I know it when I see it:” how experts and novices recognize good design. In Looking further: Design thinking beyond solution-fixation (pp. 249–266). Tanner, K., Johnson, N., & Landay, J. A. (2020). Poirot: A web inspector for designers. In Investigating design team performance (pp. 229–253). Tauemel, M., Lincke, J., Rein, P., & Hirschfeld, R. (2022). A pattern language of an exploratory programming workspace. In Achieving real innovation (pp. 111–146). Traifeh, H, Refaie, R. A., von Thienen, J., von Schmieden, K., Mayer, L., Osman, S., & Meinel, C. (2021). Mapping design thinking in the Arab world. In Translation, prototyping, and measurement (pp. 41–60). von Thienen, J. P.A., & Meinel, C. (2021). Theoretical foundations of design thinking part III: Robert H. McKim’s visual thinking theories. In Interrogating the doing (pp. 9–73). von Thienen, J., Noweski, C., Rauth, I., Meinel, C., & Lang, S. (2012). If you want to know who you are, tell me where you are: The importance of places. In Studying co-creation in practice (pp. 53–74). von Thienen, J., Perlich, A., & Meinel, C. (2015). Tele-board MED: Supporting twenty-first century medicine for mutual benefit. In Building innovators (pp. 101–130). von Thienen, J. P.A., Clancey, W. J., Corazza, G. E., & Meinel, C. (2018). Theoretical foundations of design thinking. In Making distinctions: Collaboration versus cooperation (pp. 13–41. von Thienen, J. P.A., Clancey, W. J., & Meinel, C. (2019). Theoretical foundations of design thinking part II: Robert McKim’s need-based design theory. In Looking further: Design thinking beyond solution-fixation (pp. 13–39). von Thienen, J., Szymanski, C., Santuber, J., Plank, I., Rahman, S., Weinstein, T., Bauer, M., Meinel, C. (2021). Neurodesign – Insights from practice. In Interrogating the doing (pp. 357–425). von Thienen, J., Hartmann, C., & Meinel, C. (2022). Different concepts of human needs – or: Is there a tension between the design thinking focus on needs and aspirations to radical innovation? In Achieving real innovations (pp. 209–226). Wenzel, M., & Meinel, C. (2020). Prototyper: A virtual remote prototyping space. In Investigating design team performance (pp. 171–184). Wenzel, M., Gericke, L., Thiele, C., & Meinel, C. (2016). Globalized design thinking: Bridging the gap between analog and digital for browser-based remote collaboration. In H. Plattner, C. Meinel, & L. Leifer (Eds.), Making design thinking foundational (pp. 15–34). Springer.

Decades of Alumni: Perspectives on the Impact of Project-Based Learning on Career Pathways and Implications for Design Education Sheri D. Sheppard, Helen L. Chen, George Toye, Aya Mouallem, Micah Lande, Lauren Shluzas, Timo Bunk, Nada Elfiki, Johannes J. L. Lamprecht, and Katharina Prantl

Abstract This chapter summarizes four interview-based studies exploring the impact of two graduate-level courses in mechanical engineering at Stanford University on the innovative, entrepreneurial, and collaborative capacities of alumni and, in particular, the innovative career pathways of female graduates. The research findings are situated in two frameworks: (1) the social cognitive career theory (SCCT), a well-established model of how basic academic and career interests develop and how academic and career success is obtained, and (2) the academic-workplace relational (AWR) model, a new model developed to describe the many bidirectional relationships observed between university and workplace settings. Finally, the continuing research efforts identifying how project-based learning prepares individuals for career success and how project-based learning can be improved and strengthened are outlined.

1 Introduction From 2018 to 2022, our research team has studied the “career diaspora” of engineering graduates from immersive project-based learning experiences. We do this by identifying elements from those educational experiences that contribute to S. D. Sheppard (✉) · H. L. Chen · G. Toye · L. Shluzas · T. Bunk · N. Elfiki · J. J. L. Lamprecht · K. Prantl Department of Mechanical Engineering, Stanford School of Engineering, Stanford, CA, USA e-mail: [email protected]; [email protected]; [email protected]; [email protected]; [email protected] A. Mouallem Department of Electrical Engineering, Stanford University, Stanford, USA e-mail: [email protected] M. Lande Design & Engineering Education, South Dakota School of Mines & Technology, Rapid City, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_2

25

26

S. D. Sheppard et al.

professional and career capabilities, particularly around innovation and entrepreneurial endeavors. Our test beds were two course sequences (ME 218 and ME 310) offered in the graduate program in mechanical engineering at Stanford University. In 2020 and 2021, surveys were administered to graduates of these two sequences from the last 25 years, exploring their memories and recollections of the courses, the details of their job choices and postgraduate plans, descriptions of their work behaviors, and self-assessments of their capabilities related to engineering, entrepreneurship, design, and innovation. Our research approach involved more than just surveys; we used a mixed methods approach, which also included semi-structured interviews of course graduates. We have based our publications and workshops to date largely on an analysis of survey data from 25 years of alumni from these two courses. In Sheppard et al. (2021a), we described the process of designing the ME 310 survey instrument and developing different survey items to address the research areas related to impactful course elements, career steps and work tasks, and self-efficacy assessments. Invitations to complete the survey were sent in July 2020 to ME 310 graduates from the class of 1992–1993 through the class of 2016–2017, and 301 alumni participated (41% response rate). The survey data indicated an overwhelmingly positive course assessment by the alumni, with over 90% rating their course experience with fellow Stanford teammates and global partners as positive or rewarding. Furthermore, we evaluated the respondents’ self-assessments on innovation self-efficacy (ISE), engineering task self-efficacy (ETSE), entrepreneurial self-efficacy (ESE), and design thinking self-efficacy (DTSE) via statistical analysis. Overall, we found that the ME 310 alumni respondents reported very high self-confidence across all four selfefficacy measures. Sheppard et al. (2021b) expanded upon the preliminary analyses of ME 310 data provided in Sheppard et al. (2021a). While there is no single dominant career pathway that ME 310 graduates pursue, we were able to classify and analyze the ME 310 graduates’ experiences via three analytical lenses: leadership responsibility, organizational role type, and presence of research and development (R&D) and/or design in their job function. A comparative analysis of these three subgroups revealed that greater ESE and innovative behavior were exhibited by graduates with more leadership responsibilities and/or who were a current founder. Furthermore, those whose current job involved R&D and design, and whose first job had involved at least one of these functions, exhibited greater ETSE, but their ISE or ESE scores were no greater than any of the other groups considered. In contrast, graduates whose first and current jobs did not include any R&D and/or design functions exhibited the lowest innovative behavior. More recently, Sheppard et al. (2022) described the larger mixed methods study that examined the longer-term impact of project-based learning in ME 218 and ME 310 on graduates’ career plans and pathways (Note: By the end of 2021, we collected quantitative data from ME 218, as well as ME 310 graduates, via surveys tailored to the specific course experience). In addition, Sheppard et al. (2021a) documented how we opportunistically designed complementary qualitative studies so that we could probe more deeply into the alumni experiences. These semi-structured interviews with selected alumni from both courses have allowed for triangulation across

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . .

27

datasets. We collectively designed these qualitative and quantitative research instruments and participant foci to address the following research questions: • RQ1: What career paths have graduates from immersive engineering design courses pursued? • RQ2: What are the alumni’s attitudes and perspectives on the various components of these types of design courses? • RQ3: How might course experiences influence graduates’ career paths, especially those related to being innovative and/or entrepreneurial? • RQ4: What can we learn from listening to alumni about the effectiveness of education more generally? What improvements for engineering design education do alumni suggest? In this chapter, Sect. 2 outlines the two foundational frameworks for our research followed by Sect. 3, which describes the features and characteristics of the ME 218 and ME 310 course sequences and graduates’ career pathways (as well as our justification for considering both sequences in our study). In Sect. 4, we summarize findings from four semi-structured interview-based studies of ME 218 and ME 310 graduates to identify the course elements that contributed to graduates’ entrepreneurial and innovative career success. We look at what alternate, and perhaps surprising, career paths graduates have taken and how these course experiences have prepared graduates for job changes in unexpected (e.g., COVID-19) circumstances.

2 Underlying Theoretical Frameworks 2.1

Social Cognitive Career Theory (SCCT)

Underpinning our research is the social cognitive career theory (SCCT) developed by Lent, Brown, and Hackett (1994) and Lent and Brown (2006). The work we have done investigates how the attitudes, interests, goals, and performance of engineering students and graduates relate to innovation work and “self-efficacy.” We follow Bandura’s definition of self-efficacy as one’s own beliefs in their ability to perform a specific task or action (1986). Our exploration focuses on how the factors within the SCCT model contributed to individuals’ decisions about academic and career intentions (Gilmartin et al., 2017), namely, the self-efficacy measures emphasizing innovation (Schar et al., 2017a, 2017b), entrepreneurship (De Noble et al., 1999), and design thinking (Schar, 2020). Figure 1 highlights how the key areas of the SCCT model align with our research on graduates of immersive project-based learning experiences.

2.2

Academic-Workplace Relational (AWR) Model

Academic institutions such as colleges and universities are connected to the larger world of work. Sometimes this connection is modeled in terms of a pipeline, with

28

S. D. Sheppard et al.

Fig. 1 Mapping the ME 218 and ME 310 course sequences and self-efficacy measures to the elements of social cognitive career theory (SCCT), modified from Lent et al. (1994, p. 94)

students “flowing” through an increasingly restrictive series of pipes that imply that along the way, students “leak out” before entering the world of work (Hammonds, 2016). Another model of this phenomenon that is increasingly favored and is arguably more student- and human-centric is a pathways model where students navigate and explore several possible areas of interest and future directions (National Academies of Engineering, 2018; Main et al., 2021). Students move through the academy and then into the working world, oftentimes on well-marked “trails.” Where both the pipeline and pathways models fall short is that they model the academic and working world connections as one way (i.e., students move or flow out of the academy and into the working world) and portray it as transactional (i.e., the primary function of the academy is to supply workers). Our research has been built as an observational model, seeing the connection between the academy and the working world as a series of bi- or multidirectional and evolving relationships. For example, these relationships may be about codeveloping knowledge, adapting and adopting technologies created in the working world into academic practices (or vice versa), leveraging and growing networks formed during one’s student days into professional practice support systems, strengthening career shifts between the academy and the working world, or concurrently having responsibilities and engagement in both. We illustrate the richness of these relationships in the academic-workplace relational model (AWR) shown in Fig. 2. This type of model acknowledges a nonlinear, bidirectional, collaborative, and evolving world where the academy and the workplace are explicitly connected. Neither the academy nor the working world exists in a vacuum. Furthermore, their relationships are not just transactional, with students “outflowing” into industry and “money for research” flowing into the academy. The AWR model represents the philosophy behind our research design investigating these possible relationships as students graduate from the academy and move into the working world. We chose to focus on two courses that have developed into a variety of workplace relationships (as described in Sect. 3.2) and whose many

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . .

29

Fig. 2 The academicworkplace relational (AWR) model

graduates over the last 25 years have utilized their formal education in a wide array of settings (as described in Sect. 3.3). These courses and their over 2000 graduates are “fertile ground” for exploring in new ways the relationships between the academy and the working world.

3 Context for the Four Interview Studies The four interview studies presented in this chapter consider graduates from two long-running project-based course sequences: ME 218 and ME 310. In Sect. 3.1, we describe the ways in which ME 218 and ME 310 are similar and how the courses are different. These arguments provide a justification, in part, for why the two course sequences are considered together in a common research project. Next, in Sect. 3.2, we view the courses through the academic-workplace relational model, illustrating similar histories of interconnectedness of the courses and their graduates to the larger working world. Finally, in Sect. 3.3, we consider and compare the career roles pursued by graduates of the two sequences as another part of our justification for considering both courses (and their graduates) in a common research project.

3.1

ME 218 and ME 310: Similarities and Differences

ME 218 and ME 310 have much in common. They are both yearlong graduate-level course sequences that fulfill the depth requirement for students pursuing their mechanical engineering master’s degree in design at Stanford University. Both courses immerse students in hands-on project-based learning. They share other attributes, such as explicitly promoting a sense of community among students and

30

S. D. Sheppard et al.

alumni by engaging students in teamwork, cohort collaborations, and engaging alumni as coaches. Both courses require intense time commitments. ME 218 and ME 310 courses and their students’ academic experiences are also significantly different. While students are not required to complete the entire course sequence, those who take ME 310 generally take all three quarters. The vast majority of students enroll in either ME 218 or ME 310, although some take both courses. ME 218 (Smart Products Fundamentals, Applications, Practice), a four-quarter course sequence, offers students the opportunity to learn mechatronics (i.e., smart products) from the fundamentals to practical application. Students are grounded in learning the technology and acquiring the skills to become expert designers and practitioners. The series of projects students engage in during the year differ from year to year but are typically defined by the teaching team. Projects proposed by the teaching team have carefully defined contexts, needs, and requirements that allow students to pursue a variety of design solutions. The projects require students to build prototypes and utilize technologies and skills from the course. Course projects sequentially increase in sophistication and complexity. Project milestone successes highlight debugged working prototypes that fully address the prespecified requirements. ME 310 (Global Engineering Design Thinking, Innovation, and Entrepreneurship) challenges students to work in teams to explore design innovation opportunities in areas of interest of partner companies from diverse industries. The contexts, needs, and requirements of each project are open for student teams to discover and specify for themselves. The learning journey includes iterating the processes of studying potential contexts, need finding, benchmarking, ideation, prototyping, testing, analysis, refinement, and pivoting—continuously pursuing innovative solutions. Prototype milestones occur every 2 to 3 weeks; measures of success are not based on whether the prototypes work or fail, but on how much is newly learned from each phase of building, debugging, and testing (how “well” does it work, rather than whether it works or not). The final delivered solution prototype is expected to be demonstrably functional and have credible resolution and fidelity. ME 218 and ME 310 establish varying foundations for raising innovation, engineering task, and entrepreneurial and design thinking self-efficacies. One prioritizes learning via focused skills in mechatronics (having rich technology-enabling opportunities), while the other via advanced design thinking processes (having diverse scoping and industry applicability). Notably, ME 218 and ME 310 graduate students are not isolated in silos; they often take these courses concurrently with related courses offered by the Stanford Technology Ventures Program, Stanford’s Graduate School of Business, and the Hasso Plattner Institute of Design (Stanford d. school). Ample cross-pollination contributes to students’ tacit learning experiences within Stanford’s entrepreneurship-minded ecosystem. While the two course sequences are different in many details of their projectbased learning approaches, both aim to graduate creative, confident, and technically competent engineers, who have great potential to become engineering leaders and innovators in the future. As such, we decided there was ample evidence to consider

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . .

31

the impact of both sequences on the career trajectories and motivations of graduates in a single study.

3.2

ME 218 and ME 310 Interpreted Against the Relational Model

Both the ME 218 and ME 310 course sequences exemplify relationships between the academy and the working world. For example, there are interactions between alumni and the students through coaching and mentoring. Furthermore, the projects that challenge the students in the ME 310 three-course sequence and the final course in the ME 218 sequence come from industry sponsors. However, this snapshot does not fully capture the evolving, time-based relationships. Figure 3 represents some of these relationships in ME 218 and ME 310 over the last 30 years within the context of changes in technology trends, as broadly defined. This figure illustrates how the courses have evolved because of the ongoing changes in workplace practices and technology (workplace-to-academy connection). At the same time, some of these changes in workplace practices and technologies are the result of the courses’ graduates being inspired and prepared (because of their education) to be workplace change agents (academy-to-workplace connection). The actual design of our study of ME 218 and ME 310 alumni (Sheppard et al., 2021a, 2021b, 2022) also exemplifies the academic-workplace relational model. We

Fig. 3 Pedagogical changes in ME 218 and ME 310 and industrial technological advancements over many years (Carleton & Leifer, 2009)

32

S. D. Sheppard et al.

have collected data from over 25 years of course graduates, recognizing the potential power of identifying short-term and long-term gains from educational experiences. This adds much-needed detail about what aspects and features really “stick,” as students transition via the academy-to-workplace, as shown by the arrow in Fig. 3. In addition, our research instruments have been designed to enable us to provide actionable feedback to the courses’ teaching teams as part of continuous improvement; this is the workplace-to-academy arrow in the AWR model.

3.3

Similarities in the Career Paths of ME 218 and ME 310

In the prior sections, we have laid out two arguments for why we consider it reasonable to study graduates of the ME 218 and ME 310 course sequences in a single research project. Both arguments are founded on immersive graduate-level project-based courses in mechanical engineering that focus on design (Sect. 3.1), and both are rich examples of active and varied relationships that can be built over many years between the academy and the working world (Sect. 3.2). In this section, we present a third argument. This argument speaks to the remarkably similar range of career paths pursued by graduates from the two courses. Figure 4 shows the industries in which course graduates were working at the time of the survey. The data indicate that both sets of course graduates currently work in a variety of industries, though “Professional, Scientific, and Technical Services” is at What industry sector best aligns with your current or more recent job? 0% Professional, Scientific, and Technical Services Manufacturing Information

15%

20%

13.8% 10.3%

12.1% 5.4%

9.9%

5.4%

7.1%

Educational Services

5.9%

5.2%

Management of Companies and Enterprises

6.4%

4.9%

Finance and Insurance

2.5%

2.7%

Arts, Entertainment, and Recreation

2.0%

Utilities

2.0% 0.5%

1.5%

Armed Forces

0.7%

Seldom Mentioned Industries*

1.0% 4.9%

4.4%

18.2%

19.7% 20%

25% 23.6%

Transportation and Warehousing

25%

10%

13.8%

Health Care and Social Assistance

Other (Not Listed Above)

5%

16.0%

15%

10%

5%

ME218 ME310

0%

NME218 respondents = 406 NME310 respondents = 203

Fig. 4 ME 218 alumni and ME 310 alumni respondents report the industry sector that best aligns with their current or most recent job. *Seldom mentioned industries: Accommodation and Food Services, Construction, Agriculture, Forestry, Fishing and Hunting, Other Services (Except Public Administration), Real Estate and Rental and Leasing, Retail Trade, Public Administration (Except Armed Forces), Administrative and Support and Waste Management and Remediation Services, Mining, Quarrying, and Oil and Gas Extraction, Wholesale Trade

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . . 0%

5%

10%

33 15%

Research and Development 22.7% Design

25%

18.6%

18.5%

Project Management

13.2%

15.0%

Production/Manufacturing

9.8%

9.4%

Functional Management

8.1%

8.0%

Quality/Testing

6.0%

6.4%

Teaching/Educational Outreach

3.2%

3.5%

Other Function/Business Unit

2.7%

Sales

2.7%

Marketing/Public Relations

2.6%

Support

2.5%

IT

4.8% 3.3% 3.5% 1.1% 3.5%

2.1%

Finance

1.4%

Human Resources

1.1%

Legal

0.8%

Security

0.7% 25%

20% 18.4%

20%

15%

10%

5% ME218 ME310

2.9% 1.6% 1.6% 0.5% 0%

NME218 respondents = 406 NME310 respondents = 202

Fig. 5 Business units in which ME 218 and ME 310 alumni respondents classified their current or most recent jobs. Respondents could mark all options that applied to them

or near the top for both ME 218 and ME 310 graduates. The survey data also show an “Other (Not Listed Above)” option where nearly 20% of graduates indicated industries such as architecture, aerospace, and fashion design. Approximately 5% of graduates from both courses reported “Educational Services” as their industry, suggesting that a notable percentage of graduates pursued careers in the academic field. In Fig. 5, graduates of ME 218 and ME 310 reported similar patterns in their current position, selecting from a list of 16 functions or business units. Perhaps not surprising for these engineering graduates, research and development (R&D) and design were the top 2 functions, followed by project management. Interestingly, over 3% of the alumni reported engagement in teaching or educational outreach, reinforcing the idea of the interconnectedness of the working world and the academy. Finally, in Table 1, we examine similarities between ME 218 and ME 310 graduates in terms of their future career interests. We focused on graduates currently employed in one of the three most frequently reported organizational types (employee in a medium or large firm, employee in a small or startup firm, and founder or cofounder of one’s own firm) and considered what each respondent’s future employment interests were relative to these same three options. Two key findings were identified: First, ME 218 and ME 310 graduates who are currently working in medium or large firms or in small/startup firms share a similarity in that both are fairly certain of wanting to remain in their current size or type of firm. The data show that those in small/startup firms are strongly considering a possible future that includes being a founder/cofounder, more so than those who are currently in medium/large firms. Second, ME 218 and ME 310 graduates who are currently founder/cofounders envisioned different futures from one another. The data indicate that ME 310 current founders are more committed to being founders in the future, than are ME

34

S. D. Sheppard et al.

Table 1 Future employment aspirations of ME 218 and ME 310 graduates relative to current employment Career Aspiration in the Next 5 Years1 Current Role

Medium-Large Firm Employee ME 310 ME 218 Small-Start-up Firm Employee ME 310 ME 218 Founder/Co-Founder ME 310 ME 218

Med/Large M (SD)

Small/ Start-up M (SD)

Founder/ CoFounder M (SD)

N5

3.06 (.82) 2.97 (.93)

1.79 (.94) 1.86 (.93)

1.25 (.99) 1.24 (.99)

98 190

1.90 (.90) 1.77 (.91)

3.27 (.75) 3.33 (.82)

1.75 (.92) 1.73 (1.00)

40 101

1.64 (1.20) 1.41 (1.02)

2.41 (1.55) 2.58 (1.47)

3.23 (.95)2 2.60 (1.12)3,4

34 56

1 Scale of 0–4, from Definitely Will Not to Definitely Will, with 2 being might or might not, answer for each of the eight options 2 Paired t-test between average ME 310 aspiration to found/cofound as compared with join a small firm or startup, p = 0.008 3 Paired t-test between average ME 218 aspiration to found/cofound as compared with join a small firm or startup, p = 0.942 4 Two-sample t-test between ME 310 and ME 218, average aspiration to found or cofound, p = 0.008 5 Number of survey respondents who answered both the current work (from eight options) and future plans questions (from eight options)

218 current founders (see Notes 2 and 4 in Table 1). Current founders from ME 218 also expressed interest in a future that involves either becoming a small/startup employee or becoming a founder (see Note 3).

4 Four Interview-Based Studies of ME 218 and ME 310 Graduates From 2020 to 2022, we expanded our research portfolio and methodologies through collaborations with four early career researchers who conducted interviews with graduates from ME 218 and ME 310. Each of these investigators used a grounded theory approach (Glaser & Strauss, 1967) to gain insights into the impact of the ME 218 and ME 310 course experiences, supplementing and expanding upon the quantitative research findings. Table 2 summarizes the key features of each of these studies.

Graduates who are engineering team leaders (2021), Johannes J.L. Lamprecht

Graduates who are founders (2021), Timo Bunk

Case Graduates active in innovation (2020), Nada Elfiki

Research questions RQ1. What skill sets are in common between innovative and entrepreneurial behaviors? RQ2. What insights from engineering alumni can help educators enhance entrepreneurial self-efficacy and innovation self-efficacy? RQ1. How does mechatronics education contribute to entrepreneurial careers? RQ2. What educational experiences do alumni consider valuable for their entrepreneurial careers and what gives them the selfconfidence to found a company? RQ1. What factors contribute to successful remote collaboration within engineering and design (E&D) teams? RQ2. What transferable skills from ME 218 have been especially valuable to E&D team leaders when navigating the transition to remote collaboration? Graduates of ME 218 who were founders (identified through LinkedIn profiles and ME 218 course instructor recommendations)

ME 218 alumni who were remotely leading and directing cross-collaborative teams during the COVID-19 pandemic

Engineering and design leaders working with remote teams

Unit(s) of observation: interviewee criteria Graduates of ME 310 involved in embracing new ideas (ENI)-related activities (identified through the ME 310 survey), either in the company they founded or as an intrapreneur

Entrepreneurs and founders

Unit(s) of analysis Entrepreneurs and intrapreneurs

Table 2 Summary of key features of each of the four qualitative studies of ME 310 and ME 218 graduates

9

19

N 39

(continued)

Identifying how contextual influences in combination with learning experiences (ME 218) enable leaders to adapt to changing their workplace circumstances (i.e., COVID)

Connecting learning (ME 218) to behaviors and actions (founding a company)

Relationship to SCCT (Fig. 1) Connecting learning (ME 310) to self-efficacy (ENI-SE, ISE-SE)

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . . 35

Case Graduates who are founders and who choose not to become founders (2022), Katharina Prantl

Table 2 (continued)

Research questions RQ1. Why don’t high potential individuals (measured as entrepreneurial self-efficacy, ESE) in STEM start their own tech ventures? Are there gender differences in ESE and why? RQ2. What barriers hinder the engagement of females in STEM entrepreneurship?

Unit(s) of analysis Career paths of female engineering graduates with high entrepreneurial self-efficacy

Unit(s) of observation: interviewee criteria Graduates of ME 218 or ME 310 who did or did not start technology companies N 20

Relationship to SCCT (Fig. 1) Learning experiences (ME 218 and ME 310) in combination with selfefficacy do not prescribe a unique set of actions; interests and outcome expectations play a critical role

36 S. D. Sheppard et al.

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . .

4.1

37

Graduates Active in Innovation, Nada Elfiki

Nada Elfiki (2021) focused on how academic and professional learning experiences contribute to the making of an entrepreneurial and innovative engineer. In particular, she investigated the considerable overlap between entrepreneurship and innovation. Nada organized her findings regarding how educational and experiential activities, tasks, and engagement contribute to entrepreneurial self-efficacy and innovation self-efficacy along four main themes: Mastery by doing recognizes how training and repeated exposure to tasks related to innovation and entrepreneurship provide opportunities to iteratively develop prototypes. As a result, students “build to think” while navigating ambiguity and perceiving failures as positive learning experiences. This prototyping mindset prepares them to fail fast and cheaply by engaging in iterative testing to inform their next steps in the product development process. Connections to real life focuses on the opportunities afforded by ME 310 to address industry sponsored projects and problems while collaborating on small teams. The emotional connection and personal commitment to these yearlong problems explores the “why” behind the project. Students are encouraged to come up with creative ideas while exploring the solution space from small scale prototypes to a larger systems context. Interdisciplinary exposure demonstrates how a user-centric approach to design can help students understand their customers in order to identify pain points and articulate their value propositions that address their needs. Knowledge of the business ecosystem, skills in social and emotional understanding, and interdisciplinary project management are critical for successful commercialization and long-term success of technically innovative products and services. Supportive environments emphasizes the factors that contribute to the culture and broader ecosystem of innovation and entrepreneurship. From partnering with local incubators to building coop programs with startups, these kinds of experiences facilitate on-the-ground intensive learning where, as noted earlier, failure is normalized and risk-taking is encouraged. The environment of ME 310 includes people who start their own companies and develop innovative products which positively reinforces and inspires students to believe that they too can pursue this entrepreneurial pathway.

4.2

Graduates Who Are Founders, Timo Bunk

Timo Bunk (2021, Bunk et al., 2022) explored how mechatronics education can be designed and taught to foster the development of more successful startups and individual entrepreneurs, based on the findings from studying the entrepreneurial career pathways of ME 218 alumni. The pedagogy of project-based learning (PBL) with its emphasis on open-ended problems, practical experimentation, and a

38

S. D. Sheppard et al.

hands-on approach had a significant impact on the entrepreneurial confidence and skill development shown by graduates. The final course results of a tangible product, teamwork, and question asking were fundamental activities reiterated throughout the course. While ME 218 has a reputation of being academically challenging with a demanding and stressful workload, this “hardcore” culture was also viewed by some alumni as a valuable opportunity for learning. According to one alum, “ME 218 is the closest experience I have had to the pain of doing a startup. My cofounder was also 218. If someone says they took 218, you are [thinking] ‘OK, this guy can deal with pain’.” While students were confronted with complex problems, there was a strong feeling of achievement when the problem was solved. In addition, the reputation of ME 218 beyond the Stanford community, and the knowledge and skills that alumni possess, are recognized by companies in the mechatronics field. When it came to the reasons why ME 218 alumni choose to start companies, several cited that initial naiveté: “I do not think I would have started the company if I had known how much I did not know after coming out of school. But I was just naive. I thought I knew everything, and I was wrong. And then it was too late.” Making an impact and utilizing their personal skills motivated individuals seeking to pursue new opportunities.

4.3

Graduates Who Are Engineering Team Leaders, Johannes J.L. Lamprecht

Johannes J. L. Lamprecht (2022a, 2022b, Lamprecht and Sheppard 2022) addressed a timely question exploring the impact of the COVID-19 pandemic on effective collaboration in remote environments, particularly for those engineering teams who work with hardware. While the need for more effective communication approaches among distributed team members predated the pandemic, the insights gained from the interviews conducted with ME 218 alumni have implications not only for the current course but also for educators who are preparing future engineering and design team leaders for more frequent remote collaboration. The ME 218 course experience has proved to be fertile ground for students to not only build and practice skills that have been necessary for successful remote collaboration but also to navigate the rapid emergency transition to remote work. In ME 218, students document their work, engage in time and project management, and work in an interdisciplinary context. As noted by Timo, the real-world challenges that ME 218 students investigate require them to explore a solution space that is wide open. For these problems that have not yet been defined, teams are challenged to figure out what questions to ask and to communicate their process and decisions with team members and externally with classmates. The interview findings noted how ME 218 alumni continued to leverage the knowledge and skills they developed in their course experiences as they negotiated

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . .

39

the challenges of remote teamwork and hardware development. Reliance on technology tools such as messaging and community platforms (e.g., Slack) was employed to maintain a positive team culture, as well as for project documentation and communication. Sustaining a culture of information-sharing both inside and outside the technology was critical, such as adding socializing time after each online meeting. In successful remote teams, senior leadership, managers, and supervisors were intentional in their management and facilitation of relationships among team members. Mentoring and efforts to be transparent about management decisions were critical. Communication became more high touch and frequent, between team members, between leaders and team members, and between leaders. The balance between individual agency and empowered decision-making led to shared leadership and greater ownership. Correspondingly, shorter but more frequent team meetings and check-ins ensured alignment in goals and progress, often in combination with a Scrum work framework. While this project infrastructure of communication and leadership was the foundation of effective remote collaboration, the data indicate that remote hardware design is heavily dependent on technical resources and tools, including being able to easily share pictures, video, and CAD files via the cloud. Other tools and approaches included an ongoing livestream of debugging processes and 3D printing of prototypes, which were shipped to team members. Last, any opportunities for engineers to be on site were encouraged in order to facilitate interaction and engagement with the physical work.

4.4

Graduates Who Choose Not to Become Founders, Katharina Prantl

Katharina Prantl (2022) identified alumni from both ME 310 and ME 218 to better understand the decision-making process related to entrepreneurship and innovation and, in particular, the barriers that hinder female alumni from founding a company. The interviews focused on experiences related to career trajectories, personal and professional priorities, and engagement in entrepreneurship education. The findings are represented in four personas. Opportunity seekers have always intended to start a company. They have intentionally sought out educational opportunities (e.g., taking additional entrepreneurship courses or pursuing an MBA) and have positive associations with the culture and meaning of entrepreneurship. While opportunity seekers often engage in activities related to their own venture, they may lack resources (e.g., collaborators, finances) and confidence. Mission and passion drivers have little to no interest in building their own venture; however, they often engage in comparable activities and work as that of founders, through working in a startup, founding their own club, or teaching

40

S. D. Sheppard et al.

entrepreneurship. Their biggest barriers are time and priorities outside of entrepreneurship since their focus is on mission and impact. Autonomy and flexibility drivers have high exposure to entrepreneurship through their networks or prior work experience. Many currently own small businesses; however, they do not see themselves as entrepreneurs. They have some ideas for possible ventures but value work-life balance and, as a result, have low to mid interest in entrepreneurial engagement. Barriers include deep subject matter knowledge and expertise as well as the potential trade-offs in flexibility and independence. High impact drivers often work in innovative companies that previously had been startups and have high responsibility. They may hold leadership responsibilities and view their positions and opportunities through an intrapreneurial lens within their organizations. High impact drivers have strong technical expertise and do not have any interest in pursuing their own ventures due to risk aversion, given the anticipated limitations in resources and the opportunity cost of switching from a position where their expertise and skills are recognized to a less certain entrepreneurial position where they could pursue their world changing idea. The use of the design thinking tool of personas highlights the spectrum of motivations and barriers for entrepreneurship and expands on how engineering educators can improve entrepreneurship training and education to accommodate and address these varying needs.

4.5

Researcher Impact

The four early career researchers, whose work is described in Sects. 4.1–4.4, also identified several areas of personal growth and impact resulting from their engagement in this work: • Identify a research topic that you are passionate and curious about. Like the purpose- and mission-driven motivations of many entrepreneurs, researchers should also strive to identify a topic that they are passionate and curious about. Beyond generating novel, interesting, and publishable research findings, the four researchers shared a keen interest in contributing to the growth of innovative and entrepreneurship ecosystems at universities that directly contribute to the training and development of tomorrow’s leaders. • Become part of iterative and community-based research. The environment in which these four interview studies were conducted provided iterative opportunities for constructive and critical feedback. In addition, researchers shared recommendations for the use of software to manage data collection and interview scheduling and resources for transcription and qualitative analysis tools. Reflection on their own personal growth and skills development and particularly their intellectual identities as scholars was evident during the whole research process. • Be inspired by those who are in your study. One of the researchers was motivated by having the opportunity to speak firsthand with inspiring people and described

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . .

41

it as participating in 19 personal TED talks. Overcoming one’s personal fears about asking questions and being a good listener were also important areas of growth. • Consider your own pathway as you study the pathways of others. Several of our researchers gained practical insights from not only the research findings but especially from the research process. Some aspire to pursue an entrepreneurial pathway in the future by starting their own companies, while others are considering pursuing a PhD. They found the findings immediately applicable to their own personal and professional lives.

5 Discussion and Next Steps These research findings on innovative, entrepreneurial, and technical career pathways and education’s influence on those pathways are just the tip of the iceberg. While we found that many graduates remain involved in R&D and design years after graduating, some of their most important educational learnings were about human relationships and a mindset that “embraces” failure. We also unearthed many new knowledge gaps and questions about how education (and particularly project-based learning) influences students in the short and long term, for example, the question how many of our findings are particular to the graduate populations from a highly selective university that we studied? or how should generational differences be factored into gathering and interpreting data? Recognizing that there is still more to be learned, shared, and acted on, our future work includes next steps such as the following: • Publications through the writing of journal articles and conference articles based on Nada Elfiki’s and Katharina’s Prantl’s findings. We are also developing a manuscript on the combined survey dataset from ME 218 and ME 310 that will present several models examining the factors that influence and drive innovative behavior in the workplace. • Outreach to other academic institutions, programs, and environments will involve the translation of the study findings into workable ideas, activities, and practices for educators who wish to provide their students with similar experiences or who wish to adapt and implement our instruments and methodologies for their own students, alumni, and environments. • Future research efforts are underway through the adaptation of the ME 218 and ME 310 survey instruments for two entrepreneurship and innovation-focused programs in Germany—the Technology Management program of the Center for Digital Technology and Management (CDTM), a joint institution of LudwigMaximilians-Universität München (LMU) and the Technische Universität München (TUM), and the Manage and More program of UnternehmerTUM. This research will explore the unique factors and experiences that contribute to and foster entrepreneurship and innovation education across international

42

S. D. Sheppard et al.

contexts and diverse student populations. In addition, we continue to conduct interviews with female founders and non-founders in different cultural settings to inform comparative research on career pathways, motivations, and goals with specific consideration of the implications of these findings for engineering and design educators. In closing, we reiterate how underexplored the connections and relationships are between engineering education and the world of work in terms of the development of human capital (or as one of our colleagues refers to it, “human capital formation”). There are great societal and individual investments of time, resources, and finances into formal education, and yet there is so much more to be understood about the “payouts” of education in terms of new career opportunities for the individual and new innovations and inventions for our broader society. We invite others to join us in this ongoing exploration to contribute more insights into the fundamental questions of “what ‘sticks’ about engineering education?” and “how do we improve that education?” Acknowledgments We are grateful for the support and encouragement we have received from Professor Larry Leifer and Dr. Prof. Christoph Meinel and the Hasso Plattner Design Thinking Research Program. The following individuals were also involved in designing and carrying out the research described in this chapter: Dr. Felix Kempf, Hung Pham, Dr. Mark Schar, Nicole Salazar, and Stanford’s School of Engineering Alumni Relations and Student Engagement. Others who provided critical feedback and background information include Niclas-Alexander Mauss, Lawrence Domingo, Dr. Jan Aurenhammer, Dr. Claudia Liebethal, Professor Helmut Schönenberger, Professor Mark Cutkosky (ME 310), Dr. Ed Carryer (ME 218), Crystal Pennywell, Gosia Wojciechowska, Elizabeth Mattson, Tammy Liaw, Dr. Sharon Nemeth, Jill Grinager, and members of the Designing Education Lab. Finally, we gratefully acknowledge the many graduates of ME 218 and ME 310 who completed our surveys and answered our invitations to be interviewed—without them, this research would not have been possible. They are the true stars of this work!

References Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice Hall. Bunk, T. (2021). Enhancement of university startup support in the fields of mechatronics and robotics [Unpublished master’s thesis]. Technische Universität München TUM School of Management. Bunk, T., Chen, H. L., & Sheppard, S.D. (2022, June). Exploring the impact of project-based mechatronics course design on alumni’s entrepreneurial career pathways. In Proceedings of the 2022 ASEE Annual Conference & Exposition. Minneapolis, MN. https://peer.asee.org/40650 Carleton, T., & Leifer, L. (2009, March 30–31). Stanford’s ME 310 course as an evolution of engineering design. In Proceedings of the 19th CIRP Design Conference – Competitive Design. Cranfield University, https://dspace.lib.cranfield.ac.uk/handle/1826/3648 De Noble, A., Jung, D., & Ehrlich, S. (1999). Entrepreneurial self efficacy: The development of a measure and its relationship to entrepreneurial action. In Frontiers of entrepreneurship research (pp. 73–87). Wellesley, MA: Babson College.

Decades of Alumni: Perspectives on the Impact of Project-Based Learning. . .

43

Elfiki, N. (2021). The making of an entrepreneurial and innovative engineer: Academic and professional learning experiences that promote entrepreneurial and innovation self-efficacy. [Unpublished master’s thesis]. Technische Universität München TUM School of Management. Gilmartin, S. K., Chen, H. L., Schar, M. F., Jin, Q., Toye, G., Harris, A., Cao, E., Costache, E., Reithmann, M., & Sheppard, S. D. (2017). Designing a Longitudinal Study of Engineering Students’ Innovation and Engineering Interests and Plans: The Engineering Majors Survey Project. EMS 1.0 and 2.0 Technical Report. Stanford, CA: Stanford University Designing Education Lab. Glaser, B., & Strauss, A. (1967). The discovery of grounded theory: Strategies for qualitative research. Sociology Press. Hammonds, J. (2016, January 6). Green STEM: An educational collision of epic proportion. NSF Blog. https://blog.nwf.org/2013/04/green-stem-an-educational-collision-of-epic-proportion/ Lamprecht, J. J. L. (2022a). Key components of effective remote engineering work: A leader’s lens on the impact of COVID-19 on team collaboration. [Unpublished master’s thesis]. Technische Universität München TUM School of Management. Lamprecht, J. J. L. (2022b). How to better manage remote collaboration: A leader’s perspective [Infographic]. https://bit.ly/ManageRemoteInfographic Lamprecht, J. J. L., & Sheppard, S. D. (2022). Key components of effective remote engineering work: Factors learned in school and on the job – Study motivation, design, and preliminary results. In Proceedings of the 2022 Frontiers in Education Annual Conference & Exposition, Uppsala. Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social cognitive theory of career and academic interest, choice, and performance. Journal of Vocational Behavior, 45, 79–122. Lent, R. W., & Brown, S. D. (2006). On conceptualizing and assessing social cognitive constructs in career research: A measurement guide. Journal of Career Assessment, 14(1), 12–35. Main, J. B., Griffith, A. L., Xu, X., & Dukes, A. M. (2021). Choosing an engineering major: A conceptual model of student pathways into engineering. Journal of Engineering Education, 11(1), 40–64. https://doi.org/10.1002/jee.20429 National Academy of Engineering. (2018). Understanding the educational and career pathways of engineers. The National Academies Press. https://doi.org/10.17226/25284 Prantl, K. (2022). Decoding the gap: A theory of women non-founder types in STEM in the context of entrepreneurial career barriers. [Master’s thesis under review]. Technische Universität München TUM School of Management. Schar, M. (2020). Design thinking self-efficacy. [Unpublished presentation]. Schar, M., Gilmartin, S. K., Harris, A., Rieken, B., & Sheppard, S. (2017a). Innovation selfefficacy: A very brief measure for engineering students. In Proceedings of the 2017 ASEE Annual Conference and Exposition, Columbus, . Schar, M., Gilmartin, S. K., Rieken, B., Brunhaver, S.R., Chen, H. L., & Sheppard, S. (2017b). The making of an innovative engineer: Academic and life experiences that shape engineering task and innovation self-efficacy. In Proceedings of the 2017 ASEE Annual Conference and Exposition, Columbus, OH. Sheppard, S. D., Chen, H. L., Toye, G., Kempf, F., & Elfiki, N. (2021a). Measuring the impact of project-based design engineering courses on entrepreneurial interests and intentions of alumni. In C. Meinel & L. Leifer (Eds.), Design thinking research: Translation, prototyping, and measurement (pp. 297–313). Springer. Sheppard, S., Chen, H. L., Toye, G., Kempf, F., & Elfiki, N. (2021b). Decades of alumni - What can we learn from designing a survey to examine the impact of project-based courses across generations? In Proceedings of the Annual Conference for the American Society of Engineering Education, Virtual Meeting. Sheppard, S. D., Chen, H. L., Toye, G., Bunk, T., Elfiki, N., Kempf, F., Lamprecht, J. L., & Lande, M. (2022). Decades of alumni: Designing a study on the long-term impact of design. In C. Meinel & L. Leifer (Eds.), Design thinking research: Achieving real innovation (pp. 247–269). Springer.

Part 1

Application of Design Thinking to Governance and Social Causes

Predicting Creativity and Innovation in Society: The Importance of Places, the Importance of Governance Julia von Thienen, Kim-Pascal Borchart, Detlef Bartsch, Lars Walsleben, and Christoph Meinel

Abstract Places causally impact the development of creativity and innovation. Observable patterns include “creative explosions” at some places and times, which contrast to phases of continuity at other locations and times. But what characteristics of places enhance or trim creativity? This paper draws attention to governance, which strongly impacts emerging creative work in the region. We therefore introduce a four-leaf clover model in design thinking, according to which promising innovation is (1) desirable in terms of human values, (2) feasible in terms of technology, (3) viable in terms of business models, and (4) legal in terms of governance. Making a start in quantifying the impact of governance, we share a case study on genetic engineering in European agriculture. Furthermore, the TYPE method is introduced as an approach to compare rates of creativity and innovation in a region under current regulation as opposed to a hypothetical scenario without regulation. Subsequently, we suggest key figures to trace the innovation potential in a region over time, building on the computational process model of invention. Such key figures allow comparisons between regions on behalf of their overall innovation potential, and they reveal profiles of strengths and weakness in terms of available resources for creative work. This chapter closes with a discussion of governance approaches that trim creativity and innovation, as opposed to governance approaches that raise the innovation potential in a region.

J. von Thienen (✉) · K.-P. Borchart · C. Meinel Hasso Plattner Institute for Digital Engineering, Potsdam, Germany e-mail: [email protected]; [email protected]; [email protected] D. Bartsch Bundesamt für Verbraucherschutz und Lebensmittelsicherheit (BVL), Berlin, Germany L. Walsleben Westfälische Wilhelms-Universität (WWU), Münster, Germany © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_3

47

48

J. von Thienen et al.

Fig. 1 Design thinking works with a 4P model of creativity and innovation. In class, students learn to orchestrate resources from the domains of creative people, processes, and places in order to develop worthwhile creative products. Thus, places are addressed as one out of three pillars in order to advance or predict creative outcomes

1 The “Place” Factor in Creativity and Innovation It has long been noted that places impact creativity and innovation (Fig. 1). The immediate room where people work has some effect. There is also a notable impact of large-scale places, such as the geographic region or the political context in which people operate. Overall, this chapter defines places as any environment that impacts how people feel and behave, including their inclination to be creative. Some places are welldefined in terms of physical boundaries, such as a room or building. Other places are socially defined, as when addressing people who work in a particular organization or in a “zone of psychological safety” characterized by mutual trust among colleagues. Places can be analog or virtual. Moreover, places can be relatively small, such as a desktop background, or relatively large, such as a country, continent, or planet. As research has pointed out, places impact the number and kind of creative outcomes that emerge. At some places and times, a huge number of creative solutions emerge. These developments can be so outstanding and seemingly abrupt that they have even been described as “creative explosions” (Mithen, 2005). Such episodes of innovation contrast to long phases of continuity and stasis that seem to prevail at other places and times. The archaeological record suggests that such patterns already existed in humanity’s prehistory: There are times and places in the past when a great many novel artefact forms or procedures appeared [. . .]. These seemingly fertile intervals are brought into sharp relief by comparison with times and places when very little that was new or different emerged. (Kuhn, 2012, p. 69)

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

49

In historic times, these patterns are well-documented too. The appearance of [creative] genius fluctuates over place and time. Geniuses do not pop up randomly—one in Siberia, another in Bolivia—but in groupings. Genius clusters. Athens in 450 BC. Florence in AD 1500. Certain places at certain times, produced a bumper crop of brilliant minds and good ideas. (Weiner, 2016, p. 8)

Such patterns of creative explosions that contrast to phases and places of stasis seem to be more than a coincidence. Experimental research has demonstrated that places causally impact creativity and innovation (von Thienen et al., 2012; Guegan et al., 2017; Bourgeois-Bougrine et al., 2020; Richard et al., 2018). Against this background, there is a likely question to ask: How does the location of people impact their creativity and innovation? In this chapter, we turn toward a place factor that is presumably very impactful in shaping creativity among individuals and whole societies. Still, the subject has received surprisingly little attention in design thinking research and creativity literature so far. The question is as follows: How does the political environment, or governance, impact creativity and innovation in society? The importance of this large-scale place factor has been pointed out recently by Mitchell and Bartsch (2020), leading to the redesign of an often used framework in design thinking.

2 A Four-Leaf Clover Model of Innovation In design thinking literature, innovation has often been described as emerging in the middle of a three-leaf clover. According to this model, innovation thrives when creative solutions are (1) desirable in terms of people’s needs, (2) feasible in technical terms, and (3) viable in light of business models. As Kelley and Kelley explain In every innovation program we have been involved with, there are always three factors to balance [. . .]: The first has to do with technical factors, or feasibility [. . .]. The second key element is economic viability, or what we sometimes refer to as business factors. Not only does the technology need to work, but it also needs to be produced and distributed in an economically viable way. [. . .] The third element involves people, and is sometimes referred to as human factors. It’s about deeply understanding human needs. (Kelley & Kelley, 2013, p. 20)

Mitchell and Bartsch (2020) have pointed out how successful innovation, in addition, requires the satisfaction of constraints in a fourth domain. It is also characteristic for the innovation sweet spot that respective creative developments are legally permitted, politically accepted, and maybe even encouraged. Let us consider an example from the domain of genetic engineering (e.g., Schmidt et al., 2020; Wilhelm et al., 2021). In this realm, arguably, the satisfaction of the classic three constraints is not enough for creative developments to take place. In

50

J. von Thienen et al.

terms of human values, there are stakeholders with strong concerns for more sustainable farming methods, such as having to use less pesticides or water resources while being able to feed growing numbers of people on Earth. In technological terms, it is feasible today to produce plants with looked-for traits via genetic engineering methods. For instance, plants can be developed to be more weed tolerant and then require fewer pesticides, or plants can be rendered more drought tolerant and then need less frequent watering. In terms of business prospects, there are large markets for plants of this kind around the globe. However, in terms of governance, such developments are only accepted and encouraged in some parts of the world. By contrast, other regions including the European Union discourage these kinds of developments. While the application of genetic engineering methods in agriculture is not strictly prohibited in Europe, it is governed in very strict ways. Those who would like to implement projects need to obtain permits to do so. The requirements and approval procedures are so complex that projects in Europe tailored toward market release have remained close to zero. This contrasts to other places around the globe, where many projects are being put to practice. Views on genetic engineering vary widely. While some stakeholders see great potential in this technology, others find it risky and therefore consider respective projects undesirable, no matter what the concrete project goals might be. Therefore, we want to provide a second example from a different domain. Furthermore, we wish to emphasize that our goal is to describe and model factors that impact creativity and innovation. We do not advise for or against particular political values or decisions. The impact of governance can also be seen in contexts where the development of creative solutions, desired by some stakeholders in society, threatens the power of authorities. These authorities may then take action to counter those creative developments. For instance, there are stakeholders around the globe who embrace human values of gender equality. This yields market opportunities, and viable businesses could be established based on the interests of customers, for instance, when women in a society decide they want to buy trousers for themselves. Technologically, it is certainly feasible to produce trousers for women. However, when an authoritative regime feels threatened by the idea of gender equality, they are in a position to legally forbid the wearing of trousers by women in the public. As a result, creative developments in a particular fashion domain will not unfold as they would without regulation. In this case, people in the society cannot conduct desired projects on designing, distributing, and wearing clothes in ways that signal gender equality. In short, governance plays a significant role in shaping the ability of society to generate and implement innovative solutions (Fig. 2). It can either foster or inhibit creative exploration and novel developments. Worthwhile innovation thrives in the middle of a four-leaf clover: The new solution is (1) desirable as it aligns with human values, (2) technically feasible, (3) economically viable, and (4) legally permissible, if not actively encouraged by governance.

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

51

Fig. 2 Innovation thrives at the sweet spot of four intersecting fields: (1) desirability in terms of human values, (2) feasibility in terms of technology, (3) viability in terms of business models, and (4) legality in terms of political governance

3 Key Tasks in Innovation Governance: Advancing Desirable Developments, Protecting Against Risks In the governance of creativity and innovation, it is important to note that not all creative endeavors yield positive outcomes. For example, someone may use their creativity to design a weapon with the intention of causing harm to others or the planet. Such actions are widely considered undesirable. In research, developments like this fall under the headline of “malevolent creativity” (e.g., Cropley et al., 2008, 2014; Eisenman, 2008, Hao et al., 2016; Perchtold-Stefan et al., 2021). This field of research focuses on the intentional use of creativity for harmful or malicious purposes. Notably, undesirable outcomes can emerge from creative activity when people work toward criminal ends, but also when they have good intentions and aim for desirable outcomes. In iconic cases staged colorfully in movies, protagonists have worthwhile aims in mind and then conduct risky experiments; something goes wrong and consequences are fatal. Even in real life, no matter what outcomes people develop, there can always be some unexpected negative side effects. Products might be misused, or creative processes simply culminate in a mess. An example known to many families is the motivation of kindergarten-aged children to engage in creative cooking projects. Here the kids typically have the best of intentions. They want to cook a delicious meal, potentially developing a great new recipe along the way. Yet, without the supervision and governance of caretakers, these creative projects can easily end in a number of different dramas. Children may end up with bloody fingers (from using knives), the living room

52

J. von Thienen et al.

sofa might be splattered with juice stains (due to the preparation of dishes randomly in the home), and larger quantities of food can end up in the trash (due to misjudgments about food combinations that would be enjoyable). The governance of creativity and innovation based on expectable benefits or risks is a topic that affects almost anyone in society. Questions of good governance appear in the family, in organizations, just as they appear for the state or political union. A key objective in innovation governance is to permit desirable creative developments. This may include encouragement and facilitation. At the same time, non-helpful, risky, or harmful developments are to be kept under control, which may happen via prohibition, discouragement, and/or close monitoring. Research can look into the governance approach that regulators pursue. Such an assessment needs to start from innovation domains of interest, because the same authority can decide to be liberal about creative developments in one domain but strict in another. All governance strategies can be located in a triangle of hypothetical scenarios (Fig. 3). At one extreme, authorities can decide not to regulate a particular innovation domain at all. In this case, anyone who has a creative idea can forge ahead with it. There are no particular requirements that creators need to meet. At the other extreme, authorities can opt for maximally strict regulation, which means trying to exert full control over the kinds of creative developments that occur. However, authorities still need to decide whether they want to control based on risks or benefits or some balancing of both.

Fig. 3 The innovation regulation triangle is a framework that helps to understand actual governance approaches by placing them in relation to three hypothetical extremes. These extremes are (1) no regulation whatsoever, (2) regulation solely focused on the avoidance of risks associated with taking action, and (3) regulation solely focused on maximizing particular benefits. Real-world governance approaches can fall somewhere along this spectrum. Authorities can try to design governance approaches in order to tap a sweet spot in the triangle that they consider ideal

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

53

In the hypothetical scenario of maximal regulation based on risks, no creative idea ever achieves impact. The requirements that creators would have to meet in order to obtain project approvals are so extensive, that it is practically impossible to meet them. For instance, if creators have to demonstrate that their endeavor is completely risk-free, nothing can ever be created anew, because not even the most harmless project is 100% risk-free in every possible regard. In the hypothetical scenario of maximal regulation based on benefits, a lot of creative ideas can be realized. However, permissions are given only to those creators whose project ideas serve “benefits” acknowledged by the authorities. When creators can demonstrate such benefits, their projects are approved and risks are ignored. Subsequently, we will look closely at an example of strict regulation focused on risks. To briefly mention an example of strict regulation focused on benefits, one can think back to the colonial days. Many kings and queens have been sensitive to the benefit of increasing their own wealth. Creative citizens interested in traveling, science, and/or adventure suggested projects to the rulers, where the official aim was to organize gold or other treasures for the emperors. In a number of cases, the sovereigns approved of these projects and supported them. This happened irrespective of the fact that many of these projects not only involved risks, but they caused great harm. In particular, for other countries and their inhabitants, such projects were often fatal. We will now turn to a case study, which estimates the position of European governance in the innovation regulation triangle. The case study focuses on the domain of genetic engineering for agriculture. Subsequently (in Sect. 5), we will introduce the TYPE method, which allows for the positioning of any governance approach within the innovation regulation triangle for a selective field of creative work.

4 A Case Study on the Impact of Governance: Genetic Engineering in European Agriculture Bioengineering is one area where the influence of governance can be seen (Fig. 4). Specifically, regulations for new genomic techniques (NGT) vary greatly worldwide. These technologies enable a precise design of organisms, such as crops, with particular looked-for traits. In some countries, those technologies are widely utilized, while in other countries their use is highly restricted or outright prohibited. Overall, currently four major techniques are available to develop plants with new traits, such as increased draught and weed tolerance (Table 1). All four techniques produce plants that often cannot be distinguished from naturally evolved variants. That means, without human intervention, identical plants could emerge via natural evolution. As shown in Fig. 4, Europe is known for its strict regulation of genetic engineering methods in agriculture. However, it is important to note that terms such as

54

J. von Thienen et al.

Fig. 4 Around the world, different regulatory environments (ranging from restrictive to liberal) have been established for new genomic techniques. The image is reprinted from Schmidt et al. (2020), according to the Creative Commons License CC BY 4.0 Table 1 Overview of four major techniques used for the development of new plants

Technique (1) Selective breeding (2) Chemical (3) Radiation (4) New genomic techniques (NGT)

Time demand Very high High High Low to moderate

Undesired/ random plant changes Few

Decades of documented safe use in the EU Yes

Product developments in the EU Many

Many Many Few to moderate

Yes Yes No

Many Many Hardly any

“strict” or “lenient” are relative. The level of strictness in regulation can vary, leading to a gradual decline of innovation rates in society. When the intent is to regulate an innovation domain strictly, this may involve aiming for a reduction of 50% or 70% compared to the creative developments that would be expected without regulation. Ideally, in such cases, the regulator may want to maintain the 30% or 50% of creative projects that offer the greatest benefits while having the least associated risks. Less beneficial and/or more risky creative endeavors would not be implemented due to the strict regulation. On the other hand, at the extreme end of the spectrum, we may also be facing a situation where 0% of the innovation occurs that would be expected in this particular society in the case of complete deregulation.

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

55

In order to gain insight into potential innovation rates in Europe under varying regulations, looking into applications of New Genetic Techniques (NGT) in the realm of agriculture, we hosted three rounds of expert discussions. The first round took place early in 2022 and brought together a group of N = 5 individuals with extensive knowledge of genetic engineering methods, respective regulations in Europe, as well as institutions working in this field. Subsequent discussion rounds were hosted at the end of 2022. They aimed to trace any changes in estimates given the evolving regulatory discourse and market developments. Furthermore, different estimates could reflect varying views among experts. By the end of 2022, we assembled two separate groups of experts (N = 4 and N = 6), asking them to provide the same estimations as in the initial discussion. Out of the original N = 5 group, two experts also took part in the repeated estimations, one joining the N = 4 team, while the other joined the N = 6 group. Each session followed a structured format involving two phases. The first was dedicated to an estimation of numbers under current regulation, while the second phase was concerned with a hypothetical scenario of complete deregulation. The first phase started with the following prompts, to be discussed for 10 min: • How many companies in Europe seem to pursue NGT-based product developments for agriculture at present (intended for markets inside or outside of the EU)? • On average, how many different NGT products for agriculture does each company try to develop within 10 years? • What is the average success probability with which each company achieves an NGT product they try to develop within 10 years? • Do individuals pursue NGT-based product developments in Europe? If so, how many successfully completed NGT product developments do you expect from individual work in the next 10 years? In the initial N = 5 study early in 2022, the experts reported that only a small number of companies appeared to be using NGT in agricultural projects. The experts noted that these companies seemed to focus on the development of products for markets outside of the EU. The discussion identified approximately nine companies that were pursuing NGT-based agricultural projects in Europe. These companies were estimated to conduct 20 product developments on average in the upcoming decade, each with a success probability of approximately 0.9. This led to an estimated total of 9 × 20 × 0.9 = 162 NGT-based products for agriculture to emerge in Europe in the next decade. However, by the end of 2022, the situation was perceived differently in the repeated expert discussions. More experts believed that Europeans were beginning to engage in NGT-based projects for agriculture under current regulation. While some experts still believed that only companies were currently engaged in such endeavors, others anticipated an increase in product developments in Europe from individuals as well. One expert group estimated that there would be a total of 1057

56

J. von Thienen et al.

agricultural NGT products to emerge in Europe in the next 10 years under current regulation, while the other expert group estimated 1485 products to emerge. In the second phase of the discussion, groups were instructed to estimate numbers for a hypothetical European scenario of complete deregulation. To begin, the teams obtained 10 min for orientation, working with the following prompts and available laptops for online research: • How many individuals in Europe have the know-how to conduct NGT-based product developments for agriculture? Internet research • • • •

What universities or other institutes convey knowledge in this field? How long have respective training programs existed? How many people have thus been trained in Europe by now? How many topic experts are estimated to exist at present in the EU, considering also immigration and emigration?

Subsequently, the teams obtained 10 min to arrive at numerical estimates for the deregulation scenario. This discussion phase was guided by the following prompts: • Out of all persons in the EU who have the know-how to conduct NGT-based product developments for agriculture, how many have the motivation and access to equipment in order to implement projects? • How many projects does each motivated and well-equipped expert conduct on average within a decade (considering that some projects might be conducted alone, others within companies or groups)? • What is the average success probability with which each project achieves the NGT product they try to develop within 10 years? During the initial expert discussion held early in 2022, the experts came to the following conclusion: It was estimated that there were approximately 20,000 individuals in Europe with the subject knowledge to conduct agricultural product developments using NGT. Out of this group, it was expected that 1500 individuals had the motivation and access to necessary equipment to actually implement NGT-based projects. Assuming each individual dedicated 10 years to their respective product developments and the probability of success for each year of development was 0.6, the expert discussion yielded an expected value of 1500 × 10 × 0.6 = 9000 product developments in Europe given the hypothetical scenario of complete deregulation. These estimates suggest that the current European regulation results in a reduction of around 98% of completed product developments in the sector of NGT-based plants for agriculture, when compared to the hypothetical scenario of deregulation (9000 – 162 = 8838 fewer product developments). These estimates are depicted in Fig. 5, illustrating the use of the innovation regulation triangle in the description of governance approaches.

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

57

Fig. 5 Based on estimates from an expert discussion, European governance of NGT in agriculture can be described as “very strict,” allowing for the implementation of only 2% of the product developments that would be expected in a hypothetical scenario of complete deregulation

When the expert discussion was repeated by the end of 2022, different numbers emerged. One group estimated that under a hypothetical scenario of deregulation, 135,000 agricultural NGT products could emerge in Europe within the next 10 years, while the other group arrived at an estimate of 270,000. The significant difference between these estimates reflects the uncertainties that experts face when considering hypothetical scenarios that diverge significantly from the current situation in Europe—an aspect that should be taken into account when interpreting these numbers. Despite this variation, all expert groups agree that there is a drastic reduction in the number of NGT product developments due to the current strict regulation. According to one expert discussion by the end of 2022, there is a reduction of 99% in NGT product developments in Europe compared to the deregulation scenario (133,943 products out of 135,000). By comparison, the second expert group estimates a trimming effect of 99.5% (268,515 products out of 270,000). In summary, the expert discussions indicate that the current regulation results in a reduction of potential NGT product developments for agriculture in Europe by approximately 98%, 99%, or 99.5%, compared to what would be expected under complete deregulation. These estimates provide a numerical impression of the level of strictness with which NGT are currently regulated in Europe.

5 The TYPE Method to Estimate Product Developments in a Place Without Regulation In the case studies above, the calculation of hypothetical scenarios without regulation has been conducted based on the following simple methodological approach, which can be reapplied on behalf of other innovation domains (Table 2). We introduce this method as the TYPE framework, standing for tools, lifetime years, success probability, and expected products. Out of the 4P framework of creativity and innovation (Fig. 1), this estimation method begins by determining numbers for 3P, namely, already existing products

58

J. von Thienen et al.

Table 2 Factors used to estimate numbers of successful product developments in a place for scenarios of complete deregulation Place Old products “Tools”

People “Lifetime years”

Process “Success probability”

New products “Expected products”

(tools), the lifetime years people invest in creative developments using these tools, and their process or success probability per year. These estimates are characteristic for places at certain times. For instance, in our case study, it has been estimated that currently 20,000 persons in Europe possess the domain knowledge to use NGT in agricultural projects. In the expert discussion, this appraisal depended in part on the number of universities that offer respective education in Europe and their student numbers per year. Clearly, there is an upper limit to the domain experts that can exist in a region, based on the overall population size. In 2021, the state with the smallest population size on Earth has been determined to be the Pitcairn Islands with 50 inhabitants.1 Obviously, a state that only has 50 inhabitants could not possibly have 20,000 experts in any particular knowledge field. The larger the state and its population, the larger its innovation potential in terms of people who can have the expertise and can invest work years in novel product developments. Similarly, tools are place-specific. For example, Europe in the 1950s did not have access to NGT for agricultural projects, because these technologies were only invented later in history. Similarly, some countries today may not have access to these tools because they have not kept up with the latest technological advancements. In a scenario without regulation, the likelihood of successfully completing a creative project is largely dependent on the tools used. For example, when working on plants with specific desired traits, the probability of success varies depending on the tools used. The application of NGT increases the probability of success within a specific time frame, such as one work year, in comparison to traditional selective breeding techniques when targeting the same plant trait, such as enhanced drought tolerance. In addition, the expertise of the individuals involved can also be considered. If the product developers have limited knowledge in the use of the tools, it can make sense to assign a lower probability of success compared to those who have received state-of-the-art education in the subject domain. In a scenario that includes regulation, the probability of successfully completing a creative project depends on a range of other factors as well. For instance, if product developers have to invest 50% of their work time in documentation, product developments take at least twice as long, because the active work time in projects has been halved. In addition, there might be switching costs for creators, who need to take care of two types of processes: creative and administrative. 1

https://de.statista.com/statistik/daten/studie/689720/umfrage/bevoelkerungsaermste-laender-derwelt/

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

59

To sum it up, the TYPE method aims to estimate realistic numbers of successful product developments in an innovation domain of interest. These estimates can be calculated for regions under current regulation, looking some years into the future. For comparative purposes, estimates can also be calculated on behalf of hypothetical scenarios of complete deregulation. In the remaining part of this chapter, we will look into another modeling approach. It is dedicated to the calculation of key figures, where we specifically suggest the monitoring of a factor called innovation potential. It is an integrative measure of the resources for creative work that a community has available. These resources provide the basis for creative productivity in the future. To some degree, the TYPE method and calculations of the innovation potential integrate information from the same variables. In particular, both approaches look into tools, invested lifetime years, and the yearly success probability in creative developments. These factors reappear in the modeling approaches, as they do in creativity literature at large. To provide further pointers, for the impact of tools, we refer to Enquist et al. (2008) and Kolodny et al. (2015). For the impact of population size or person-time, we refer to Henrich (2004), Derex et al. (2013), and Creanza et al. (2017). For the impact of project success probabilities, we refer to Arnold (2016), who discusses various creativity approaches from random trials to more systematic searches and their different lengthiness and success probabilities. Notably, the two methods introduced in this chapter—which integrate information on tools, people, and success probabilities—serve different purposes. In contrast to the TYPE method, the calculation of the innovation potential does not yield realistic numbers of expected product developments in a region. Instead, it is introduced as a key figure to characterize the creative output that is possible in a region, numerically integrating aspects of output quantity and output variety.

6 Introducing the Computational Process Model of Invention to Calculate Key Figures of Innovation Potential in a Region In economics and business administration, it is common to calculate key figures that provide an overview of economic situations and their development over time. One example is the gross national income or the gross national product. These indicators aggregate and integrate information from various economic variables, condensing them into a single overall number. Key figures are used to identify problems, determine strengths and weaknesses, gather information, control or monitor, document, or plan actions. They provide interested parties with a quick overview of economic developments. This becomes possible by collecting data and calculating key figures at regular intervals, allowing for comparisons and trend analyses. We propose that creativity and innovation studies should also develop key figures to provide an overview of the innovation potential in a specific place, to facilitate

60

J. von Thienen et al.

Fig. 6 The computational process model distinguishes seven mechanisms of invention, which have been found to resurface in human creative activities. The model builds on the lucky leap modeling approach (Kolodny et al., 2015). In its present form, the model is discussed in detail by Corazza and von Thienen (2021) as well as von Thienen et al. (2023)

comparisons and trace developments over time. Such key figures can be determined for places like a state or country and its entire population, but also for smaller-scale places such as an organization and its employees. To begin, we introduce an indicator for the innovation potential in a region. It can be calculated based on the computational process model of invention. This model identifies seven different mechanisms of invention that have been found to be recurrent in human creative processes throughout history and prehistory (Fig. 6). Governance approaches can impact each of these invention mechanisms. On such grounds, profiles can be created that depict strengths and weaknesses of innovation policy in a region. Those governance profiles must be formulated in relation to policy goals. For example, if a region wants to promote innovation in a specific field, it should support the application of invention mechanisms in that work area. By contrast, if a region has ethical or risk concerns about certain types of innovation, policy can discourage the use of those invention mechanisms in order to decrease the respective innovation potential in the community. Against this background, strengths of current governance approaches can be described when policy wants to encourage innovation in a work area and the taken policy measures effectively serve this end, or when policy aims to reduce creative developments in a particular application field and the taken policy measures actually lead to such reductions. In order to prepare for the calculation of the innovation potential in a region, let us briefly look into the computational process model and the seven mechanisms of invention it distinguishes:

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

61

1. Basic inventions emerge (see Boxes 1, 2, and 3 in Fig. 6). This is when an invention appears for the first time. For instance, at some point in human (pre-) history, agriculture and the wheel appeared for the first time. 2. Basic inventions evolve into invention branches. These branches exhibit two major characteristics: diversification and refinement (e.g., Boxes 1.1, 1.11, or 1.12 in Fig. 6). For instance, agriculture diversified into livestock production and crop production, with specialist knowledge expanding in each field. Refinement means that solutions become more effective over time, in light of a continuous value or development goal. For instance, crop yields have increased over time. 3. Invention combinations emerge (e.g., boxes 1&2, 2&3 and 1&2&3 in Fig. 6). For instance, combining the basic inventions of agriculture and the wheel yields vehicles for agriculture, like a hay cart. 4. Toolkits are developed (e.g., Boxes 1a, 1b, 1c, and 1d in Fig. 6). Toolkits make the implementation of a basic invention more efficient and/or effective. For example, alongside the invention of agriculture, toolkit inventions such as the plow and water management systems were developed, so that agriculture could be implemented in more effective ways. 5. Exaptations emerge (e.g., boxes 1′, 1*, and 1♦ in Fig. 6). Novel uses are found for already existing technology. This occurs by identifying basic principles in past inventions and reusing them to invent something new. For instance, a basic principle of the wheel is something spinning around an axis. This principle can be reused to create a windmill or pulley. 6. Values are extracted, maintained, or changed (e.g., Boxes 1-α and 1-β in Fig. 6). Every invention conveys values and can elicit value discussions. When values stay constant, incremental innovation emerges, which means that already established solutions become fine-tuned by changing details. For example, in agriculture, there is an ongoing concern for maximal crop yields, and so plants are bred constantly for this purpose. When societal values change, innovation is led to explore novel directions, and leap innovation is likely to emerge. In this case, novel solution avenues are found that vary greatly from past solution strategies. For instance, novel values of food taste or sustainability can guide innovation projects toward novel kinds of crops, even when this entails compromises regarding crop yields and the use of nontraditional breeding technologies. 7. Finally, inventions can become game changers for creativity and innovation in one or multiple fields (cf. surrounding frame in Fig. 6). In such cases, an invention greatly facilitates the ability of the entire community to engage successfully in creative projects. For example, the development of the printing press enabled knowledge to be passed on more reliably to future generations, and it became easier to disseminate knowledge across wider geographical regions. All inventions taken together (e.g., all boxes shown in Fig. 6) can be addressed as the overall set of tools that a particular community has available.

62

6.1

J. von Thienen et al.

Calculating the Baseline Innovation Potential in Society

Based on the computational process model and in line with the TYPE estimation method, three positive factors are proposed for calculating the baseline innovation potential of a community. (+) Tools: These are past inventions, or existing products, that are available to creators (e.g., every box in the computational process model). (+) Lifetime years: This refers to the number of work-ready persons in the community or more specifically to the years of lifetime they dedicate to creative explorations with the given tools. (+) Success probability: This refers to the probability of each lifetime year invested in creative exploration to yield successful outcomes, i.e., novel solutions, inventions (a new box in the computational process model). Apart from these three positive factors, there is one that can prove a hindrance (cf. Katz, 1990; Ram, 1987; Heidenreich & Spieth, 2013). (-) Innovation resistance: This factor captures the loss of creative outcomes in a community that is not interested in or opposed to innovation. The resistance factor quantifies the loss of tools after their invention due to lack of support from the community. When innovation resistance is at its maximum, the community does not use or preserve novel tools provided by inventors, so every invention that is made is lost again afterward. By contrast, when innovation resistance is zero, the community’s set of tools grows at its maximum rate. Every novel solution that inventors provide is secured afterward in the community; nothing gets lost. Against this background, we propose a simple model to overview key factors that impact the innovation potential in society. Definition 1 The innovation potential is the product of lifetime years, tools, and the yearly project success probability, minus innovation resistance, so that IP = LY
T
SP - IR: Since some creative projects can be shorter than a year, one could also choose a different or more abstract time unit. However, empirical data on how long it takes humans to make notable creative contributions in the world suggests that a year is a rather fine-granular and meaningful unit (cf. Sect. 7). Moreover, it is certainly possible to calculate with numbers, such as 0.25 LY, equaling 3 months, or any smaller or bigger number. Overall, the innovation potential is a key figure that integrates information on possible output quantity and variety of creative work in a region. More specifically, the yearly success probability is a quantitative metric of how many projects will be completed per year, due to the successful development of an invention. The tools variable reflects the diversity of solutions that a community already has at its

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

63

disposal. Lifetimes years are multiplied with both factors. By investing their lifetime, individuals can turn abstract success probabilities into tangible outcomes, resulting in a specific number of new inventions (quantity aspect). Additionally, by investing their lifetime, individuals can also explore different combinations of available tools (variety aspect). However, the baseline innovation potential can be limited or reduced by innovation resistance, which hinders the adoption of new inventions in the community.

6.2

How Governance Can Impact the Innovation Potential in Society

The baseline model to compute the innovation potential in society comprises four factors, three of them positive (increasing creative potential) and one of them negative (reducing creative potential). Each of these factors can be impacted by governance. Furthermore, interventions can have either positive or negative impacts on each factor. Sometimes, governance is complex. Some measures increase the innovation potential, while other measures trim it. For this reason, it seems helpful to keep these factors apart, to allow for more pinpointed analyses of the impact that each intervention has. Definition 2 (a) The innovation potential in a regulated environment equals “lifetime years plus regulated lifetime year gains minus regulated lifetime year losses” multiplied by “tools plus regulated tool gains minus regulated tool losses” multiplied by the “yearly project success probability plus the regulated gains in success probability minus the regulated losses in success probability” minus the “innovation resistance plus the regulated gains in innovation resistance minus the regulated losses in innovation resistance,” so that rIP = LY þ rLYgain - rLYloss
T þ rTgain - rTloss
SP þ rSPgain - rSPloss - IR þ rIRgain - rIRloss : This can also be expressed more shortly as follows. Definition 2 (b) The innovation potential in a regulated environment equals regulated lifetime years, multiplied by regulated tools, multiplied by the regulated yearly project success probability, minus the regulated innovation resistance, so that. rIP = rLY
rT
rSP - rIR: In a next step, we consider sample governance interventions that impact different factors in the model. Some of the considered governance approaches are “hard interventions” in terms of changing the law code. Other approaches are “soft interventions,” such as

64

J. von Thienen et al.

facilitating or sponsoring desired cultural activities. The discussion begins with factors that impede creativity and innovation in society (Sect. 7), followed by a discussion of facilitating factors (Sect. 8). Of course, this overview of possible political interventions does not claim to be complete. However, the model can help to clarify political objectives (e.g., make access to a tool easier or more difficult) in order to increase or decrease the innovation potential in a particular work area. This may facilitate the selection or development of suitable governance interventions. In addition, it becomes easier to monitor in descriptive terms how policies differ between countries or how interventions evolve within one country.

7 Governance Interventions that Impede Creativity and Innovation Subsequently, we will discuss three major avenues of how governance can impede creativity and innovation in a region. This occurs via (1) prohibition, (2) administration, and (3) legal inertia. All of these factors reduce the innovation potential that otherwise would exist in the community. Readers are welcome to reflect on further governance interventions they can think of, which also trim creativity and innovation in the region.

7.1

Prohibition

Prohibition occurs when the reuse of an invention is not legally permitted in a regulatory environment. This prohibition can be generic—the invention must not be reused in any project. Such a prohibition is seldom. In most countries, even inventions that are considered highly dangerous can be “reused” in the sense of experts authoring scientific publications about them or displaying artifacts in museums. Therefore, typically, when regulation restricts the reuse of an invention, this limitation is context-specific. For instance, in many countries, gene-editing technology must not be used to alter the human genome, but in some countries it may be used in animal germ lines. In terms of modeling, regulatory prohibitions reduce the amount of tools that members of society can use in creative pursuits, so we are quantifying the factor rTloss. To consider an example, let us assume there is an interest in the breeding of crops for sustainability purposes. A community with the following tools and no prohibition can explore at least two invention branches. Community 1, Deregulated Basic Inventions: Crop Cultivation, Conventional Breeding, New Genomic Techniques.

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

65

Values: Traditional Farming, Sustainability. One invention branch in this community can investigate the following combination: Crop Cultivation, Conventional Breeding, Traditional Farming, Sustainability. This path will investigate the potential of sustainable crop cultivation using traditional farming methods, in this case conventional breeding. Another invention branch can explore the following combination: Crop Cultivation, New Genomic Techniques, Sustainability. This path will examine the potential for achieving maximal sustainability in crop cultivation through the use of leap innovation, probing approaches that deviate significantly from traditional farming methods. When the application of NGT is restricted in crop cultivation, it limits the available tools for breeders, resulting in a narrower range of options. Despite NGT being a recognized technology in society, it is excluded as a potential tool for developing new crops. Community 1, Regulated. Basic Inventions: Crop Cultivation, Conventional Breeding. Values: Traditional Farming, Sustainability. With this reduced set, only the first invention branch can be explored by members of society. In this modeling example, rTloss = 1. Regulation reduces the set of tools from 5 (crop cultivation, conventional breeding, new genomic techniques, traditional farming, sustainability) to 4 (crop cultivation, conventional breeding, traditional farming, sustainability).

7.2

Administration

Lifetime is one of the key bottlenecks for high-ranking creativity. In research on creative processes of humans across various subject domains, it was found that individuals commonly need at least 10 years of training before they can author internationally recognized creative works in the field (Bloom, 1985; Hayes, 1989; Ericsson et al., 1993). The time difference between the beginning of an individual’s study time and their most highly appreciated creative work in life is even much greater. A typical finding is that at least another decade passes between the person’s first publication in a field, which is already based on substantial training, and their best creative output (Simonton 1997, 2000; Kaufman & Kaufman 2007). Even for experts who are already well-trained, lifetime devoted in concentrated ways to a creative project remains essential for success. A lifetime that is imbued with parallel or scattered tasks, due to interruptions at work, is worth little compared to lifetime that can be devoted seamlessly and in self-determined ways to the task at hand. When people have to think about project-external tasks alongside creative work, this induces cognitive load, which is known to reduce performance in the long term. Moreover, enduring cognitive load and interruptions at work can be

66

J. von Thienen et al.

burdensome to the extent of posing a significant health risk (Mizuno et al., 2011; Baethge & Rigotti, 2013; Sonnentag & Frese, 2012; Keller et al., 2020; McKee 2022). In this context, it is important to emphasize that high-ranking creativity is cognitively very demanding. The “hobby-type creativity” where people produce a mediocre clay sculpture to decorate their homes has little if any impact on the set of tools in society as depicted in the computational process model. The model addresses inventions such as agriculture or the wheel—examples of notable, nontrivial creative outcomes according to common classifications of creative achievement (Kaufman & Beghetto, 2009). Especially projects tailored toward leap innovation usually require the full devotion of all cognitive resources by the creators. When regulatory environments force creators to run through formal admission procedures prior to conducting creative projects, this does not only cause a loss of lifetime (Fig. 7). Depending on the amount of administrative effort, a few, many, or all members of society can lose interest in creative projects concerning that domain, as it costs “too much time and energy.” Even when people remain willing to work creatively in the field, they can lose a considerable amount of their lifetime years through lengthy administrative processes. It should also be noted that especially high-performing creators are often passionate about the creative work they conduct (Arnold, 1959/2016; Amabile, 1996; Csikszentmihalyi, 1997; Hennessey et al., 2015). These people are exceptionally engaged and successful in their creative pursuits in part because they love what they do. If they cannot do what they love, because they are forced to engage in bureaucratic administrative activities instead, a lot of frustration can result. Especially highperforming creators thus can be tempted to change their environments (potentially leaving their organization or country) in order to find better working conditions, where they experience more creative freedom and face fewer disturbances. For all these reasons combined, administration typically contributes to the factor rLYloss, which represents the loss of lifetime years that society invests in creative endeavours to achieve innovation. Another factor to consider is that administration usually costs the applicants money. Financial resources impact the project success probability, as more money means that creators have better opportunities to organize themselves in a way to succeed in their endeavor. Therefore, administrative procedures usually also feed into the factor rSPloss.

7.3

Legal Inertia

Innovation resistance is a general factor that reduces the innovation potential in society. There can be at least two major sources for this resistance. One is a cultural factor, in particular the fear of novelty or a preference for traditional solutions in the community (Zajonc, 2001; Blair & Mumford, 2007; Mueller et al., 2012; von

Fig. 7 When administrative processes are bureaucratically organized, they can cause massive damage to creativity and innovation in society. In particular, creators are forced to invest their limited resources (learning time, money, patience, etc.) in the administrative rather than in the creative process, where these resources would be greatly needed

Predicting Creativity and Innovation in Society: The Importance of Places,. . . 67

68

J. von Thienen et al.

Thienen et al., 2019). It has also been noted that cultures sometimes seem to display a “tiredness of change,” and then continuity and adherence to tradition become dominant values (Katz, 1990). Beyond these cultural influences, regulatory environments can also induce resistance. In particular, laws can be outdated—for instance, if they were made for earlier technologies that are quite unlike currently emerging tools. There are by now case studies from numerous innovation domains showing how outdated laws pose severe obstacles in the handling of novel solutions (Armitage et al., 2017; Becher, 2021; Schmidt et al., 2020). In these cases, new solutions cannot be seamlessly integrated into the existing regulatory context, leading to time delays and sometimes even harsh, long-enduring legal disputes that bind many resources. The pace of technological change is clearly challenging for regulators. Against such a background, De Barro et al. (2011) provide an interesting case example of a creative and rather improvisational solution to cover a novel technical development under existing regulation. Legal inertia refers to the time delay between the invention of a novel tool and revisions in legal texts to address both old and novel tools in appropriate ways. As a proxy for quantifying legal inertia, we suggest a function that incorporates the number of years since the law was last updated in light of technological advancements. For example, in Europe, the use of NGT for food and feed is still regulated based on the state of knowledge and technology as of 2001. This has led to a processing of more recent technology that many European scientists find “bewildering” (Schmidt et al., 2020). In terms of modeling, legal inertia flows into the factor rIRgain.

8 Governance Approaches to Encourage Creativity and Innovation Regulatory environments can encourage creativity and innovation with regard to all four factors that determine innovation potential in the model. This means to look out for measures that increase lifetime years, tools, and the project success probability while minimizing innovation resistance. Such interventions feed into the factors rLYgain, rTgain, rSPgain, and rIRloss in the model.

8.1

Measures to Increase Lifetime Years

Lifetime years dedicated to creative work in a particular domain are a major predictor of inventions and other creative solutions that can be expected in the work area. Lifetime years depend on the number of persons in society, who possess the required domain knowledge, access to tools, and motivation to make creative developments

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

69

in the subject area. At any location (e.g., in a particular town or country), policymakers can reshape the environment to court or hinder creative work in a specific topic domain. Measures can pinpoint all four factors that impact lifetime years. 1. Measures in terms of migration, housing, etc. that increase the number of persons living in a particular area raise the innovation potential at this location. 2. Schools, universities, museums, etc. in the area, which convey knowledge regarding a particular work domain, increase the innovation potential at this location in the subject domain. 3. Measures to increase tool-related factors are discussed in Sect. 8.2. 4. There are many ways of motivating people to be more creative in a particular field, thereby increasing the respective innovation potential. One strategy is to provide financial incentives for successful projects, which can be facilitated by intellectual property rights or patent law. Another approach is to increase visibility for exemplary projects in the desired areas of activity through the support for exhibitions, workshops, and other cultural events. In addition, creative projects thrive most profusely in a “maker community.” These communities can be highly domain-specific: An area that has a creative community regarding custom cars is not necessarily the same as an area that places its emphasis on creative gardening projects. Policy can provide support for maker communities. Similarly, we already discussed how administration typically reduces lifetime years. Therefore, minimizing currently existing administration demands is another way to increase lifetime years.

8.2

Measures to Increase the Availability and Use of Tools

Tool-related factors can be increased in several ways. The first important measure is to maintain what already exists. Unless tools are actively maintained (e.g., by preserving books in libraries and artifacts in museums), prior inventions often get lost (Henrich, 2004; Enquist et al., 2008; Mesoudi & Whiten, 2008; Castro & Toro, 2014). In this context, preserving tools also includes the active teaching of what exists and how to use it as part of education programs. The next important question concerns access to tools, which includes issues of infrastructure. Stores and streets allow people to purchase certain tools. “Rental stations” like libraries can even grant access for free. The tools within reach for creators determine the kind of projects they can conduct. Because the development of novel tools is typically time- and labor-intensive (Corazza & von Thienen, 2021; von Thienen et al., 2023), a next important intervention is to help society learn from solutions that already exist in other cultures (Creanza et al., 2017). Measures in this context span a variety of approaches. We see, for instance, how policy can foster friendly cultural exchange with other nations and communities. Immigrants can be encouraged to actively share their knowledge and

70

J. von Thienen et al.

tools from their country of origin. International import-export solutions and customs agreements can be designed in ways that facilitate the exchange of goods. At the same time, in some cultural areas, it is also important to clarify potential complications. Especially in the biological field, “goods” cannot be exchanged without a closer look. There is a notable danger of introducing a foreign species, which can become problematic for the native species. This means some cultural areas exist, in which the exchange of “goods” needs to be regulated in more cautious ways. When it comes to the production of novel tools in a community, again a number of measures can be taken. This may begin with political decisions to welcome creativity and innovation. Generally, interventions that increase lifetime years dedicated to creative pursuits and the project success probability while reducing the innovation resistance will eventually increase newly emerging tools. Notably, when policymakers become creative and think up novel solutions for governance, this also means adding novel solutions to the overall set of tools in the community.

8.3

Measures to Increase the Project Success Probability

A number of governance strategies can help to increase the success probability in creative projects. For instance, there can be direct project funding from the state or region. Similarly, tax burdens can be reduced to support creative work in a desired work domain. In addition, economic measures that generally increase funding in an area (via investors, banks, etc.) also benefit innovation. One important measure in this context can be laws to secure benefits for the creator in the case of successful developments (e.g., via patent law). Investors are more willing to finance creative projects when good project outcomes translate reliably into financial gains. Another key measure is to support education for creativity and innovation. Trained community members can conduct creative projects in better informed, more experienced, and ultimately more successful ways. For instance, design thinking provides such a general training to enhance capacities for creativity and innovation, irrespective of the specific project content (Plattner et al., 2009; Brown & Katz, 2009; Kelley & Kelley, 2013; Roth, 2015). In this regard, the government can assess the number of training facilities in a region, next to percentages of people from the overall population who have been educated in creativity and innovation. Furthermore, policy can help establish support systems, such as services to interconnect experts, to facilitate their exchange of insights. This can include the support on an active club landscape and/or fair events. Here, persons with interests in a creative work domain can come together to exchange ideas and experiences, as well as to help each other. Apart from these already existing approaches, there is a potential to fundamentally redesign administrative processes, turning them into creativity support systems.

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

71

This redesign can build on the insight that creativity and administration share major goals and, in this way, they can become more synergistically interconnected. Ultimately, the creative process itself is geared toward developing new and helpful products, which usually should be as harmless as possible. Goals such as banning risks are therefore typically identical in the creative process and in the administrative process. In light of this, administrative processes could become more seamlessly interconnected with ongoing creative work. For instance, by answering administrative questions, creators could see themselves making important progress in their creative work and coming closer to high-quality outcomes. These are still possibilities to be investigated in the future.

8.4

Measures to Reduce the Innovation Resistance

Since innovation resistance reduces the innovation potential in society, measures to minimize the innovation resistance benefit the potential. There are straightforward governance approaches to help counter the cultural resistance against innovation, especially the fear of novel solutions in society (cf. Sect. 7.3). These measures help people feel safe. They convey an understanding that novel products are not dangerous. Such measures include safety checks for novel products, as currently realized via administrative procedures prior to the release of novel products. Of course, finding feasible ways to conduct safety checks is a creative challenge in itself (Committee, 2016). In addition, the regulator can introduce liability law and provide a legal basis for redress. Beyond this, policy can facilitate innovation that is respectful of needs and interests across the whole society. When a creator develops inventions that directly counter the interests of other parties, resistance is the natural reaction. For instance, an invention that renders cars obsolete endangers the economic basis and investments of people who work for the automobile industry, who have bought cars, or people who have merely invested time and money in a driving license. The novel solution can also endanger personal passions and ways of experiencing joy, as when someone loves cars and driving. Affected persons will try to secure their existing benefits, typically by rejecting or even actively fighting the novel solution. Policy can facilitate the discourse in society about which values matter to whom and which needs might be important to address in all innovation projects. Beyond public debates, education matters. Design thinking, as one approach to innovation education, highlights responsibilities of the innovator (e.g., Meinel & von Thienen, 2022). Policy can help educate and prepare creators for their influential role in society. Innovators can be guided in producing those inventions that are mindful and respectful of a broad range of needs while learning to avoid producing potentially harmful new products and services. In this sense, it may even become a country’s “good mark” and reputation that they have specifically well-trained innovators whose products are especially well thought out. These products would

72

J. von Thienen et al.

be preferred on the market as it could be assumed that they have many advantages and the most minimal detrimental effects possible. Finally, policy and regulation need to provide an environment where different interests across society and their occasional clashes can be discussed productively. At least, conflicts of interest should be balanced in a civilized, tolerant, and foreseeable way—not according to the “law of the strongest.” Just having a dependable legal system at all is an important factor in this regard.

9 Conclusion and Outlook It has long been acknowledged that a person’s environment—or place—strongly impacts their prospects for creativity and innovation. This chapter makes a start in modeling such environmental impacts, with a specific focus on the role of governance. It introduces a number of methods for studying and modeling the impact of places. The TYPE method aims at calculating realistic numbers of product developments in a region. It focuses on a selective innovation domain of interest. Here, the method allows comparisons between expected product developments based on the current regulation, compared to a hypothetical scenario of deregulation. In the latter case, citizens can work creatively in the innovation domain without having to meet particular legal requirements. The innovation regulation triangle allows a characterization of governance strategies in a field between three extreme poles: (1) zero regulation, (2) maximal regulation focused on risks, and (3) maximal regulation focused on benefits. In the future, this framework could be developed further to help find ideal or desirable positions in the spectrum. A likely goal could be to find a governance approach that is open to some innovation, being mindful of risks associated with stagnation in a country. The looked-for governance approach might also integrate considerations of product risks and benefits. To illustrate the TYPE method and innovation regulation triangle, we have conducted small case studies on genetic engineering for agriculture in the European Union. Clearly, numerical estimates could be determined in even more elaborate ways, and we hope that more comprehensive case studies will emerge in the future. In the meantime, structured expert discussions have proven feasible to gain some first orienting numbers. In three such discussion rounds taking place at different times in 2022, the experts estimated that the European Union faces a trimming effect in the development of NGT products for agriculture of 98%, 99% or 99.5% within the next decade under the current regulation compared to expectable numbers in a scenario of complete deregulation. Finally, based on the computational process model, we have suggested some key figures for creativity and innovation studies. In particular, they permit the calculation of an innovation potential in society, based on the resources that are available for creative work in a particular region. This model helps to facilitate discussions and

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

73

research as to how exactly governance approaches impact creativity and innovation in society. In particular, political interventions can cause gains and losses in a number of key parameters that feed into the overall innovation potential. On this basis, we have reviewed three sample governance approaches that reduce resources for creativity and innovation in a region: (1) prohibiting the reuse of particular tools, (2) introducing administrative procedures, and (3) establishing a legal system that displays inertia. In addition, we have reviewed a number of different governance approaches that increase the innovation potential, i.e., resources for creativity and innovation in society. These are (1) measures to increase the lifetime years that citizens can invest in creative work, (2) measures to increase tool accessibility and operating knowledge among citizens, next to the diversity of tools in the region, (3) measures to help creators work more successfully in their projects, and finally (iv) measures to reduce the innovation resistance in the community. Next steps can be taken in terms of mathematical modeling approaches. Both the TYPE method and the calculation of the innovation potential are currently designed to be as simple and straightforward as possible. This has led to the choice of a static modeling approach. To complement this, it might be desirable to formulate a dynamic modeling approach, where the success probability of people working in a particular creative project increases from year to year. After all, people learn in these projects, and they make progress. Furthermore, in order to understand the regulatory impact on creativity and innovation even better, additional refinements can be introduced. For instance, the chapter addresses “liability law” merely as a keyword. Of course, there are many different options on how “liability law” can be formulated and worked out in detail, with somewhat varying effects on creativity and innovation. Here, the model provides a framework to scan systematically facilitating and hindering effects of different regulatory choices, and it hopefully supports informed governance decisions. As future developments can focus more narrowly on specific factors of interest, it is also possible to widen the scope away from the specific impact of governance. This means modeling the impact of places more generally. The basic structure used in the current approach seems suitable for reuse in other contexts. It could be adapted to model environmental impacts in a broad range of application fields, including the impact of organizational environments or the impact of class environments in education. Overall, we hope to have raised awareness for the importance of governance as a key place factor in the development of innovation. It would be a pleasure to see research that is more active, and discussion emerge in this field. Acknowledgments We thank colleague Ilia Berg for the graphic design in Figs. 1, 2, and 7, as well as Dr. Sharon Nemeth for the English proofreading.

74

J. von Thienen et al.

References Amabile, T. M. (1996). Creativity in context. Westview Press. Armitage, A., Cordova, A. K., & Siegel, R. (2017). Design thinking: The answer to the impasse between innovation and regulation. Geo. L. Tech. Rev., 2, 3. Arnold, J. E. (2016). Creative engineering. In W. J. Clancey (Ed.), Creative engineering: Promoting innovation by thinking differently (pp. 59–150). Stanford Digital Repository. Retrieved from http://purl.stanford.edu/jb100vs5745. (Original manuscript 1959). Baethge, A., & Rigotti, T. (2013). Interruptions to workflow: Their relationship with irritation and satisfaction with performance, and the mediating roles of time pressure and mental demands. Work & Stress, 27(1), 43–63. Becher, S. (2021). Re-examining the law and policy framework of new meat(s). Paper presented at the conference Sustainable Development and the Law: Potential and Challenges of Using Behavioural Insights, Hamburg University, Germany, September 23–24. Blair, C. S., & Mumford, M. D. (2007). Errors in idea evaluation: Preference for the unoriginal? The Journal of Creative Behavior, 41(3), 197–222. Bloom, B. (1985). Developing talent in young people. BoD–Books on Demand. Bourgeois-Bougrine, S., Richard, P., Burkhardt, J. M., Frantz, B., & Lubart, T. (2020). The expression of users’ creative potential in virtual and real environments: An exploratory study. In Creative learning in digital and virtual environments (pp. 96–117). Routledge. Brown, T., & Katz, B. (2009). Change by design. Harper Business. Castro, L., & Toro, M. A. (2014). Cumulative cultural evolution: The role of teaching. Journal of Theoretical Biology, 347, 74–83. Committee, E. S. (2016). Guidance to develop specific protection goals options for environmental risk assessment at EFSA, in relation to biodiversity and ecosystem services. EFSA Journal, 14(6), e04499. Corazza, G. E., & von Thienen, J. P. A. (2021). Invention. In V. P. Glăveanu (Ed.), The Palgrave encyclopedia of the possible. Palgrave Macmillan. Creanza, N., Kolodny, O., & Feldman, M. W. (2017). Greater than the sum of its parts? Modelling population contact and interaction of cultural repertoires. Journal of the Royal Society Interface, 14(130), 20170171. Cropley, D. H., Kaufman, J. C., & Cropley, A. J. (2008). Malevolent creativity: A functional model of creativity in terrorism and crime. Creativity Research Journal, 20(2), 105–115. Cropley, D. H., Kaufman, J. C., White, A. E., & Chiera, B. A. (2014). Layperson perceptions of malevolent creativity: The good, the bad, and the ambiguous. Psychology of Aesthetics, Creativity, and the Arts, 8(4), 400. Csikszentmihalyi, M. (1997). Creativity: Flow and the psychology of discovery and invention. Harper Collins. De Barro, P. J., Murphy, B., Jansen, C. C., & Murray, J. (2011). The proposed release of the yellow fever mosquito, Aedes aegypti containing a naturally occurring strain of Wolbachia pipientis, a question of regulatory responsibility. Journal für Verbraucherschutz und Lebensmittelsicherheit, 6(1), 33–40. Derex, M., Beugin, M. P., Godelle, B., & Raymond, M. (2013). Experimental evidence for the influence of group size on cultural complexity. Nature, 503(7476), 389–391. Eisenman, R. (2008). Malevolent creativity in criminals. Creativity Research Journal, 20(2), 116–119. Enquist, M., Ghirlanda, S., Jarrick, A., & Wachtmeister, C. A. (2008). Why does human culture increase exponentially? Theoretical Population Biology, 74(1), 46–55. Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100(3), 363. Guegan, J., Nelson, J., & Lubart, T. (2017). The relationship between contextual cues in virtual environments and creative processes. Cyberpsychology, Behavior and Social Networking, 20(3), 202–206.

Predicting Creativity and Innovation in Society: The Importance of Places,. . .

75

Hao, N., Tang, M., Yang, J., Wang, Q., & Runco, M. A. (2016). A new tool to measure malevolent creativity: The malevolent creativity behavior scale. Frontiers in Psychology, 7, 682. Hayes, J. (1989). The complete problem solver. Routledge. Heidenreich, S., & Spieth, P. (2013). Why innovations fail—The case of passive and active innovation resistance. International Journal of Innovation Management, 17(05), 1350021. Hennessey, B., Moran, S., Altringer, B., & Amabile, T. M. (2015). Extrinsic and intrinsic motivation. Wiley Encyclopedia of Management, 1–4. Henrich, J. (2004). Demography and cultural evolution: How adaptive cultural processes can produce maladaptive losses—The Tasmanian case. American Antiquity, 69(2), 197–214. Katz, B. M. (1990). Technology and culture: A historical romance. Alumni Association. Kaufman, J. C., & Beghetto, R. A. (2009). Beyond big and little: The four c model of creativity. Review of General Psychology, 13(1), 1–12. Kaufman, S. B., & Kaufman, J. C. (2007). Ten years to expertise, many more to greatness: An investigation of modern writers. The Journal of Creative Behavior, 41(2), 114–124. Keller, A. C., Meier, L. L., Elfering, A., & Semmer, N. K. (2020). Please wait until I am done! Longitudinal effects of work interruptions on employee well-being. Work & Stress, 34(2), 148–167. Kelley, T., & Kelley, D. (2013). Creative confidence: Unleashing the creative potential within us all. Currency. Kolodny, O., Creanza, N., & Feldman, M. W. (2015). Evolution in leaps: The punctuated accumulation and loss of cultural innovations. Proceedings of the National Academy of Sciences, 112(49), E6762–E6769. Kuhn, S. L. (2012). Emergent patterns of creativity and innovation in early technologies. Developments in Quaternary Sciences, 16(69–87), 2022. McKee, H. A. (2022). Automatic detection of the flow mental state in the context of creative collaboration. Hasso Plattner Institute for Digital Engineering. Meinel, C., & von Thienen, J. P. A. (2022). Design thinking – Enabling digital engineering students to be creative and innovate. In Design thinking in education (pp. 9–23). Springer. Mesoudi, A., & Whiten, A. (2008). The multiple roles of cultural transmission experiments in understanding human cultural evolution. Philosophical Transactions of the Royal Society B: Biological Sciences, 363(1509), 3489–3501. Mitchell, H. J., & Bartsch, D. (2020). Regulation of GM organisms for invasive species control. Frontiers in Bioengineering and Biotechnology, 7, 454. Mithen, S. (Ed.). (2005). Creativity in human evolution and prehistory. Routledge. Mizuno, K., Tanaka, M., Yamaguti, K., Kajimoto, O., Kuratsune, H., & Watanabe, Y. (2011). Mental fatigue caused by prolonged cognitive load associated with sympathetic hyperactivity. Behavioral and Brain Functions, 7(1), 1–7. Mueller, J. S., Melwani, S., & Goncalo, J. A. (2012). The bias against creativity: Why people desire but reject creative ideas. Psychological Science, 23(1), 13–17. Perchtold-Stefan, C. M., Fink, A., Rominger, C., & Papousek, I. (2021). Creative, antagonistic, and angry? Exploring the roots of malevolent creativity with a real-world idea generation task. The Journal of Creative Behavior, 55(3), 710–722. Plattner, H., Meinel, C., & Weinberg, U. (2009). Design thinking. Mi-Fachverlag. Ram, S. (1987). A model of innovation resistance. ACR North American Advances. Richard, P., Burkhardt, J. M., Lubart, T., Bourgeois-Bougrine, S., & Barré, J. (2018). The effect of virtual environment and user/designer collaboration on the creative co-design process. In Congress of the international ergonomics association (pp. 605–614). Springer. Roth, B. (2015). The achievement habit. Harper Collins. Schmidt, S. M., Belisle, M., & Frommer, W. B. (2020). The evolving landscape around genome editing in agriculture: Many countries have exempted or move to exempt forms of genome editing from GMO regulation of crop plants. EMBO Reports, 21(6), e50680. Simonton, D. K. (1997). Creative productivity: A predictive and explanatory model of career trajectories and landmarks. Psychological Review, 104(1), 66.

76

J. von Thienen et al.

Simonton, D. K. (2000). Creative development as acquired expertise: Theoretical issues and an empirical test. Developmental Review, 20(2), 283–318. Sonnentag, S., & Frese, M. (2012). Dynamic performance. In S. W. Kozlowski (Ed.), The Oxford handbook of organizational psychology (Vol. 1). Oxford University Press. von Thienen, J. P. A., Noweski, C., Rauth, I., Meinel, C., & Lang, S. (2012). If you want to know who you are, tell me where you are: The importance of places. In Design thinking research (pp. 53–73). Springer. von Thienen, J. P. A., Kolodny, O., & Meinel, C. (2023). Neurodesign: The biology, psychology and engineering of creative thinking and innovation. In N. Rezaei (Ed.), Brain, decision-making, and mental health (pp. 617–659). Springer. von Thienen, J. P. A., Ney, S., & Meinel, C. (2019). Estimator socialization in design thinking: The dynamic process of learning how to judge creative work. In Dynamic perspectives on creativity (pp. 67–99). Springer. Weiner, E. (2016). The geography of genius: A search for the world's most creative places from ancient Athens to Silicon Valley. Simon and Schuster. Wilhelm, R., Bartsch, D., Consmüller, N., de Witte, T., Sprink, T., et al. (2021). Bericht zu möglichen Synergien der Nutzung neuer molekularbiologischer Techniken für eine nachhaltige Landwirtschaft. Berichte aus dem Julius Kühn-Institut. Zajonc, R. B. (2001). Mere exposure: A gateway to the subliminal. Current Directions in Psychological Science, 10(6), 224–228.

An Exploration of Agile Governance in Rwandan Public Service Delivery Reem Abou Refaie, Lena Mayer, Karen von Schmieden, Hanadi Traifeh, and Christoph Meinel

Abstract While derived from software engineering, “agile” as a concept has become increasingly sought-after among public sector organizations in the last decade as the means to becoming more responsive, flexible, adaptive, and rapid in their behavior. However, there is a limited understanding of what it means to be agile in the public organization context. How do public sector organizations change when agile approaches are introduced. This chapter contributes to the literature by exploring the practitioners’ understanding and approach to agile governance and identifies the knowledge gaps facing their applications and practices. To carry out our investigation, we conducted in-depth semi-structured interviews (n = 14) with public service providers from Rwanda’s “one-stop shop” platform, where agile methods are used in the development of digital public services.

1 Introduction As public organizations strive toward flexibility, adaptability, and seamless service delivery, many governments have adopted new working practices from (strategic) design to agile governance practices (Mergel et al., 2020; Gong et al., 2020). Agile working practices often focus on design-led processes such as prototyping, experimentation, and user testing in iterative loops, relying on epistemological frameworks from action research and ethnography. Hence, these practices depart from or transition away from linear processes of research, validation, implementation, assessment, and adaptation (Learmonth & Harding, 2006). Agile governance also incorporates novel organizational, cultural, and mindset elements—such as tolerance of failure— which depart from traditional Weberian bureaucracy logic. Additionally, agile governance challenges elements of strict individual performance management and

R. A. Refaie (✉) · L. Mayer · K. von Schmieden · H. Traifeh · C. Meinel HPI School of Design Thinking, Hasso Plattner Institute for Digital Engineering, Potsdam, Germany e-mail: [email protected]; [email protected]; offi[email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_4

77

78

R. A. Refaie et al.

performance appraisal put forth by the new public management. Instead agile governance propagates more collaborative and team-driven managerial approaches as well as iterative evaluation cycles that hallmark people and their creativity. As an emerging area of research and practice, agile governance has its roots in the software engineering domain. However, agile governance practices go beyond agile software development to cover other domains such as agile project management, agile acquisition, agile evaluation, and agile regulation (Mergel et al., 2018). Created as an alternative to waterfall approaches, “The Agile Manifesto” (Beck, 2001) stressed collaboration, adaptation, and iterative reviews rather than sequential and linear steps (Beck, 2001). In this sense, agile values are useful approaches in an era of rapid change. More recently, public organizations began to realize the added value of agile beyond software development, which led to the rise of “agile” as a new way of governance that academics and policymakers are grappling with. Examples of this increased interest include the Agile Nations Charter (2020) spearheaded by the UK government and the Agile Regulation Toolkit issued by the World Economic Forum (2020). Described as an “evolving concept” (Mergel et al., 2020, p. 161), agile governance remains an opaque concept and a contested practice. With the exception of Mergel et al. (2020), and McBride et al. (2021), who critically engaged in what “agile” is, and what it is not, agility has not yet been fully captured in the governmental context. In Mergel’s (Mergel’s, 2016; Mergel’s et al., 2020) view, agile governance is “a holistic concept that does not refer to an isolated area of agility, such as software development or project management.” On the other hand, McBride’s study makes the distinction between “being agile” and “doing agile.” In this sense, the study argued that agility is a modifier to government, which already exhibits agility. Based on the previous literature, we argue that gaining a deeper understanding of agile governance requires the consideration of the entanglement of three different domains (i.e., technology, innovation, and organization). We address this entanglement from a practice approach; thus, our unit of analysis is the “doings and sayings of entangled configurations through which phenomena are produced” (Schultze et al., 2020, p. 817). A practice approach is needed to “explain agile transitions as an ongoing, continuous, long-term transformation, rather than clearly circumscribed stages of agile adoption” (Hoda & Noble, 2017 p. 141). We also draw from institutional theory, where we define “institutional logics” as “socially constructed, historical patterns of cultural symbols and material practices, assumptions, values and beliefs by which individuals produce and reproduce their material subsistence, organize time and space, and provide meaning to their daily activity” (Thornton et al., 2013, p. 50). We conducted in-depth semi-structured interviews (n = 14) with public service providers from Rwanda’s “one-stop shop” platform (iremboGov, 2022), where agile methods are used in the development of digital public services. Using an inductive qualitative research design, this chapter identifies understandings of agile governance among public service providers, the agile methods they use, and how they are implemented.

An Exploration of Agile Governance in Rwandan Public Service Delivery

79

RQ1: What characterizes agile governance approaches in the public service providers’ perspectives and practices? RQ2: How do public organizations change when agile methods are introduced? We contribute to the literature by offering a better understanding of how public service providers perceive agile governance and how these perspectives are translated into work practices, which can lead to the reshaping of practices at the organizational level. Based on our findings, practitioners can gain insights on how to better leverage agile governance approaches to develop digital public services. The chapter is structured as follows. Section 2 presents an overview of digital transformation, agile governance, and institutional logic presenting the theoretical background and existing related works. In Sect. 3, we describe the methods applied, the methodological quality, data collection, and data management and analysis, while Sect. 4 describes the case study context. In Sect. 5, we present the results of this study, addressing the research question and discussing theory usage. Section 6 concludes, discussing contributions and study limitations and offering recommendations for subsequent research on agile governance.

2 Related Work Agile governance has an interdisciplinary character, therefore requiring a combination of theoretical backgrounds. We draw from two relevant bodies of knowledge: (1) digital transformation in the literature of public organizations to understand the organizational context in which agile governance is implemented in light of digital technologies’ impact on public organizations and (2) institutional logic theory, which is relevant in understanding agile governance as a new and competing logic and practice.

2.1

Digital Transformation and the Rise of Agile Governance in Public Organizations

Public organizations are experiencing immense institutional changes as a result of digital transformation, and implementing agile governance is one way they have attempted to manage these changes. In the context of public services, the development of digital public services requires redesigning and reengineering government services from the ground up to fulfill changing user needs. At the center of these efforts are users—both internal and external users—of digital services who are included in the digital transformation efforts. Such organizational change or transformation is grounded in the ongoing perceptions and practices of organizational actors and emerges out of their explicit or implicit actions and experiments with the everyday possibilities, failures, opportunities, and unintended consequences that

80

R. A. Refaie et al.

they encounter. To have any effect, reform initiatives must ultimately result in changes in the work processes of public organizations and in the attitudes and behavior of employees who work in these organizations (Ahmad et al., 2020). Therefore, to make sense of the introduction of agile as a “new way of governing” (Mergel et al., 2020), in the context of digital transformation in public organizations, we draw from institutional logic theory.

2.2

Adoption of Agile Methods as a New Institutional Logic and Practice

Agile adoption or the implementation of agile methods is a relatively new phenomenon in public management. Previous research (e.g., Patanakul & Rufo-McCarron, 2018) from digital engineering has been overly focused on how specific agile methods are adopted and how high-level agile properties were rolled out in digital government projects. This understanding of agile adoption often follows multistage adoption models, which put forth a predefined set of agile practices over time. These are process-heavy and inflexible and ignore contextual differences across jurisdictions and policy sectors. Departing from sequential prescriptive adoption frameworks for the understanding of agile governance, and drawing on institutional logic theory, we study agile governance as an organizing principle for an organizational field, which is made up of a community of actors grouped by their joint values and beliefs (Friedland & Alford, 1991; Scott et al., 2000). In this sense, “logics are important in understanding institutional change because a change in the field’s dominant logic is fundamental to conceptualisations of institutional change” (Reay & Hinings, 2009 p. 630). They form the basis of the guiding behavior of field-level actors, and they refer to the belief systems, perceptions, and related practices that dominate an organizational field (Scott, 2014). A core proposition of institutional logics is that interests, identities, values, and assumptions of individuals are embedded within the prevailing institutional logic. In this perspective, institutional change is understood as the movement from one dominant logic to another (e.g., Greenwood et al., 2002; Hoffman, 2017). Therefore, to understand agile transition in the context of public sector organizations, we must be concerned with the perceptions, practices, and beliefs of the people within the target organization. In applying institutional logic theory to organizational change, previous studies have shown how a new logic may be introduced to a field and become dominant— providing new guidance for field members (Kitchener, 2002; Hensmans, 2003; Scott et al., 2000). This is the case for many government reforms that proceeded agile governance, e.g., new public management. Others have also shown that the introduction of a new logic can instigate tension with the existing dominant logic as was illustrated in the case of Scott (2014), who pointed out the inherent rivalry when

An Exploration of Agile Governance in Rwandan Public Service Delivery

81

multiple logics exist. The study proposed that many organizational fields are characterized by two or more competing belief systems. Studies also proposed that conflicting logics coexist during transition times until one side or the other wins and the field reforms around the winning dominant logic (Hensmans, 2003; Hoffman, 2017) or a new logic that is a hybrid version of two previous logics (Glynn & Lounsbury, 2005; Thornton et al., 2013). Our goal is to identify agile governance as a new institutional logic in public organizations that has the potential of becoming a dominant logic. This chapter explores how field-level actors—in this case, public service providers—perceive agile governance both as a concept and how they engage with it as a work practice. We are also interested in the impact of this new institutional logic on the organizational culture and practices.

3 Methodology Our research design followed an inductive qualitative research using a single case study approach (Yin, 2008). We analyzed the case study to gain an in-depth knowledge of the practitioners’ understanding and approach to agile governance and to identify the knowledge gaps facing their applications and practices. A qualitative approach to researching agile governance provides an opportunity to concurrently explore individual perspectives and practices as well as broader processes and contextual factors (Bryman, 2004). The case study was drawn from the Central African country of Rwanda, where we explore the provision of public service delivery. We chose this case for two main reasons: (1) Despite theoretical and empirical developments in this direction, insufficient attention has been paid to developing country contexts, and (2) access to firsthand insights on their doings and sayings with regard to the public sector’s utilization of agile approaches to develop digital services. The selection of Rwanda’s one-stop shop “IremboGov” was also made on the basis of high expectations about the information content it would provide. For our study, we sought an organization that was set up to produce high-end, seamless digital solutions for citizens and businesses. The second criterion was that the chosen organization uses agile approaches to develop digital public services. The third criterion was that the case should include collaboration between several public agencies and sectors, which can give us access to different perspectives. A fourth and final criterion when selecting the case was that it has had been running for some time (8+ years), thus having the potential to yield information about agile experiences. The lead author conducted an in-depth case study of Rwanda’s public service delivery system, in the context of a research study with IremboGov, the country’s one-stop shop platform for delivering public services online. In 2014, the Irembo project was established as a public-private partnership (PPP) initiative by the

82

R. A. Refaie et al.

government of Rwanda (GoR) together with RwandaOnline Platform Ltd. (now Irembo) to digitalize all public services into a single window platform.

4 Data Collection and Analysis To obtain an adequate set of interviewees, we used purposive and snowball sampling (Van Thiel, 2022). We used this sampling to recruit interviewees from various central level administrative positions across seven main departments, namely, engineering, product and user support, partnerships, strategy, communication, administration, and human resources (n = 14). The first round of semi-structured interviews were conducted in English during December/January 2020/2021, and the second round took place in January 2022. Due to the limitations imposed by the pandemic, all interviews were conducted remotely via Zoom and recorded (audio and video) with the informed consent of the interviewees and transcribed using Otter.ai, a transcription software. To triangulate our findings, we utilized multiple data sources including digital regulations and national strategies. Especially relevant for this chapter was the observation of monthly governance cluster meetings conducted by the first author during a research stay in Kigali, Rwanda, in 2018/2019. Recognizing the nascent stage of topic knowledge, we adopted a predominately inductive approach. We collected and analyzed data iteratively, shifting between empirical data and theoretical concepts in a cycle between interviewing, transcribing, analyzing, and checking back with the theoretic body of knowledge and our focus on agile governance perceptions and practices. To analyze the data, we pursued a stepwise coding, which consisted of open, axial, and selective coding using Atlas.ti (Gioia, 2020). After the first round of interview write-ups and summaries, we employed the open-coding stage to generate first codes, which were used to condense the transcripts and obtain an initial overview of all case data (Yin, 2008).

5 Findings This section contains the findings gained from the analysis of the data of the 14 experts interviewed. They are supported by quotes from the interviews as this illustrates the experts’ different perspectives, helps to enhance the transparency of the research process, and enables other researchers to follow the reasoning. We asked the interviewees what they believe agile governance is like. Many noted that above all agile governance is about adaptive and collaborative management, an experimentation mindset, and delivering services that respond to the needs of citizens.

An Exploration of Agile Governance in Rwandan Public Service Delivery

5.1

83

Adaptive and Collaborative Management

Many explained that agile governance is about having political leadership that exhibits a strong willingness to collaborate with other key stakeholders in order to achieve a shared vision and coordinate joint action in order to achieve it: “We discussed this in quite a lot of different African governments. And I feel like in Rwanda, there is a mix of this. In Rwanda, when we want to achieve something, everyone tries to go in the same direction” (P14). Agile governance is when the leadership allows for quick changes in direction and guides the response to a fast-changing context with various partners and workflow interdependence: “You never know what this is going to be because maybe you had in mind that you wanted to add new services or you wanted to redefine the experience for this specific service. Then there is an incident with the payment platform because your partner bank had an infrastructure incident. This is what happened recently. There is a fire in the house, and you cannot redesign anything, and you should just focus on doing the minimal, which is fixing the issues” (P9). At the project management level, many noted that short projects and flexible deadlines play a very important role by providing the framework within which decisions are made for project development and implementation to achieve the intended strategic goals. This is in line with the transition from sequential to concurrent product development. This process/method significantly shortens development cycles as involved in agile software development and lean management. One administrator pointed out that agile governance entails having short targeted projects, which allow practitioners to pivot implementation strategies while still working toward the original objective of the project and the wider organization. We don’t have long projects with things that are set in stone; it’s more like things are moving. Even though we have a common objective, the way to achieve it can change. (P1).

Another set of project management practices that are associated with agile governance are having regular team meetings, where priorities are discussed and altered depending on the most recent developments. The practice of weekly meetings or sprints was highlighted by several interviewees. And if I look at how it works currently, they have the product and engineering team, for instance. They have weekly sprints, and they focus on what should be the priority for the week. (P9).

Most of the respondents pointed that in practice, adaptive and collaborative management require open communication and regular team exchanges and discussions. One respondent emphasized the usage of digital collaboration tools to maintain smooth and flexible communication, which plays a dominant role in enabling adaptability and flexible decision-making. By having short deadlines and teams that are communicating a lot, I would say, yeah, on a daily basis, of course, using the right tools, for instance, WhatsApp, JIRA, Slack, etc., is essential. This way we ensure that whenever some things pop up, we can reprioritize. (P13)

84

R. A. Refaie et al.

Another respondent added that this regular exchange via digital communication tools allows managers/leadership to be in constant touch with the rest of the team. This constant interaction provides the opportunity to make the necessary high-level decisions in collaboration with the relevant stakeholders: The tools that enable the cooperation and strong leadership that can also do this trade-off because it’s not always the junior engineer that is going to be able to know what he or she should focus on solving. (P8)

In the same vein, another respondent clarified that a key practice in adaptive management is to hold open team discussions in order to reach a common understanding before a course of action is determined. It’s not like there’s no discussion. We always give our input as to what we are requested to do. There’s always some discussion. But after we’ve come to a common understanding, it’s imperative that we deliver that very quickly. (P2).

5.2

Organizational Culture of Experimentation and Iteration

When asked to explain agile governance, the majority of the respondents spoke about elements of repeating processes and specific tasks to reach better results, as well as remaining curious, trying something new, incorporating the learnings, and trying again. In this sense, agile governance is perceived as an organizational culture that allows mistakes and errors and ensures the use of new methods of iterations and experimentation. One respondent explained iteration as a way of working, whereby developing digital services undergo continuous improvement based on gathering feedback and implementing swift changes: “It’s not really like you do something and then you put it out there and then that’s it. It’s a continuous process of gathering feedback and then taking action on the feedback as quick as possible” (P1). Most of the interviewees described their work as guided by an experimentation mindset within an agile environment. Most highlighted experimentation as a remedy to the complexity of the work done by public service providers in Rwanda’s digital public service delivery system. Although some praised experimentation as a way to take pressure off of public administrators as they are able to iterate on the way they do things, there was still a reference to fear of getting something wrong. This implies a lingering culture of risk aversion yet signals a transition to a tolerance of failure as iteration is embraced. I would say that the complex work we do is supported by working in agile methodology as well as an agile environment, which helps us to remain experimental and curious. Also it removes the pressure from getting something wrong, as I can constantly recheck and iterate on the way I do my work and the way I collect the data. (P3)

Most of the interviews highlighted agile governance in the context of iterative problem analysis, whereby one respondent elaborated on the problem-analysis approach by continuously asking whether the solutions proposed add value to citizens.

An Exploration of Agile Governance in Rwandan Public Service Delivery

85

We go through the problem, and we ask ourselves how can this work better for the citizens? We talk to the product managers, they put it in writing, they go back to the developers, and then, after the development of the service, I come back to check it from the citizens’ point of view. After these steps, we can then launch the improvements. (P2)

Applying iteration to the public service design process, one respondent explained that iteration also involves going further into problem analysis, stepping beyond the high-level symptom and really examining the root cause from various perspectives before a solution is tested. I can go deeper into a problem and try to analyze what happened. Is this the way the service is designed? Is it the result of a mistake we made while in the development journey? That’s where I go deeper and try to talk to product managers to develop. Why is it like this? Is it our issue? Is it our partners? You know, we don’t do everything on our own; we have partners. Sometimes they also have problems and issues that can affect the public. We also have to go through that cycle to check where the problem is. (P10)

Remunerating ways to dig deeper into problem analysis, another respondent recalled extracting data from calls and service reports. This helps to zero in on recurrent issues and at the same time provides leverage to push for changes vis-à-vis the engineering team. We analyze our calls, and we generate reports to see the highlights. Then we drill down into those issues and say, why is this particular issue coming back? Then we’ll go back and look at the way the service is designed and see if there’s any particular thing that keeps coming back to maybe make it harder for citizens to apply. We then sit with the product team to tell them, we propose that the changes are made this way, you can improve this particular service, because we have numbers to back it up. (P12)

As an integral part of an iterative problem-solving process, respondents also mentioned testing solutions before implementation. One respondent explained that a key aspect of testing is leveraging the diverse mandates and areas of expertise within the team. For example, he distinguished between the engineering team members and the other departments who do not understand the codes behind the services and will, therefore, only be concerned with usability of the service. From the technical side, we come up with a possible solution to what we call a testing environment where production team people that don’t know but codes in everything can just use the front end, yeah, to test now, is this service working perfectly. After validation, they ship it to production and that’s when citizens can start using it. (P12)

In reference to the process of designing digital public services, another engineer concurred by describing the testing environment in detail: “After doing our technical tests, development testing environment, where we bring in products, partnership operation, then we show them how what we developed works. So, from that operation and partner partnership, department will be like, oh, okay, so this is how things work. We show them and then they are the ones who go to the agents and other partners to show them” (P8). One respondent stated that testing environments are really about looking at the service from the citizen-user’s point of view. In this example, the engineer

86

R. A. Refaie et al.

empathizes with the citizen by putting himself in the citizen’s shoes and roleplaying an alternative scenario in order to reveal issues with the service. We do this both ways because there is this path of testing as an engineer as you get into codes. And also there’s this other part that you work out as an end user. Even if you read the requirements, you ask how it fits the needs of an end user. Do you understand it as a person, you think the end user (who doesn’t have an idea of what you’re doing) will understand it? The way you understand it? That’s my role. (P9)

Another respondent highlighted testing environments as a form of monitoring where the partnership department represents the needs of their clients vis-a-vis the engineering team, thereby ensuring that the designed digital services are iterated in ways that deeply reflect the citizens’ needs. This implies an act of balancing the different perspectives within one team in order to optimize the designed services: “I would say that, by having this strong partnership and user ops team. The partnership team is close to the partners, and the UserOps team, close to the day-to-day needs of the citizens. This helps to put pressure on the engineering team to say, let’s not just do it, because it’s easier for you or because you think it makes sense but listen to what we tell you and what the partners did, and what the user said.” (P6).

5.3

Citizen Orientation and Engagement

Central aspects of agile governance are the citizen engagement and user-centricity of services, which are supported by design tools that allow continuous interaction with citizens to better meet their needs. One respondent candidly explained it: “We just want to make the process better and better. The question is how we deliver more citizen centric services, to the point that you know, even the hardest of services today can still be simpler to access” (P12). Elaborating on this, several respondents described Irembo as “an agile company, because one of our main goals is to delight our users, who are mainly citizens” (P8). However, the multiplicity of users makes the task of delighting all users a difficult one. A product designer working for Irembo emphasized that despite the strides made with regard to enhancing the accessibility and usability of digital services, limited digital knowledge and literacy remain a common issue faced by many citizens. “Before the design of the services was really complicated, even if some people in a developed country don’t really have the digital knowledge, and literacy is really low. Even applying for a small service was quite demanding for some people. So the idea behind that was just to make a service really simple for even those without much experience in applying for a service” (P5). Nevertheless, the product designer reaffirmed that a dominant perception that drives public service providers is the design of “more seamless, more accessible and inclusive services for everyone” (P5).

An Exploration of Agile Governance in Rwandan Public Service Delivery

87

Achieving this, however, necessitates ongoing consultations and frequent course corrections in order to maintain effective service delivery. This was reflected in many of the respondents’ narratives pointing to the fact that in practice: “For me, I think that as an agile organization, the government of Rwanda did the right thing in designing the services, whereby the partners [private sector, government institutions] were consulted, meaning that they let us know what they wanted, the services they wanted, and on our end, we presented what citizens wanted. We came up with an agreement for service provision” (P5). Citizen orientation is embedded in an assessment of satisfaction and an evaluation of the citizens’ response to digital public services. One respondent explained that the design of the digital services is dictated by how the engineering team responds to citizen feedback. This often takes the form of adjusting the design, features, and functionality of the services, at times even at a drastic level. Whatever we do, we are first looking at how the citizen responds to this. With that in mind, you will find that from a software engineering perspective, we find ourselves changing too much. Today we’re doing something and then when we put it out there; it’s not fully received as we expected, and then there is change to something else. And there’s also a lot of research involved, because we need to make sure what we deliver is actually what citizens want. (P10)

To do this, another respondent elaborated on how they gather data through social media in order to assess citizen satisfaction with the digital public services. “What we usually get out of the interaction is that it gives us a general picture of how people feel about the product, and then we’re able to go back and improve it. The feedback that we get from radio and social media leads us back to the drawing board. There we discuss some of the issues that we found on ground with the product team and our partner institutions” (P14). Outcomes of such assessments have a direct effect on the digital public service delivery strategy and infrastructure. One respondent recalled that despite having a one-stop shop for all public services, public feedback revealed that people still perceive applying for services as cumbersome. In response to this, Irembo established an intermediary system of around 4000 agents around the country with the mandate to support citizens with accessing digital services in exchange for a small fee. We had many issues; people would feel uncomfortable to apply for services for themselves. We had to go through Irembo agents, people whose jobs were just to assist Irembo. For us, we were okay with that. Because if people were not comfortable and they paid a little money to those people to apply for them, that was okay. (P13)

6 Discussion and Conclusion This study set out to investigate what characterizes the new institutional logic of agile governance in the public service providers’ perspectives and practices.

88

R. A. Refaie et al.

Our findings provide vivid examples in feedback comments that an agile approach to governance supports the Rwandan public service provider’s desire to show that they can respond and adapt to changes in the technological and market environments. This includes citizens’ increased expectations for efficient and effective online services. This is translated into a narrative on agile governance as a bundle of methods, a public service management approach, and a service delivery logic. Results revealed that as a new institutional logic in the digital public services system, agile governance was clearly prevalent in the perception and practices of field-level actors. However, the extent to which this can be characterized as fundamental change in governance requires further investigation over a longer period, which allows for temporal comparisons. An important limitation is that our findings reflect the specific situation in Rwanda and should be validated in other countries with different state structures, cultures, and maturities in e-government. Additionally, we would like to address the issue of the long-term adoption of agile government. Due to the situated and discourse nature of our findings, it is difficult to predict the long-term perceptions and practices associated with agile governance. Future research is necessary to characterize how those manifested patterns of agile practices in public organizations develop and evolve over time. It should also consider factors like level of agile knowledge, digital literacy, and size of the organization. Upcoming research can further investigate the circumstances leading to agile governance being a successful approach for digitalization. Acknowledgments We thank the HPI-Stanford Design Thinking Program and our colleagues in Rwanda for enabling this study and research collaboration. Many thanks to Dr. Sharon Nemeth for copyediting and language support.

References Agile Nations Charter (Accessible Webpage Version). (2020). GOV.UK. November 25, 2020. https://www.gov.uk/government/publications/agile-nations-charter/agile-nations-charter-acces sible-webpage-version Ahmad, A. B., Straatmann, T., Mueller, K., & Liu, B. (2020). Employees’ change support in the public sector—A multi-time field study examining the formation of intentions and behaviors. Public Administration Review, 18(2), 231. https://doi.org/10.1111/puar.13275 Beck, K. (2001). Manifesto for Agile Software Development. Agilemanifesto.org. Agile Alliance. 2001. http://agilemanifesto.org/ Bryman, A. (2004). Social research methods (2nd ed.). Oxford University Press. Friedland, R., & Alford, R. R. (1991). Bringing society back In: Symbols, practices, and institutional contradictions. In P. J. DiMaggio (Ed.), The new institutionalism in organizational analysis (pp. 232–263). University of Chicago Press. Gioia, D. (2020). A systematic methodology for doing qualitative research. The Journal of Applied Behavioral Science, 57(1), 20–29. https://doi.org/10.1177/0021886320982715

An Exploration of Agile Governance in Rwandan Public Service Delivery

89

Glynn, M. A., & Lounsbury, M. (2005). From the critics’ corner: Logic blending, discursive change and authenticity in a cultural production system*. Journal of Management Studies, 42(5), 1031–1055. https://doi.org/10.1111/j.1467-6486.2005.00531.x Gong, Y., Yang, J., & Shi, X. (2020). Towards a comprehensive understanding of digital transformation in government: Analysis of flexibility and Enterprise architecture. Government Information Quarterly, 37(3), 101487. https://doi.org/10.1016/j.giq.2020.101487 Greenwood, R., Suddaby, R., & Hinings, C. R. (2002). Theorizing change: The role of professional associations in the transformation of institutionalized fields. Academy of Management Journal, 45(1), 58–80. https://doi.org/10.5465/3069285 Hensmans, M. (2003). Social movement organizations: A metaphor for strategic actors in institutional fields. Organization Studies, 24(3), 355–381. https://doi.org/10.1177/ 0170840603024003908 Hoda, R., & Noble, J. (2017). Becoming agile: A grounded theory of agile transitions in practice. In 2017 IEEE/ACM 39th international conference on software engineering (ICSE), May, 141–151. https://doi.org/10.1109/icse.2017.21 Hoffman, A. J. (2017). Institutional evolution and change: Environmentalism and the US chemical industry. SSRN Electronic Journal, 42(4). https://doi.org/10.2139/ssrn.2940277 “IremboGov.” (2022). Irembo.gov.rw. 2022. https://irembo.gov.rw/home/citizen/all_services Kitchener, M. (2002). Mobilizing the logic of managerialism in professional fields: The case of academic health centre mergers. Organization Studies, 23(3), 391–420. https://doi.org/10.1177/ 0170840602233004 Learmonth, M., & Harding, N. (2006). Evidence-based management: The very IDEA. Public Administration, 84(2), 245–266. https://doi.org/10.1111/j.1467-9299.2006.00001.x McBride, K., Kupi, M., & Bryson, J. (2021). Untangling agile government: On the dual necessities of structure and agility. In M. Stephens, R. Awamleh, & F. Salem (Eds.), Agile government: Concepts and practice for future-proof public administration. World Scientific Publishing. Mergel, I. (2016). Agile innovation management in government: A research agenda. Government Information Quarterly, 33(3), 516–523. https://doi.org/10.1016/j.giq.2016.07.004 Mergel, I., Ganapati, S., & Whitford, A. B. (2020). Agile: A new way of governing. Public Administration Review, 81(1), 161–165. https://doi.org/10.1111/puar.13202 Mergel, I., Gong, Y., & Bertot, J. (2018). Agile government: Systematic literature review and future research. Government Information Quarterly, 35(2), 291–298. https://doi.org/10.1016/j.giq. 2018.04.003 Patanakul, P., & Rufo-McCarron, R. (2018). Transitioning to agile software development: Lessons learned from a government-contracted program. The Journal of High Technology Management Research, 29(2), 181–192. https://doi.org/10.1016/j.hitech.2018.10.002 Reay, T., & Hinings, C. R. (2009). Managing the rivalry of competing institutional logics. Organization Studies, 30(6), 629–652. https://doi.org/10.1177/0170840609104803 Schultze, U., van den Heuvel, G., & Niemimaa, M. (2020). Enacting accountability in IS research after the Sociomaterial turn(Ing). Journal of the Association for Information Systems, 21, 811–835. https://doi.org/10.17705/1jais.00620 Scott, W. R. (2014). Institutions and organizations: Ideas, interests and identities (4th ed.). Sage. Scott, W. R., et al. (2000). Institutional change and healthcare organizations: From professional dominance to managed care. University Of Chicago Press. Thornton, P. H., Lounsbury, M., & Ocasio, W. (2013). The institutional logics perspective: A new approach to culture, structure, and process. Oxford University Press. van Thiel, S. (2022). Research methods in public administration and public management: An introduction. In Routledge masters in public management (2nd ed.). Routledge. World Economic Forum. (2020). Agile regulation for the fourth industrial revolution: A toolkit for regulators. World Economic Forum. December 2020. https://www.weforum.org/about/agileregulation-for-the-fourth-industrial-revolution-a-toolkit-for-regulator Yin, R. K. (2008). Case study research: Design and methods (4th ed.). Sage Publications Ltd.

Voices from the Field: Exploring Connections Between Design Thinking Approaches and Sustainability Challenges Nicole M. Ardoin, Alison W. Bowers, and Daniella Lumkong

Abstract The range and complexity of the United Nations’ 17 Sustainable Development Goals make clear the need for innovative, effective, and efficient problemsolving approaches involving the people most impacted by the issues being addressed. In previous work, we argued that design thinking provides a useful and appropriate process for generating sustainability-oriented solutions and outlined five characteristics of design thinking that support this argument. In this chapter, we present findings from 20 semi-structured, qualitative interviews with sustainability practitioners, designers, and university professors from 9 countries who have used design thinking approaches in their work. Based on a deductive coding analysis of the interview data, we found support for all five of our suggested design thinking characteristics as facilitators of sustainability solutions. Nearly all interviewees appeared to value design thinking as a tool for sustainability work because design thinking is participatory and people-focused. Many interviewees also emphasized that design thinking encourages creativity, diversity in thought and action, systems thinking, and a streamlined approach to action. These findings add further support for calls to use design thinking as a tool in sustainability work and suggest a need for more in-depth exploration to solidify a theoretical understanding of why, how, and

N. M. Ardoin (✉) Social Sciences Division, Doerr School of Sustainability, Stanford University, Stanford, CA, USA Emmett Interdisciplinary Program in Environment and Resources, Stanford University, Stanford, CA, USA Stanford Woods Institute for the Environment, Stanford University, Stanford, CA, USA e-mail: [email protected] A. W. Bowers Social Sciences Division, Doerr School of Sustainability, Stanford University, Stanford, CA, USA e-mail: [email protected] D. Lumkong School of Engineering, Stanford University, Stanford, CA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_5

91

92

N. M. Ardoin et al.

under what conditions design thinking approaches are effective in the quest for addressing sustainability challenges.

1 Introduction In 2015, the 193 United Nations member states approved the 2030 Agenda for Sustainable Development (United Nations, 2015). This ambitious document served as a call to action to developing and developed countries to work toward global peace and prosperity, guided by 17 Sustainable Development Goals (SDGs) and 169 targets. The 17 goals focus on a range of issues including, but not limited to, poverty, hunger, education, gender equality, human health, energy, and industry. Although these goals provide a set of discrete targets, they were articulated knowing that sustainable development requires consideration of the complex ways in which people and their environment are intertwined. This complexity makes addressing issues of sustainable development difficult, time consuming, and resource intensive and demands new ways of thinking and problem-solving. Given the complexity of achieving sustainable development, we previously asserted that design thinking provides a useful, appropriate, and perhaps even necessary approach for addressing sustainability issues (Ardoin et al., 2022). Based on findings from an exploratory literature review (Ardoin et al., 2022), we identified five design thinking characteristics that support design thinking as a beneficial process for seeking solutions to climate change, biodiversity loss, water scarcity, and other wicked socioecological challenges. Through the process of conducting the exploratory literature review of design thinking and sustainability research, we confirmed that many professionals are already using design thinking approaches to tackle sustainability issues. As a result, we became curious about the experiences of academics and practitioners working in fields at the nexus of design and sustainability, wondering whether they would agree with the five design thinking principles we identified as supporting sustainability solutions. To explore this, we conducted 20 semistructured interviews with professionals who had used design thinking in their sustainability work. The 20 interviewees included designers, sustainability practitioners, and university professors from 9 countries. In this chapter, we present an analysis of the interview data designed to explore the level of support for the five design thinking characteristics we previously identified (Ardoin et al., 2022).

2 Background Many scholars have acknowledged the role of design in sustainability (Ardoin et al., 2022; Ceschin & Gaziulusoy, 2016; Dewberry & Sherwin, 2002; Fry, 2020; Manzini, 2007). As Papanek (2022, chap. 2) writes, “Design must be the bridge

Voices from the Field: Exploring Connections Between Design. . .

93

between human needs, culture and ecology.” Design can contribute to sustainable development through numerous pathways: Designers and researchers have created new approaches and tools such as eco-design, design for sustainability, and life cycle assessment (Bhamra & Lofthouse, 2016; Finnveden et al., 2009; Karlsson & Luttropp, 2006) to name a few. In a previous chapter for this book series, we argued that design thinking provides an ideal set of tools and a mindset for identifying and implementing sustainability solutions (Ardoin et al., 2022). Scholars in related fields and areas of interest—such as education and the learning sciences (Koh et al., 2015), public health (Roberts et al., 2016; Oliveira et al., 2021), and the social sector more broadly (Liedtka et al., 2017)—have encouraged the use of design thinking, noting its success in addressing complex, wicked problems. Building on previous discussions linking design thinking and sustainability (e.g., Clark et al., 2020; Young, 2010) and our experiences with design thinking approaches, we examined the use of design thinking to address sustainability issues (Ardoin et al., 2022). We explored five characteristics of design more intensively through the literature and in practice: inspiring creativity; participatory or peoplefocused; diversity in thought and action; a holistic, systems thinking mindset; and a streamlined, action-oriented approach. (See Table 1 for a description of how each of the characteristics helps address sustainability issues.)

3 Methods To explore the level of support for the five identified design thinking characteristics, we conducted qualitative interviews with professionals (academics and practitioners) who have used design thinking in their sustainability work. We collected and analyzed the interview data as described below.

3.1

Data Collection

As the interviews marked our first collection of qualitative data from people who had used design thinking in sustainability work, we sought to ensure that we addressed a core of similar topics in each interview. To do so, we developed a semi-structured interview guide (Corbin & Strauss, 2015; Rubin & Rubin, 2012) using a limited number of prepared questions (see Table 2). Three members of the research team conducted the interviews and adapted the interview guide and wording as appropriate, asking clarifying questions and probing when appropriate to the interviewee’s context. The interviewers strove to leave time for the interviewees to share additional information, as desired. For the interviews, we sought to identify professionals who had used design thinking approaches in their sustainability work. We used four methods to identify potential participants: (1) professional contacts in the design world with a

94

N. M. Ardoin et al.

Table 1 Design thinking characteristics and their connection to sustainability solutions Design thinking characteristic Inspires creativity

Participatory and people-focused

Encourages and inspires diversity in thought and action

Adopts a holistic, systems thinking mindset

Offers a streamlined, actionoriented approach

Connection to sustainability solutions Design thinking emphasizes process over product, meaning that the journey of working toward a solution can be just as important as whether one actually gets to a single “right” answer in the end. The creative confidence mindset and iterative nature of design thinking encourage participants to try taking action and working toward a solution or many possible solutions. Without fear of failure or judgment, individual and collective creativity can lead to accelerated innovation and new ways of thinking, which are critically needed in the sustainability space Rather than telling people what to do in a top-down manner, a design thinking approach supports curiosity, collaboration, and active, participatory learning. Design thinking lowers, and at times removes, the barrier to entry during the initial solution generation process as a broader group of participants are encouraged to contribute in unencumbered ways. Focusing on people’s needs and experiences builds ownership and enthusiasm, concurrently encouraging participants to construct their own knowledge about sustainability, grounded in personal and socioculturally relevant experience, including the causes, impacts, and potential solutions The design thinking process signals that we are all in this together, similar to the collective mindset recognizing that sustainability challenges require perspectives from all experiences, walks of life, and points of view. Design thinking engages, and indeed requires, diverse teams; this diversity enhances the richness of content, innovation, and action that arises from the group process. Diverse groups provide a greater number and variety of resources upon which the sustainability solution process can draw and helps ensure that the solution set produced is equitable, inclusive, and appropriate Combining design thinking and sustainability means emphasizing the multiple spatial and temporal scales of sustainability throughout the design process. By encouraging and employing a systems lens, design thinking helps participants envision the interconnectedness of complex sustainability challenges and, in the process, imagine more creative, collaborative, context-appropriate solutions Design thinking’s supportive environment can empower people or teams to take action, which can be especially important in situations where they might otherwise be daunted by a large-scale problem or one without an immediate solution. Design thinking provides structures and support to organize the problem-solving process in an efficient, streamlined way, encouraging communication and collaboration among a diverse range of stakeholders, engaged across complex sustainability challenges. The action orientation focuses everyone on a solution space in a collective, community-oriented way

Note: From Ardoin et al. (2022). Reprinted with permission

Voices from the Field: Exploring Connections Between Design. . .

95

Table 2 Semi-structured interview guide Participant’s background • Can you tell me about your background in design/design thinking and sustainability? Design thinking • How would you define design thinking? • What do you see as the benefits of design thinking? In your opinion, what makes for a successful design thinking process? • What are the limitations of the design thinking approach? What might be some possible dangers or challenges? Design thinking and sustainability/climate change • How might design thinking approaches be applied to issues of climate change and sustainability? • Can you describe for me a time when you used a design thinking approach to address a problem related to sustainability (probe: e.g., something related to climate change, biodiversity loss, water scarcity/drought)? Teaching about design thinking and sustainability/climate change • What competencies do you think people need to address sustainability issues? • Imagine you’re designing a course/workshop on design thinking for sustainability. What would you include in it?

sustainability focus and who were known to a member of the research team; (2) Internet searches using keywords such as “sustainability,” “designer,” “climate,” and “design thinking;” (3) a listing of designers on the Climate Designers website (https://www.climatedesigners.org/); and (4) chain-referral (“snowball”) sampling in which interview participants recommended others we should interview (Noy, 2008). Following protocol approved by Stanford’s Institutional Review Board (IRB), we contacted potential participants by email and we invited them for an interview. We shared information about the study and consent procedures. We scheduled interviews with all those who responded and agreed to be interviewed. Between July and October 2022, we conducted 20 interviews with 10 designers, 6 sustainability practitioners, and 4 university professors. The 20 participants came from 9 countries (9 from the USA, 3 from Australia, 2 from Italy, and 1 each from Finland, France, Kenya, India, Singapore, and South Africa). In terms of gender, ten participants identified as male and eight identified as female. (Two participants did not provide information about their gender identity.) Interviews averaged 41 min in length and were conducted over Zoom. After receiving consent from the interviewees to participate in the interview and allow us to record, we recorded the interviews in Zoom. We uploaded the audio files to otter.ai to generate an initial transcript. A team member then listened to the audio file, compared the audio with the transcript, and edited the transcript to improve accuracy. For analytic purposes, we uploaded the final transcripts into NVivo, a software program that facilitates analysis of qualitative data.

96

3.2

N. M. Ardoin et al.

Data Analysis

Initial qualitative coding of the data involved descriptive coding as two team members read each transcript and coded for general patterns and trends (Saldaña, 2021). In this chapter, we focus on the second round of coding in which we deductively coded the transcripts using the five design thinking principles that we identified in previous work (Ardoin et al., 2022): inspiring creativity; participatory or people-focused; diversity in thought and action; holistic, systems thinking mindset; and a streamlined, action-oriented approach. One team member acted as the primary coder and coded excerpts of text when those excerpts aligned with one of the five principles. A second research team member reviewed the coded data for consistency and alignment of the a priori codes. We discussed any coding discrepancies until consensus was reached (Bingham & Witkowsky, 2022). Once we completed the second round of coding, we examined the frequencies associated with each of the five main codes (the five design thinking characteristics). We iteratively reviewed the coded data to identify any patterns within each code. The primary coder used the qualitative technique of memoing (Keane, 2022) to record reflections and questions about the data. In the Findings section below, we share the level of support we found for each of the five design thinking characteristics and present selected excerpts from the transcripts that illustrate participant agreement with each characteristic. We use pseudonyms for the participants to ensure confidentiality, and we edited the excerpts for clarity while retaining the intended meaning.

4 Findings The 20 participants described how they used design thinking in an array of sustainability contexts, including climate action, fossil fuel divestment, ethics and climate change, food systems, and small-scale fisheries. We found support for all five of the design thinking principles identified in Ardoin et al. (2022) (see Table 3). Almost all interviewees (n = 17) provided responses indicating they valued design thinking as an approach to sustainability issues as design thinking is participatory and Table 3 Number of interviews coded at each design thinking principle Design thinking principle (Ardoin et al., 2022) Participatory and people-focused Inspires creativity Encourages diversity in thought and action Adopts a holistic, systems thinking mindset Offers a streamlined, action-oriented process

Number of interviews (n = 20) demonstrating support 17 10 10 9 9

Voices from the Field: Exploring Connections Between Design. . .

97

people-focused. Half of the interviews (n = 10) mentioned creativity and diversity as important design thinking principles when addressing sustainability. Finally, we coded nine interviews as showing support for the systems thinking and streamlined, action-oriented aspects of design thinking.

4.1

Design Thinking Is Participatory and People-Focused

Of the five characteristics, we coded the design thinking principle of being participatory and people-focused as having the most support in our interview data. Seventeen participants stated they valued design thinking in their sustainability work because the process centers on people, giving a voice to those most impacted by sustainability problems and helping ensure solutions are relevant, practical, and accepted. For example, two interviewees said: I think that, in my experience, knowing what I know, I think design thinking can only be community focused. So learning about collaborative co-creation, I have my process whenever I work with a freelance client or something like that. I always bring them on the journey and ensure that I’ve heard them, that they feel heard and understand what I do and that they see value in it before any real work happens—because without that human connection and understanding, there’s not enough trust, there’s not enough belief, there’s not enough understanding or insight into the process. So I think that for design thinking to work it needs 100% to have people that you’re solving problems for involved in the process, especially when it comes to climate. Because you might think that you understand the anxiety that young people have until you actually meet them, and they tell you face to face, how they’re feeling—and that’s a whole different experience. (Aidan, Sustainability practitioner) I also think that there’s a moral—that’s the wrong word—ethical . . . (pause) I’m not a fan of the idea of imposing a solution on anyone, and I think that, when you’re working in sustainability, and you’re working with social challenges, it can be very easy to fall into a situation where you’re imposing a solution on somebody. I think that the power of having people involved in that process, even if it’s not the final people that receive the benefit of that service—but people who are proxies for them, if that makes sense—I think that has a lot of moral strength. This is not just in terms of the “do good” sense of the word but means more that “we worked with you to achieve this.” (Gina, Sustainability practitioner)

As alluded to in both Aidan’s and Gina’s responses, the design thinking process can help build trust and relationships, an idea that interviewee Kara explicitly mentioned as well. It can also build closer relationships with communities in which you’re working, which is important. It can help build trust, and I think those are also important components to reduce the power dynamics between designers and people you’re designing for or with. It can be nice to find ways of involving folks. I think we find the most success when there’s more of that collaboration happening. . .How human-centered that the process is—I think that’s something that we really love about design thinking, is that it is so focused on putting people’s needs and goals first. (Kara, Sustainability practitioner)

98

N. M. Ardoin et al.

Several participants described the role empathy plays in design thinking and how this is a major strength of design thinking. One of the big parts of design thinking is building empathy and learning about the problem from the people who are facing the problem directly versus just assuming that you know what the problem is and applying your biases to that. . . .The biggest thing about being and working directly with the stakeholder is building empathy and trying to understand their point of view, as they call it. What is the problem from their perspective? And how is it that they’re uncovering this problem? . . .But until you’re there and really walking through that. . .you start to test those assumptions and you learn from people firsthand what the problem really is. So, it’s unlocking those really unique insights that I think is the most valuable part of design thinking, and a big part of that is working directly with the people, talking to them, doing these types of interviews, living their experiences, going out fishing with them at 3 am. (Alexander, Sustainability practitioner) Really going through the motions of that first kind of empathy building stage—it’s something that I think people speed through. And even within the world of applied design thinking, I think it often can be given short-change. But for us, that’s usually the most valuable, continuing piece of design thinking . . . forcing us to sit in that empathy-building work and really experiencing it and, you know, really questioning a lot of our assumptions about what people are doing and why. (Doug, Designer) What’s the core thing about design thinking? It’s the human-centered design aspect to it. And I think that’s probably also the most important thing for solutions—looking to address sustainability or climate change issues, which means really being aware of all of the stakeholders. And that step of empathizing is really key for that, too. I know a lot of these issues are impacting people differently and impacting specific communities in certain ways. And I think using that methodology when addressing climate change or sustainability issues [and learning] to be extra aware of the perspective and the experience of those different stakeholder groups is key. I think that could be one of the biggest advantages of the design thinking methodology . . . not just for being aware of [the specific] perspective, but also for understanding how solutions . . . will be feasible or not in those contexts and how they can actually be implemented effectively. (Elena, Designer)

Finally, design thinking’s people-centered focus also goes to the root of sustainability issues as it recognizes that, as Lorenzo says, sustainability problems are people problems. The involvement of people, in my opinion, is very important because we are talking about sustainability. We are talking about human behavior . . . because the climate crisis is 100% human-driven and so the involvement of citizens. . .is crucial. (Lorenzo, Designer)

4.2

Design Thinking Inspires Creativity

Ten interviewees discussed how design thinking provides space for creativity and innovation to flourish and how creative, innovative solutions are needed in the sustainability space.

Voices from the Field: Exploring Connections Between Design. . .

99

. . .the last element is certainly creativity. . .For me personally, the benefit is the freedom for approaching a problem that you’ve probably already addressed, but in a new way, and to be more creative about thinking it through. (Alexander, Sustainability practitioner) It encourages innovation, because it encourages you to explore a whole world of potential solutions that are available to you from the craziest, wildest ideas to the most traditional ones. (Elena, Designer) The very first [design principle], which just keeps coming up over and over in my mind, is innovation, innovation, innovation, innovation. It creates surprises that enable you to find novel solutions that you wouldn’t have otherwise thought of, and that creates a competitive advantage as a company. I’ve got countless little examples of that in my current practice. (Gina, Sustainability practitioner) And really, what we’re doing is . . . playing the facilitation role, but we’re also pushing an innovation mindset with people and giving them permission to come up with solutions that they may not have felt authorized to come up with previously. . . So, as a facilitator, you’re changing their mind and authorizing them to think big and, as a content subject matter expert, you are giving them new ideas, new thinking, new tools, making linkages that they might not otherwise make. (Dale, Designer) So, I think the benefits are that . . . it could lead to this way of thinking [that is] innovative or [that brings] unexpected solutions because designers look for inspiration in all different places. And so the way a designer might address a problem—even though the design thinking process is presented as linear—is, in fact, less linear than maybe other disciplines, you know. The inspiration might come from the strangest place. So I think it’s that creativity and the unexpected connections and the serendipity. I think all of this is what could possibly lead to more innovative solutions. (Lisa, Designer)

Some participants explicitly acknowledged the need for creativity because of the complexity of sustainability challenges, noting that the design thinking process helped people think about problems and solutions in new ways. I think there’s huge potential for this framework in this process to be used for all sorts of environmental challenges. I think, in general, creativity is going to be one of our best ways forward. And I think there’s always room for innovation. I think that’s something that design thinking also does so well: It gets people out of these standard solutions. . . . It’s kind of the “how might we” . . . Also, a great framing that I think came from design thinking [is] just thinking about problems differently. That’s really exciting to me because . . . the world’s going to only get more complex, and so I think we’re going to need these frameworks to be able to help us imagine something new. We’re going to need to continue to be thinking in new ways about how to solve these difficult challenges. (Kara, Sustainability practitioner)

4.3

Design Thinking Encourages and Inspires Diversity in Thought and Action

Ten participants discussed how design thinking excels at bringing together different types of people, leading to more innovative and effective solutions.

100

N. M. Ardoin et al.

The other thing I love about design thinking for this kind of challenge is it has the opportunity to be really collaborative and to help people work together across differences. So, I think we’re going to need a super diverse group of people working on environmental challenges. Design thinking, I think, is a great framework to include a lot of different voices in that conversation and to help build a solution that is the result of a lot of people’s ideas. (Kara, Sustainability practitioner)

A design-focused interviewee shared several scenarios in which the diversity that design thinking promotes leads to transformative ideas. There’re two things I’ll call out. One is about co-design, which we see as a totally non-discretionary part of design thinking. Co-design for us means that you’re bringing lots of different voices and perspectives to the design table, shifting from a consultation way of thinking into a collaborative way of designing a change. That happens in a couple of different ways. It might be happening in workshops with stakeholders and representatives who have lived experience of something all mixed together. Or it might be happening in user research, when we’re talking to people and bringing their perspectives in. Or it might be happening in testing and, when I say testing, I don’t necessarily mean testing user interface. It might be about testing a concept, or testing a story, as a way of expressing a regulatory or legal or policy change and making sure that those voices are heard back at the design table. So, I think, when you co-design well, you avoid negotiating down to the least-worst solution and, instead, you find yourself actually transforming things, like thinking about the fundamentals differently. So, that’s a good co-design process and that’s because there’s all these different perspectives that have suddenly been clashed together, and a lot of assumptions die in that conflict and a lot of new ideas get sprung out as well. (Dale, Designer)

When discussing diversity among those involved in the design thinking process, some participants valued the diversity among themselves (as the facilitators/lead designers) and the “user groups” with whom they work. For example, a university professor, Dario, described the value of working directly with farmers: “I think that this was perhaps the most interesting case to mention because these were definitely the type of stakeholders that were most different from us . . .this made the exchange between us very rich.” Other participants valued the diversity that existed among the people with whom they were working. Two university professors described the value that people from different disciplines and fields of study bring to the process: Something I’ve learned increasingly [through] teaching conservation biology is we can know all the biology in the world, but if we can’t work with social scientists, humanists, and others at communicating what we know, then what we’ve learned is not that useful. So integrating lessons from sociology, from economics, from political science, from art, and communication and social media [is important]. I think there’s an enormous need to tap into the interface of how people are influenced and what information is doing the influencing. . .I think academia is ideally positioned to participate in design thinking. I mean, most academicians aren’t familiar with design-thinking methods. So people who are, I think, have an enormous potential to tap into the knowledge of academicians on the latest findings, what works, what resonates with people, how to communicate with people. I know there’s a tremendous amount of research on all of these topics, and I think there’s an enormous potential for design thinking. (Theo, University professor) It may not be that somebody sketching on a storyboard in a workshop comes up with the brilliant idea, but then, when you step back and you look at the five storyboards that you’ve generated, and then you apply that experience and the expertise from different disciplines—

Voices from the Field: Exploring Connections Between Design. . .

101

you can come up with something really different and unique. (Gabriella, University professor)

4.4

Design Thinking Adopts a Holistic, Systems Thinking Mindset

Nine interviewees emphasized the importance of how design thinking approaches encourage people to take a holistic view of the problem and solution, acknowledging the value of systems thinking. I think [a benefit of design thinking] is the opportunity to take a holistic look at a systems level of an issue, and then engage a much wider group of brains in working out what to do about it. A lot of problems are solved at a simplistic or reductionist level. Design thinking done right is a prophylaxis against that. (Dale, Designer) [Design thinking is valuable in] taking a step back and looking at the whole picture and identifying where there are opportunities to continue with processes that you’ve done before and are static and normal and relatively straightforward. (Eddie, Sustainability practitioner)

One interviewee, a university professor, described how she values the systems thinking approach inherent in design thinking, noting that it encourages examination of both human and nonhuman components of a system. With the approach of design thinking, including the different perspectives and also the non-human perspectives, or their role, and just the existence of those non-human elements in the world around us—that’s definitely something that is really precious. (Gabriella, University professor)

Finally, some participants viewed a systems thinking approach as separate from what they see as design thinking but also described how they combine systems thinking with design thinking. The way we put it simply is that we use systems thinking and design thinking as two different and complementary approaches to understand the system. Systems thinking in general helps us to zoom out and understand the whole and design thinking helps us to zoom in . . . Understanding the micro to connect it with the macro, but also to envision change in the sense that some of the interactions that we have at small scale can influence the bigger picture. (Dario, University professor)

4.5

Design Thinking Offers a Streamlined, Action-Oriented Approach

Nine participants valued design thinking for the way the approach can spur action, often by offering a series of suggested steps, a framework, and guidelines. I think having concrete steps is really helpful, and we hear from organizations that many of them just aren’t really sure where to start. It’s kind of like “Design a behavior change

102

N. M. Ardoin et al.

campaign” and I mean, like, great, how do we do that? . . . Similarly with design thinking, just having this tested process of, “Okay, start here, then try this,” [is helpful]. Even though it’s nonlinear, I think the ability to repeat steps is really helpful—or to be able to start somewhere and go back. I think working within the journey is helpful as well. But having that very concrete guidance to create something is really helpful. We got some feedback that it helps people be more intentional in really thinking through their solutions. I think that’s great. (Kara, Sustainability practitioner) I think some of the benefits would be that it is a guide. If you feel as if a project is going off the tracks, you can, you know, do your research, and see if there’s something within that methodology that you can grab onto to bring it back on track. (Matthew, Designer)

Other participants valued design thinking because it offers an organized process that can be used even with and by those unfamiliar with design. It’s kind of creating a framework in which we can organize our ideas, our activities. It’s providing us a method. So instead of going random, we know that every time we have a project or any deal we want to develop, there is a first phase in which we need to really know what we’re talking about. So [first there is] this research phase. Then we can start to frame the specific problem we want to solve. . .So another thing in which design thinking for us is useful, it gives the opportunity to people that are not used to design thinking or design in general to use the methods, the tools, and to try to readapt them in different contexts, even if they’re not an expert. So, it’s very useful, in my opinion, [as a] framework and a guide that can be well explained and, if supported, can also be used by those who do not know design. (Lorenzo, Designer)

One interviewee, a sustainability practitioner, focused on climate change as a behavior change problem and described how the design thinking approach facilitates shifting behaviors especially in the context of undertaking rapid action. Well, I think design thinking and [related] strategies are basically behavior change strategies. A lot of our challenges in the climate space are behavior challenges. How do we change people’s behavior in terms of either their specific actions, individually or collectively? This is why I say that the psychology of habit formation and habit change really is important to think about. Once you understand that, then you can use design strategies to say, “Well, how do we break those habit cycles? And how do we help institute new strategies?” And [I use] my little exercise, which is very simple, and anybody who looks at it will say, “Oh, yes, that’s design 101, maybe.” But it’s been amazing to me how many people in the climate space are just like, “Oh, this is really good. This will get people moving.” I think, you know, we have a behavior problem. We have a psychological problem when it comes to climate change, so we need to use psychology and design strategies to get people moving. It won’t solve the problems, but it will get people to start taking action. (Sam, Sustainability practitioner)

A designer whom we interviewed also valued the efficiency of design thinking approaches: Just like having a blueprint, [design thinking] is very cyclical. It really prioritizes rapid prototyping, getting tangible stuff in people’s hands, and getting real customer feedback earlier than maybe you would otherwise be comfortable showing something to someone. . . .Especially in the world of physical goods, where it’s so easy to just spend years designing a product and never wanting to actually hand it over to customers until it’s absolutely perfect. A lot of products and companies can fall into that trap. But design thinking reminds us that it’s so much better to provide a very low-resolution prototype to people early and to catch

Voices from the Field: Exploring Connections Between Design. . .

103

basic mistakes as fast as possible so that you can learn and improve without wasting a ton of time and money on something that isn’t going to actually go anywhere. (Doug, Designer)

5 Discussion Since writing our previous chapter (Ardoin et al., 2022), several new studies have been published reporting examples of design thinking being used in a sustainability context. Törnroth et al. (2022), for example, developed and tested a design thinking workshop for citizen designers in Sweden to engage with urban renewable energy. Gozzoli et al. (2022) reported on the use of design thinking with a local community in Thailand to address sustainability in urban community development. In both instances, the authors emphasized design thinking’s ability to encourage collaboration and give voice to impacted communities. Massari et al. (2022) created a design thinking model called CEASE (communities, engagement, actions, shareability, ecosystems) to address sustainable food-related behaviors; the model highlights the value of empathy and creativity in the design thinking process. He and Ortiz (2021) further developed a design thinking framework for sustainable business models and explored its use with a Chinese company. Several studies discussed the use of design thinking in higher education settings. Avsec and Jagiełło-Kowalczyk (2021) reported on the use of design thinking with undergraduate architecture students as part of an effort to address the United Nations SDGs in architectural education. Their study found that design thinking increased self-directed learning among architecture undergraduates with improvements in a range of skills needed to create sustainable architecture design. Similarly, Manna et al. (2022) sought to bring an awareness of sustainability to undergraduate marketing education. These researchers developed a multicourse initiative that used design thinking to connect students with a local community organization to explore sustainability issues in a wetland area. Although the authors reported challenges with implementing the design thinking approach, they asserted that the benefits of the approach outweighed the challenges, commenting, “It is time for educators designing marketing curricula to prioritize the development of ‘big picture’ critical thinking skills that marketing needs, particularly in these times of global uncertainty” (Manna et al., 2022, p. 372). Taimur and Onuki (2022) described design thinking as a transformative learning approach needed to create competent leaders who can address complex, sustainability issues. They reported on courses from universities in Japan and Germany that used a design thinking approach in digital and hybrid learning. The findings from our interview-based study contribute to this growing evidence base by analyzing qualitative data to consider the level of support for five core design characteristics hypothesized to contribute to the search for sustainability solutions. Data from the interviews tapped into the direct experiences of practitioners, designers, and university professors working in the sustainability field. Across the 20 interviewees, we found support, at varying levels, for the design thinking

104

N. M. Ardoin et al.

characteristics of inspiring creativity; being participatory and people-focused; encouraging diversity in thought and action; taking a holistic, systems thinking perspective; and offering a streamlined, action-oriented approach. The semi-structured interview protocol did not explicitly share or prompt for these specific design thinking characteristics; rather, the protocol provided participants with the opportunity to offer thoughts on what may make design thinking approaches appropriate for use in the context of sustainability-related challenges. Deductive coding revealed the presence of the five design thinking characteristics in many of the interviews. The characteristic of participatory and people-focused received the strongest support, with 17 of the 20 interviews coded as demonstrating this connection with design thinking. Half or nearly half of the interviews (ten or nine interviews) were coded for each of the remaining four characteristics. This research thus demonstrates support in practice for the five design thinking characteristics initially identified in Ardoin et al. (2022).

6 Reflections and Conclusion Our findings suggest immediate implications for research and practice. Additional qualitative or mixed-methods research—including in-depth case studies, ethnographic work, and focus groups—could contribute to theory development regarding how design thinking augments the search for sustainability solutions. A logical next step building on this study’s findings would be to incorporate a larger number and more diverse perspectives and experiences of those who use design thinking in their sustainability work through employing, for example, a Delphi study or leveraging survey methods. Practical implications involve further use of design thinking in addressing sustainability challenges and, for those unfamiliar with design thinking, providing training in design thinking approaches, including specifically addressing their application in sustainability contexts. The interviews provided insight into which sustainability issues have been addressed by practitioners as well as recommendations for building capacity related to using design thinking in sustainability work. Further analysis of these data, combined with future studies, will provide additional insights addressing questions of scope, scale, and pedagogy for efforts combining design thinking and sustainability. Reflecting on both this study and our literature review (Ardoin et al., 2022), we see long-term implications for research and practice involving the intersection of design thinking and sustainability solutions. Our initial work has focused on existing literature and practitioner experiences, with our findings indicating an opportunity for researchers and practitioners to engage more deeply with the conceptualization of design thinking. A common theme across both studies was a popular view of design thinking as a step-by-step formula (Carter, 2016). We suggest moving the discussion beyond the basic principles of design thinking—such as its participatory, actionoriented nature—to delve more deeply into the underlying processes that promote creativity and problem-solving. This direction for research and practice might

Voices from the Field: Exploring Connections Between Design. . .

105

include engagement with the core abilities of design thinking, which cover aspects such as navigating ambiguity, synthesis, and abstraction, among others (Carter, 2016; Stanford d.school, n.d.). With optimism, we look forward to expanding upon this initial work to better understand the ways in which design thinking approaches can support researchers and practitioners alike in efforts to pursue a more sustainable world. Acknowledgments This research was supported by the Hasso Plattner Design Thinking Research Program (HPDTRP). We are grateful for the feedback received at the HPDTRP workshops and appreciate the support received from Jill Grinager and Sharon Therese Nemeth. Thanks to Carissa Carter and Sarah Stein Greenberg of the Stanford d.school for their guidance at various stages of this work. We appreciate research and editorial assistance from past and current members of the Stanford Social Ecology Lab, particularly Theo Bamberger, Estelle Gaillard, Pari Ghorbani, Veronica Lin, Kathleen O’Connor, Sarayu Pai, Indira Phukan, and Wilson Sherman. Finally, we are deeply grateful to the interview participants who shared their experiences and reflections.

References Ardoin, N. M., Bowers, A. W., Lin, V., & Phukan, I. (2022). Design thinking as a catalyst and support for sustainability solutions. In C. Meinel & L. Leifer (Eds.), Design thinking research: Achieving real innovation (pp. 325–340). Springer International Publishing. https://doi.org/10. 1007/978-3-031-09297-8_16 Avsec, S., & Jagiełło-Kowalczyk, M. (2021). Investigating possibilities of developing self-directed learning in architecture students using design thinking. Sustainability, 13(8), 8. https://doi.org/ 10.3390/su13084369 Bhamra, T., & Lofthouse, V. (2016). Design for sustainability: A practical approach. Routledge. https://doi.org/10.4324/9781315576664 Bingham, A. J., & Witkowsky, P. (2022). Deductive and inductive approaches to qualitative data analysis. In C. F. Vanover, P. A. Mihas, & J. Saldana (Eds.), Analyzing and interpreting qualitative research: After the interview (1st ed., pp. 133–148). Sage. Carter, C. (2016, October 11). Let’s stop talking about THE design process. Medium. https:// medium.com/stanford-d-school/lets-stop-talking-about-the-design-process-7446e52c13e8 Ceschin, F., & Gaziulusoy, I. (2016). Evolution of design for sustainability: From product design to design for system innovations and transitions. Design Studies, 47, 118–163. https://doi.org/10. 1016/j.destud.2016.09.002 Clark, R. M., Stabryla, L. M., & Gilbertson, L. M. (2020). Sustainability coursework: Student perspectives and reflections on design thinking. International Journal of Sustainability in Higher Education, 21(3), 593–611. https://doi.org/10.1108/IJSHE-09-2019-0275 Corbin, J., & Strauss, A. (2015). Basics of qualitative research: Techniques and procedures for developing grounded theory (4th ed.). Sage. Dewberry, E., & Sherwin, C. (2002). Visioning sustainability through design. Greener Management International, 37, 125–138. Finnveden, G., Hauschild, M. Z., Ekvall, T., Guinée, J., Heijungs, R., Hellweg, S., Koehler, A., Pennington, D., & Suh, S. (2009). Recent developments in life cycle assessment. Journal of Environmental Management, 91(1), 1–21. https://doi.org/10.1016/j.jenvman.2009.06.018 Fry, T. (2020). Defuturing: A new design philosophy. Bloomsbury Visual Arts. Gozzoli, P. C., Rongrat, T., & Gozzoli, R. B. (2022). Design thinking and urban community development: East Bangkok. Sustainability, 14(7), 7. https://doi.org/10.3390/su14074117

106

N. M. Ardoin et al.

He, J., & Ortiz, J. (2021). Sustainable business modeling: The need for innovative design thinking. Journal of Cleaner Production, 298, 126751. https://doi.org/10.1016/j.jclepro.2021.126751 Karlsson, R., & Luttropp, C. (2006). EcoDesign: What’s happening? An overview of the subject area of EcoDesign and of the papers in this special issue. Journal of Cleaner Production, 14(15), 1291–1298. https://doi.org/10.1016/j.jclepro.2005.11.010 Keane, E. (2022). Critical analytic memoing. In C. F. Vanover, P. A. Mihas, & J. Saldana (Eds.), Analyzing and interpreting qualitative research: After the interview (1st ed., pp. 259–274). Sage. Koh, J. H. L., Chai, C. S., Wong, B., & Hong, H.-Y. (2015). Design thinking for education. Springer. https://doi.org/10.1007/978-981-287-444-3 Liedtka, J., Salzman, R., & Azer, D. (2017). Design thinking for the greater good: Innovation in the social sector. Columbia University Press. Manna, V., Rombach, M., Dean, D., & Rennie, H. G. (2022). A design thinking approach to teaching sustainability. Journal of Marketing Education, 44(3), 362–374. https://doi.org/10. 1177/02734753211068865 Manzini, E. (2007). Design research for sustainable social innovation. In Design research now: Essays and selected projects (pp. 233–245). Birkhäuser. https://doi.org/10.1007/978-3-76438472-2_14 Massari, S., Principato, L., Antonelli, M., & Pratesi, C. A. (2022). Learning from and designing after pandemics. CEASE: A design thinking approach to maintaining food consumer behaviour and achieving zero waste. Socio-Economic Planning Sciences, 82, 101143. https://doi.org/10. 1016/j.seps.2021.101143 Noy, C. (2008). Sampling knowledge: The hermeneutics of snowball sampling in qualitative research. International Journal of Social Research Methodology, 11(4), 327–344. https://doi. org/10.1080/13645570701401305 Oliveira, M., Zancul, E., & Fleury, A. L. (2021). Design thinking as an approach for innovation in healthcare: Systematic review and research avenues. BMJ Innovations, 7(2), 491. https://doi. org/10.1136/bmjinnov-2020-000428 Papanek, V. (2022). The green imperative: Ecology and ethics in design and architecture. Thames & Hudson. Roberts, J. P., Fisher, T. R., Trowbridge, M. J., & Bent, C. (2016). A design thinking framework for healthcare management and innovation. Healthcare, 4(1), 11–14. https://doi.org/10.1016/j. hjdsi.2015.12.002 Rubin, H. J., & Rubin, I. S. (2012). Qualitative interviewing: The art of hearing data (3rd ed.). Sage. Saldaña, J. (2021). The coding manual for qualitative researchers (4th ed.). Sage. Stanford d.school. (n.d.). d.abilities. d.Student Handbook. https://www.dschoolstudent.com/ design-abilities Taimur, S., & Onuki, M. (2022). Design thinking as digital transformative pedagogy in higher sustainability education: Cases from Japan and Germany. International Journal of Educational Research, 114, 101994. https://doi.org/10.1016/j.ijer.2022.101994 Törnroth, S., Wikberg Nilsson, Å., & Luciani, A. (2022). Design thinking for the everyday aestheticisation of urban renewable energy. Design Studies, 79, 101096. https://doi.org/10. 1016/j.destud.2022.101096 United Nations. (2015). Transforming our world: The 2030 agenda for sustainable development. United Nations. https://sdgs.un.org/sites/default/files/publications/21252030%20Agenda%20 for%20Sustainable%20Development%20web.pdf Young, G. (2010). Design thinking and sustainability (p. 27). https://zum.io/wp-content/ uploads/2010/06/Design-thinking-and-sustainability.pdf

Part 2

Prototyping

User Perceptions of Privacy Interfaces in the Workplace Michelle S. Lam, Matthew Jörke, Jennifer King, Nava Haghighi, and James A. Landay

Abstract Intelligent workplace systems that support well-being offer the potential to reduce stress and burnout, promote physical activity, or increase creativity and collaboration while at work. However, such systems rely on the collection of sensitive personal information that can pose significant privacy risks to users. In this chapter, we investigate how user perceptions of privacy vary with privacy interface designs and framing scenarios. In a user study with 60 participants, we present participants with four privacy interfaces based on different privacy frameworks and study how perceptions of comfort and control vary depending on the owner of the sensing technology and the user’s relationship with that owner. We find that participants express greater comfort and control with interfaces that foreground contextual information and provide relationship-based access control. Moreover, participants display lower feelings of comfort and control when the technology is deployed company-wide or by a manager with whom they have a negative relationship. Concerningly, we find that interfaces based on technical privacy metrics are poorly understood and have the potential to promote a false sense of security. Taken together, our findings have implications for the design of privacy interfaces and can inform future large-scale studies on privacy attitudes in the workplace.

M. S. Lam · M. Jörke (✉) · N. Haghighi · J. A. Landay Department Computer Science, Stanford School of Engineering, Stanford University, Stanford, CA, USA e-mail: [email protected]; [email protected]; [email protected]; [email protected] J. King Institute for Human-Centered Artificial Intelligence, Stanford University, CA, Stanford, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_6

109

110

M. S. Lam et al.

1 Introduction Intelligent workplace systems that support well-being offer the potential to reduce stress (Adler et al., 2022; Howe et al., 2022), increase creativity and collaboration (Jun et al., 2019; Mandryk & Inkpen, 2004), encourage physical activity (Conn et al., 2009), and optimize productivity (Costa et al., 2019; Kimani et al., 2019; Mark et al., 2017) while at work. Such systems necessarily rely on data collection in the workplace, such as through passive monitoring using wearables or logging software. However, amid growing academic and public concern around workplace surveillance (Kantor & Sundaram, 2022) and “bossware” (Corbyn, 2022), user privacy is a fundamental concern. Data collected for the purpose of improving well-being can expose sensitive health information (Harwell, 2019), can be shared by employers with unwanted third parties (e.g., insurance companies) (Sloat, 2022), or can be misused to monitor and surveillance workers (Kantor & Sundaram, 2022). For instance, a survey of mental health apps by the Mozilla Foundation in 2022 reports that many employer-provided mental health apps share sensitive information with employers, often unbeknownst to the employee (Sloat, 2022). Thus, data privacy must be an integral concern in the design of workplace well-being support systems (Diel et al., 2022; Partnership on AI, 2022). Benefits to employee well-being will not be realized if they come at the expense of employee privacy. In this work, we present our findings from an online user study (N = 60) to assess how users’ perceptions of privacy vary depending on the owner of the sensing technology and depending on whether the user’s relationship with the owner is positive or negative. In our study, we present each participant with four privacy interfaces inspired by several privacy frameworks in the literature and manipulate a framing scenario that describes the context of deployment. We measure participants’ perception of comfort and control across the various conditions and framing scenarios, as well as their comprehension of and confidence in privacy-preserving features. We find that participants expressed increased comfort and control with interfaces that provide contextual information about sensed data and allow relationship-based access control. However, these preferences are highly dependent on the deployment scenario. Participants expressed decreased comfort and control in conditions where the system was deployed company-wide and in conditions where they had a negative relationship with their manager. Finally, interfaces that heavily rely on technical privacy metrics were not only poorly understood in general but led to a false sense of security and adoption of less private settings among users with low comprehension. Our work has implications for the design of workplace well-being support systems, namely, that privacy interfaces must be evaluated with consideration of the deployment context and workplace power dynamics. We additionally find that technical privacy metrics should be coupled with user-centered design approaches to ensure that users make informed decisions about their privacy.

User Perceptions of Privacy Interfaces in the Workplace

111

2 Background The interface designs and framing scenarios employed in our study draw upon four privacy frameworks from the literature, which we briefly summarize here. Nissenbaum’s theory of privacy as contextual integrity (Nissenbaum, 2009) frames privacy violations as inappropriate flows of personal information that violate contextual information norms. Five contextual parameters govern information norms: the data subject, data sender, data recipient, information type, and form of transmission. This theory of privacy informed the design of our scenario-based interface, which highlights these key contextual parameters, as well as the technology-owner dimension of our framing scenarios. Next, we designed a relationship-based interface that primarily broke down privacy controls in terms of user relationships in line with work on relationshipbased access control (Fong, 2011; Petronio, 2002). Relationship-based access control is a mechanism for granting access to personal data based on relationships among users, which may also be context dependent. Drawing on work in multiparty approaches to privacy that highlights the collective, interdependent nature of online privacy (Humbert et al., 2019), we then designed a socially conscious interface that surfaced the privacy behavior of trusted peers and enabled users to share endorsements or warnings about privacy settings. Notably, users may be susceptible to peer influence in social settings (Mendel & Toch, 2017; Tyagi et al., 2016), adopting both beneficial and detrimental privacy practices based on the behavior of their peers and their relationships with those peers. Finally, we designed a privacy metrics-based interface that incorporates user controls based on common metrics from the technical privacy literature. In particular, we explore two technical privacy approaches. First, we explore differential privacy (Dwork, 2008), which injects carefully chosen noise into each individual’s data such that their information is private in isolation but inferences can be made in aggregation over larger populations. The second approach we explore is k-anonymization (Sweeney, 2002), whereby data is masked or aggregated such that each individual cannot be distinguished from at least (k - 1) other individuals in the dataset. Prior work has shown that lay users frequently have inaccurate mental models of technical privacy solutions (Dechand et al., 2019; Kang et al., 2015; Wu & Zappala, 2018) and that technically savvy users often adopt risky privacy practices despite their technical knowledge (Barth et al., 2019; Kang et al., 2015).

3 Study Design Our work aims to understand how privacy design frameworks play out in realistic workplace deployments. To what extent are user reactions to privacy-focused design patterns impacted by details of the deployment scenario?

112

3.1

M. S. Lam et al.

Research Questions

Across the four different privacy design approaches we identified, we sought to address the following core research questions: RQ1: To what extent does the owner of the sensing technology impact user perceptions of comfort and control? RQ2: To what extent does the valence of the user’s relationship with the sensing technology owner impact user perceptions of comfort and control? In the context of this study, the user of the technology is an employee, the owner of the technology is the organization or individual that controls the system and its data, and valence refers to whether the relationship between the user and owner is positive or negative. Our decision to study privacy perceptions as a function of the technology owner and the valence of the user’s relationship with that owner reflects the power dynamics present in workplace settings and is informed by the privacy frameworks discussed in the previous section. To tackle these questions, we instantiate each of the four privacy design frameworks as a concrete privacy interface in the workplace context, as described in Sect. 3.2. Then, we present these privacy interfaces to everyday users in a survey experiment to understand user preferences in various contexts. Each user is assigned a random framing scenario, which modifies factors such as the owner of the technology and the valence of the user’s relationship, as described in Sect. 3.4.

3.2

Privacy Interfaces

We prototyped four user interfaces that each captured different privacy design frameworks: (1) a scenario-based interface, (2) a relationship-based interface, (3) a socially conscious interface, and (4) a privacy metrics interface. Across these interfaces, participants viewed hypothetical data sources organized into two categories: physical activity (daily step count, heart rate, flights climbed, and daily sleep hours) and work activity (stress level and browser history). The socially conscious interface only displayed the first three data sources (daily step count, heart rate, and flights climbed) due to space constraints.

3.2.1

Scenario-Based Interface

The first interface, drawing upon the contextual integrity theory of privacy (Nissenbaum, 2009), places emphasis on communicating the key contextual parameters of the sensing technology to the user. The interface is broken into categories by the kind of data collected (e.g., “physical activity” vs. “work activity”), and listed within those categories are specific data sources (e.g., “daily step count” and “heart

User Perceptions of Privacy Interfaces in the Workplace

113

Fig. 1 Scenario-based interface

rate”) (Fig. 1). Each of these data sources can be expanded to view additional details; in this view, details include information about the what (the requested data), who (the audience of the data), why (the purpose for the data collection), and how (the data collection strategy). The user can then enable or disable each of these data sources based on the provided information.

3.2.2

Relationship-Based Interface

The next interface draws upon the proposals for relationship-based access control based on theories of relational privacy (Petronio, 2002) to primarily orient privacy settings around relationships with other users (here, coworkers). This page allows users to create custom groups of coworkers and specify data permission settings for these different groups (Fig. 2). Once a user group is created, the user can specify a custom set of data access settings that will only apply to that group of users. These settings include the data aggregation level and the same individual data sources and category groupings present in the scenario-based interface. Again, each data source can be individually enabled or disabled from this page. For example, the user could create a custom group of “trusted teammates” and opt to grant them higher level of data access than “superiors.”

114

M. S. Lam et al.

Fig. 2 Relationship-based interface

3.2.3

Socially Conscious Interface

The third interface captures ideas related to multiparty approaches to privacy (Humbert et al., 2019) by surfacing the privacy choices made by coworkers and allowing users to factor this information into their own privacy decisions (Fig. 3). This page allows users to create a custom set of “selected peers” whom they trust, and the members of the peer group agree to share their privacy habits to help each other stay safe. Then, for each data source, users can either select that the data collection be “allowed” or “blocked,” and they can view additional information about the data source. Below each option, they can see icons showing the profile photos of the peers in their group who have selected that option (either to allow or to block that particular data source). Finally, users can choose to select the “endorse choice” or “warn peers” buttons to take a more active stance on their privacy settings. These buttons will trigger push notifications to their peers with a custom message explaining why the privacy setting should be allowed (for endorsements) or blocked (for warnings).

User Perceptions of Privacy Interfaces in the Workplace

Fig. 3 Socially conscious interface

Fig. 4 Privacy metrics interface

115

116

3.2.4

M. S. Lam et al.

Privacy Metrics Interface

The last interface is based on technical treatments of privacy: differential privacy (Dwork, 2008) and k-anonymity (Sweeney, 2002). Similar to the scenario-based interface, this page features different data sources with collapsible sections (Fig. 4). When expanded, each section displays detailed information on the data collected in terms of the anonymization level and privacy-preserving noise level. Each of these metrics is accompanied with a visual representation of the privacy metric and text description providing an interpretation of that metric in plain language. Users can select to enable or disable collection of that data based on this information.

3.3

Framing Scenarios

We then set up framing scenarios to address our research questions around the impact of technology owners and the valence of a user’s relationship with these owners. We create four tiers of technology owners in a workplace context: (1) company, (2) team, (3) manager, and (4) teammate. Then, for the last two owners which involve a personal relationship with an individual (manager and teammate), we explore relationships with either a positive or negative valence. Based on these, we develop six scenario variants to prime our study participants before they evaluate our privacy interfaces. 1. Company-oriented scenario where a large tech company deployed the sensing technology and the privacy interface. 2. Team-oriented scenario where the user’s particular team within the company owned the technology. 3. Manager-oriented (positive) scenario where the user’s team manager owned the technology and the user had a positive relationship with the manager. 4. Manager-oriented (negative) scenario where the user’s team manager owned the technology and the user had a negative relationship with the manager. 5. Teammate-oriented (positive) scenario where another member on the team owned the technology and the user had a positive relationship with the teammate. 6. Teammate-oriented (negative) scenario where another member on the team owned the technology and the user had a negative relationship with the teammate. Each of the scenarios is conveyed to study participants through a brief paragraph at the start of the survey. This paragraph describes the participant’s role, the new sensing technology that is going to be introduced, the owner of that technology, the purpose of the technology, and, if applicable, the valence of the participant’s relationship with the technology owner. Below is an example of the paragraph for the manager-oriented (positive) scenario. The full text for all six scenarios can be found in Appendix Section 1.

User Perceptions of Privacy Interfaces in the Workplace

117

You are an engineer on a small eight-person team, and you have a very positive, strong relationship with your manager, Jamie. You’ve worked with Jamie for 2 years already, and you trust that they make great decisions for the team. Jamie has recently started a pilot program to improve team members’ general wellbeing by collecting data using work-issued devices like laptops and phones. To allow team members to manage the privacy of this data, Jamie has created an interface where employees can control what data can be analyzed.

3.4

Survey Design

Our survey experiment was conducted using Qualtrics, a survey management platform that allowed us to deploy random assignment to our framing scenarios and random ordering of our privacy interface sections. The survey was broken into the following sections: (1) a brief introduction to the survey; (2) a single randomly selected framing scenario from the six described in Sect. 3.4; (3) a page with questions about each of the four privacy interfaces, shown in randomized order; and (4) a final overall rating of all interfaces and a concluding page to provide demographic information. On the framing scenario page (Fig. 5a), we explained that we would be showing four different designs for privacy settings pages in the following sections of the survey. We asked for participants to answer the questions for each design based on the context we provided with the framing scenario. Since it was important that participants indeed absorbed the content of the framing scenario before proceeding, we added attention check questions below the scenario paragraph. We asked about which entity owned the technology and, if applicable, about the valence of the participant’s relationship with that entity. Participants needed to answer these questions correctly before they would be allowed to proceed to the next page. On each privacy interface page (Fig. 5b), there was an image of the interface and a detailed description of its functionality, as in Sect. 3.2. Each of the interfaces was paired with a set of questions designed to gauge the participant’s sense of comfort and control. We also included a “feature importance” question where we listed several key features of the particular interface and, for each, asked participants to indicate the importance of that feature in informing their privacy choices. Finally, we included additional interface-specific questions for the socially conscious interface and privacy metrics interface. For the socially conscious interface, we asked participants to indicate their likelihood of endorsing one of their privacy settings or warning peers about choosing a setting that differed from their own. For the privacy metrics interface, we also asked questions to examine the participant’s objective comprehension of the metrics as well as their confidence in their level of comprehension. These answers would help us to assess the participant’s ability to make informed decisions about privacy settings. Each of the privacy interface pages also included an optional free-text question to provide any additional comments about that interface.

118

M. S. Lam et al.

Fig. 5 Qualtrics survey. (a) The framing scenario survey page, here displaying the team-oriented scenario. (b) A privacy interface survey page, here displaying the scenario-based interface

User Perceptions of Privacy Interfaces in the Workplace

3.5

119

Participant Recruitment

We recruited study participants using Prolific, a platform for sourcing research participants. Our study included 60 participants, providing about 10 participants for each framing scenario condition. Since the survey required approximately 10 min, we compensated participants $2.00 for taking part in our study. The study was advertised on the platform as a “10-minute Workplace Privacy Settings Survey.” We filtered to participants located in the USA who had a minimum approval rate of 95% on the platform.

4 Results Our study (N = 60) found that participants were influenced by the framing scenario manipulation.

4.1

Participant Demographics

Based on the demographic questions from our survey, participants in our study consisted of 27 women and 33 men; for race, 36 participants identified as White, 11 as Black or African-American, 5 as Asian, 3 as multiracial, 2 as American Indian or Alaska Native, 1 as Hispanic, and 2 preferred not to state their racial identity. For age, we had 14 participants from 18 to 24, 30 from 25 to 34, 11 from 35 to 44, and 5 from 45 to 54. Lastly, for educational background, we had 4 high school graduates, 13 with some college, 10 with an associate’s degree, 20 with a bachelor’s degree, 10 with a master’s degree, and 3 with a professional degree. Due to random assignment, there were not exactly ten participants in each of the framing scenario conditions. There were 10 in the company-oriented scenario, 11 in the team-oriented scenario, 10 in the manager-oriented (positive) scenario, 10 in the manager-oriented (negative) scenario, 8 in the teammate-oriented (positive) scenario, and 11 in the teammate-oriented (negative) scenario.

4.2 4.2.1

User Perceptions of Comfort and Control Across Interfaces User Comfort

Across the interfaces, there was generally greater comfort in team-oriented rather than company-oriented framings, with 66% vs. 38% positive comfort ratings

120

M. S. Lam et al.

a Response

Somewhat uncomfortable

Very uncomfortable

Neither comfortable nor uncomfortable

Somewhat comfortable

Very comfortable

b 3: Manager-oriented (positive) 2: Team-oriented 6: Teammate-oriented (negative) 4: Manager-oriented (negative) 1: Company-oriented 5: Teammate-oriented (positive)

70% 66% 61% 42% 38% 34%

18% 7% 16% 20% 20% 28%

12% 27% 23% 38% 42% 38% 100

50

0

50

100

Percentage

c UI2: Relationship-based

28%

15%

57%

UI1: Scenario-based

28%

17%

55%

UI4: Privacy Metrics

28%

18%

53%

UI3: Socially-conscious

33%

20%

100

50

0

47% 50

100

Percentage

Fig. 6 User comfort results. (a) Legend for comfort ratings. (b) Comfort results, by framing scenario. (c) Comfort results, by privacy interface

(Fig. 6b). There was greater comfort for positive rather than negative manageroriented scenarios, with 70% vs. 42% positive comfort ratings. Somewhat surprisingly, there was greater comfort for negative rather than positive teammate-oriented scenarios (61% vs. 34% comfort). Among the four interfaces, participants appeared least comfortable with the socially conscious interface (47% comfort), but comfort levels appeared comparable among the other interfaces, ranging between 53% and 57% comfort (Fig. 6c). Users expressed less comfort with this interface especially for the team-oriented (45% comfort compared to a mean of 73% comfort on other interfaces) and negative manager-oriented scenarios (30% comfort compared to a mean of 46.7% comfort on other interfaces).

4.2.2

User Control

Similar to comfort, participants felt a greater sense of control in team-oriented rather than company-oriented framings, with 77% vs. 52% positive control ratings (Fig. 7a). There was a greater sense of control for positive rather than negative manager-oriented scenarios, with 80% vs. 60% positive control ratings. Surprisingly again, there was a greater sense of control for negative rather than positive teammate-oriented scenarios (70% vs. 44% control). Among the four interfaces, participants expressed greater control with the relationship-based interface (73% control); the level of control appeared comparable among the other interfaces, ranging from 58% to 65% control (Fig. 7b).

User Perceptions of Privacy Interfaces in the Workplace

121

a Response

Very insufficent control

Somewhat insufficient control

Neither sufficient nor insufficent control

Somewhat sufficent control

Very sufficient control

b 3: Manager-oriented (positive) 2: Team-oriented 6: Teammate-oriented (negative) 4: Manager-oriented (negative) 1: Company-oriented 5: Teammate-oriented (positive)

80% 77% 70% 60% 52% 44%

2% 11% 16% 12% 30% 38%

18% 11% 14% 28% 18% 19% 50

100

0

50

100

Percentage

c 12%

15%

73%

UI1: Scenario-based

22%

13%

65%

UI4: Privacy Metrics

17%

20%

63%

UI3: Socially-conscious

20%

22%

UI2: Relationship-based

100

50

0

58% 50

100

Percentage

Fig. 7 User control results. (a) Legend for control ratings. (b) Control results, by framing scenario. (c) Control results, by privacy interface

4.3

User Reactions to Privacy Metrics

Next, we investigated our participants’ responses to technical privacy metrics in the privacy metrics interface. For the two privacy metrics—privacy-preserving noise level and anonymization level—we explored participants’ objective comprehension, confidence about their comprehension, and preferred privacy metric setting.

4.3.1

Privacy-Preserving Noise

Objective comprehension. To assess participants’ understanding of privacypreserving noise, we asked them to select all answer choices that are true about a high privacy-preserving noise level based on our description. There were six answer choices, and three of them were indeed true. We calculated a correctness score for each participant where their score improved by +1 for each answer that they selected that was correct and decreased by -1 for each answer that they selected that was incorrect. This means that a score of +3 indicates that a user only selected correct answers, and a score of -3 indicates that a user only selected incorrect answers. We found that there was substantial spread in participant comprehension; only 17 participants displayed full comprehension as measured by our survey question (Fig. 8). Comprehension vs. confidence. We then compared these participant comprehension scores against participants’ self-rated confidence in their comprehension of the privacy-preserving noise metric. We found that generally, higher comprehension was associated with higher confidence (Fig. 9a). Among participants with full comprehension, 77% indicated that they were either somewhat confident or very confident in their comprehension, compared to 63.7% of participants with

M. S. Lam et al.

122

Correctness score

15

-3 -2 -1 0 1 2 3

count

10

5

0 -3

-2

-1

0

1

2

3

Privacy-preserving noise answer correctness

Fig. 8 Participant comprehension of privacy-preserving noise

a Privacy-preserving noise conf. 3

18%

12%

71%

2

40%

0%

60%

1

20%

20%

60%

0

33%

33%

33%

-3

44%

11%

44%

-2

67%

0%

33%

-1

31%

31%

38%

50

100 Response

Very unsure

Somewhat unsure

0 Percentage Neither confident nor unsure

50 Somewhat confident

Higher comprehension

Lower comprehension

100 Very confident

b Ul4: Metics 3

35%

24%

41%

2

20%

40%

40%

1

20%

0%

80%

0

67%

0%

33%

-3

11%

11%

78%

-2

100%

0%

0%

-1

15%

31%

54%

100 Response

50 Very uncomfortable

Somewhat uncomfortable

0 Percentage Neither comfortable nor uncomfortable

50 Somewhat comfortable

Higher comprehension

Lower comprehension

100 Very comfortable

c Ul4: Metics 3

35%

24%

41%

2

20%

0%

80%

1

10%

0%

90%

0

0%

0%

100%

-3

11%

11%

78%

-2

33%

67%

0%

-1

0%

38%

62%

100 Response

Very insufficent control

50 Somewhat insufficient control

0 Percentage Neither sufficient nor insufficient control

50 Somewhat sufficient control

Higher comprehension

Lower comprehension

100 Very sufficient

Fig. 9 Privacy-preserving noise: comprehension results. (a) Privacy-preserving noise: comprehension vs. confidence. (b) Privacy-preserving noise: comprehension vs. comfort. (c) Privacy-preserving noise: comprehension vs. control

User Perceptions of Privacy Interfaces in the Workplace

123

correctness scores above zero and 37.0% of participants with correctness scores of zero or below. However, these same figures indicate that there was still quite high confidence among those with low comprehension (overconfidence), and there was substantial lack of confidence among those with high comprehension (underconfidence). Then, comparing participant comprehension against comfort and control ratings, we found that the highest levels of comfort and control appear to be tied to participants who had a moderate-to-low comprehension level (Fig. 9b, c). Preferred setting. Lastly, we looked at participants’ stated preferences in setting the privacy-preserving noise level. For this metric, participants had the option to set their noise level at “Low: a low amount of added noise, so a low level of privacy, but high quality evaluations of your well-being”; “Medium: a medium amount of added noise, so a moderate level of privacy and moderate quality evaluations of your wellbeing”; or “High: a high amount of added noise, so a high level of privacy, but low quality evaluations of your well-being.” The majority of participants placed the noise setting at a “Medium” level (Fig. 10a). Breaking down these results by comprehension level, we found that users with lower comprehension more often chose a “Low” or “Medium” noise level, which corresponds to less privacy (Fig. 10b). Meanwhile, participants with higher comprehension more often chose the “Medium” or “High” noise level, but still more participants tended to choose the “Medium” rather than “High” noise level.

a

40

count

30 noise level LOW MEDIUM HIGH

20 10 0 LOW

MEDIUM

HIGH

Privacy-preserving noise level

b

Privacy-preserving noise level

10

count

noise level LOW MEDIUM HIGH

5

0 -3

-2

-1

0

1

2

3

privCorrect

Fig. 10 Privacy-preserving noise: preferred settings. (a) Privacy-preserving noise: overall preferred setting. (b) Privacy-preserving noise: preferred setting by comprehension

124

M. S. Lam et al.

4.3.2

Anonymization Level

Objective comprehension. To assess participant comprehension of anonymization level, we asked them to select the most precise and correct interpretation of an anonymization level of “5.” There was only one correct answer choice for this question, so this served as a binary indicator of comprehension. We found that more participants appeared to understand anonymization level (selecting the correct option 4), but many participants selected a nearly correct answer (option 3), which was not the most precise option (Fig. 11). Comprehension vs. confidence. We then compared participants’ comprehension of anonymization level against their confidence in their comprehension. We again found that comprehension was tied to slightly higher confidence (Fig. 12a); participants who correctly answered the comprehension question reported slightly higher confidence (60% somewhat or very confident) than those who did not (50% somewhat or very confident). Nonetheless, we found that there was overconfidence among those with incorrect comprehension, and there was still substantial underconfidence among those with correct comprehension. Comparing comprehension to perceptions of comfort and control (Fig. 12b, c), we found that users with a correct understanding of anonymization level appeared to have lower levels of comfort (47% somewhat or very comfortable) and control (53% somewhat or very sufficient control) than participants who had an incorrect understanding (60% somewhat or very comfortable; 73% somewhat or very sufficient control). Preferred setting. We then compared participants’ selected anonymization level settings against their comprehension and framing scenario. For this metric, participants had the option to set anonymization level X at any value from 0 to 20, with the prompt “In the database, you would be indistinguishable from X other individuals.” Overall, participants set their anonymization level at 8.47 (SD = 5.39). Across participant comprehension levels, we found similar anonymization level settings for participants who correctly (M = 8.5, SD = 5.88) or incorrectly (M = 8.43, SD = 4.95) understood the metric (Fig. 13a). Across the framing scenarios, we observed trends similar to those we found in the comfort and control measures. There appeared to be a higher requested anonymization level for the companyoriented rather than team-oriented scenario and for the positive rather than negative

30 answer choice

20 count

There are 5 data fields that 5 individuals share in the database There are 5 individual database entries from you There are 5 individuals who have the same values in the database There are 5 other individuals who have the same values as you in the database

10

0 1

2

3

4

Anonymization level quiz answer

Fig. 11 Participant comprehension of anonymization level

User Perceptions of Privacy Interfaces in the Workplace

125

a Anonymization level conf. 23%

17%

60%

Correct

23%

27%

50%

Incorrect

100

50

Response

Very unsure

50

0 Percentage Somewhat unsure

Neither confident nor unsure

100

Somewhat confident

Very confident

b UI 4: Metrics 37%

17%

47%

Correct

20%

20%

60%

Incorrect

100

50

Response

Very uncomfortable

0 Percentage

Somewhat uncomfortable

c

50

Neither comfortable nor uncomfortable

100

Somewhat comfortable

Very comfortable

UI 4: Metrics 27%

20%

7% 100

Response

Very insufficient control

0% 50

Somewhat insufficient control

0 Percentage

50

Neither sufficient nor insufficient control

53%

Correct

73%

Incorrect

100

Somewhat sufficient control

Very sufficient

Fig. 12 Anonymization level: comprehension results. (a) Anonymization level: comprehension vs. confidence. (b) Anonymization level: comprehension vs. comfort. (c) Anonymization level: comprehension vs. control

teammate-oriented scenario (Fig. 13b). The positive and negative manager-oriented scenarios appeared to have more comparable anonymization levels.

4.4

User Perceptions of Privacy Interface Features

Based on our feature importance questions, we noted several trends among participants’ perceptions of privacy interface features (Fig. 14). Generally, the name of the data source (e.g., “daily step count”) was viewed as having lower importance than other details. For the scenario-based interface, the what, who, how, and why had similar levels of importance. For the relationship-based interface, the user set was slightly more important than the data aggregation level. For the socially conscious interface, the particular peers in the trusted user set were more important than the number of peers, and warnings were viewed as having a similar level of importance as endorsements. For the privacy metrics interface, anonymization level (k-anonymity) was rated as slightly more important than privacy-preserving noise level (differential privacy).

126

M. S. Lam et al.

a

20

Anonym level

15

10

5

0 Incorrect

Correct Comprehension

b 20

Anonym level

15

10

5

0 Company-oriented Team-oriented

Manager-oriented Manager-oriented (positive) (negative)

Team-oriented Teammate-oriented (negative) (positive)

Scenario

Fig. 13 Anonymization level: preferred settings. (a) Anonymization level: preferred setting by comprehension. (b) Anonymization level: preferred setting by framing scenario

Diving a bit more into the socially conscious interface, we asked participants about the likelihood of endorsing or warning using this page (Fig. 15). We observed that participants seemed slightly more inclined to endorse a privacy option than to warn their peers about a different privacy option (Fig. 15b). Across the different framing scenarios, we find that participants seemed more likely to endorse in

User Perceptions of Privacy Interfaces in the Workplace

127

a Response

Very unimportant

Somewhat unimportant

Neither important nor unimportant

Somewhat important

Very important

b What (requested data) How (data collection strategy) Who (audience)

8% 7% 12% 10%

Metric name

13%

Why (purpose)

7% 8% 5% 8%

85% 85% 83% 82%

8%

100

50

78%

0 Percentage

50

100

c User set

3%

10%

87%

Aggregation level

8%

8%

83%

Metric name

13%

12%

75%

0 Percentage

50

100

100

50

d Metric name

20%

8%

72%

PARTICULAR peers

15%

17%

68%

Warning push notif

15%

18%

67%

Endorsement push notif

13%

23%

NUMBER of peers

20%

17%

100

e

Anonym level DESCRIPTION Anonym level RATING

Priv-preserving noise DESCRIPTION Metric name Priv-preseving noise RATING

12% 7% 10% 18% 10% 100

63% 63%

0 Percentage

50

50

100

8% 13% 15% 7% 18% 50

0 Percentage

80% 80% 75% 75% 72% 50

100

Fig. 14 Interface feature importance results. (a) Legend for feature importance ratings. (b) Scenario-based interface, feature importance results. (c) Relationship-based interface, feature importance results. (d) Socially conscious interface, feature importance results. (e) Privacy metrics interface, feature importance results

positive-oriented relationship scenarios. Meanwhile, participants did not seem more likely to warn in negative-oriented relationship scenarios.

5 Discussion We found that participants reported the highest comfort and control for the relationship-based interface, closely followed by the scenario-based and privacy metrics interfaces. This indicates that participants appreciated relationship-based access control, detailed information about key contextual information, and additional information about privacy-preserving metrics. We suspect that lower levels of comfort and control in the socially conscious interface are due to privacy preferences being shared with peers. However, these interface ratings also varied widely depending on the framing scenario.

128

M. S. Lam et al.

a Response

Very unlikely

Somewhat unlikely

Neither likely nor unlikely

Somewhat likely

Very likely

b Endorse likelihood

30%

Warn likelihood

25%

25%

45%

37%

100

50

38%

0 Percentage

50

100

c Endorse likelihood 6: Teammate-oriented (negative)

36%

18%

45%

5: Teammate-oriented (positive)

25%

12%

62%

4: Manager-oriented (negative)

50%

20%

30%

3: Manager-oriented (positive)

10%

20%

70%

2: Team-oriented

27%

36%

36%

1: Company-oriented

30%

40%

30%

6: Teammate-oriented (negative)

27%

36%

36%

5: Teammate-oriented (positive)

25%

25%

50%

4: Manager-oriented (negative)

30%

40%

30%

3: Manager-oriented (positive)

20%

50%

30%

2: Team-oriented

18%

45%

36%

1: Company-oriented

30%

20%

Warn likelihood

100

50

0 Percentage

50% 50

100

Fig. 15 Socially conscious interface: endorse-vs.-warn results. (a) Legend for endorse-vs.-warn ratings. (b) Overall likelihood of endorsing and warning. (c) Likelihood of endorsing and warning, by framing scenario

Across all of the interfaces, participants assigned to the company-oriented scenario displayed consistently lower feelings of comfort and control relative to those who were assigned to the team-oriented scenario. We observed a similar result among those who were assigned to the manager-oriented scenario: Participants who had a negative relationship with the manager displayed lower feelings of comfort and control than those who had a positive relationship with the manager. These results suggest that users may be more likely to feel comfortable and in control of data shared among smaller groups with known and trusted individuals in workplace settings. However, these results flip among the teammate-oriented scenarios. Surprisingly, participants displayed greater feelings of comfort and control when a teammate with whom they had a negative and distrustful relationship had developed the system. Moreover, participants selected a lower level of anonymization in the negative teammate-oriented scenario, further confirming this result. We note that the negative

User Perceptions of Privacy Interfaces in the Workplace

129

teammate-oriented scenario explicitly mentions a positive relationship with their manager and other teammates, so we suspect that participants perceived a greater overall benefit of privacy interfaces in this scenario due to the contrast in relationship valence, though future work is needed to further investigate this finding. These findings indicate that it is important not just to focus on the literal design of our privacy interfaces but also to evaluate the designs with an explicit consideration of how the larger context of factors, such as power structures and social relationships, might result in varying levels of trust and differing reactions to a privacy intervention. Participants often struggled to fully understand technical privacy metrics even when they were explained in words rather than with strict numerical values, mirroring prior work on inaccurate user mental models of technical privacy measures (Dechand et al., 2019; Kang et al., 2015; Wu & Zappala, 2018). This lack of comprehension was paired with an inconsistent pattern of user confidence in their level of comprehension. While participants who correctly understood the metrics were generally more confident than those who did not, sizeable portions of the highcomprehension participants had low confidence in their understanding. Conversely, sizeable portions of the low-comprehension participants had high confidence in their understanding. We also found that participants with mid-to-low comprehension of these metrics felt a greater sense of comfort and control than those with high comprehension. Finally, we observed that users tended to choose similarly low anonymization levels regardless of their comprehension level. Users with higher comprehension of differential privacy selected higher levels of noise only slightly more often than those with low comprehension, and the “Medium” noise level remained the most common choice across user comprehension levels. Taken together, these results suggest that interfaces which heavily rely on privacy metrics appear to lend a sense of legitimacy and may provide users with a false sense of security. Moreover, users who do not fully understand these metrics are most likely to feel an inflated sense of security, even as they select less private settings. While these findings are preliminary, they suggest that technical privacy measures should be used with caution, as they have the potential to mislead users. These results support the notion that user-centered, design-based approaches to privacy are critical to complement technical, systems-oriented approaches.

5.1

Limitations

Our study was run with a relatively small number of participants, which limited our ability to confirm hypotheses with statistical tests or perform more robust subgroup analyses. Thus, our results should be considered exploratory and are intended to inform larger-scale privacy experiments. We also evaluated a narrow space of possible privacy interfaces: A larger-scale study with more fine-grained differences between interfaces could further elucidate which interface design patterns drive differences in privacy perceptions. Moreover, studies in which participants interact

130

M. S. Lam et al.

with functional interfaces using their own personal data could increase ecological validity. Finally, as indicated by our results on privacy metrics, further work is needed to design privacy interfaces that not only encourage perceptions of comfort and control but also lead users to make meaningful and informed privacy decisions in real workplace scenarios.

6 Conclusion In this work, we studied user perceptions of privacy interfaces in the context of workplace well-being support. Our findings indicate that interface designs based on privacy frameworks such as relationship-based access control and contextual integrity increase user perceptions of comfort and control. Moreover, the perceptions vary widely based on the owner of technology and the valence of relationships. Concerningly, technical privacy metrics were both poorly understood and led to a false sense of security for many participants. This work has implications for future studies on privacy interfaces in the workplace. Our research indicates that interfaces should not be evaluated in isolation of the deployment context, and solutions that heavily rely on technical privacy metrics have the potential to mislead users.

Appendix 1. Framing Scenarios 1.1 Company-Oriented You are an employee at a large tech company called SearchCo. The company has recently started a pilot program to improve employees’ general well-being by collecting data using work-issued devices like laptops and phones. To allow employees to manage the privacy of this data, the company has created an interface where employees can control what data the company is allowed to analyze.

1.2 Team-Oriented You are an engineer on the News App team at a large tech company called SearchCo. Your team (the News App team) has recently started a pilot program to improve team members’ general well-being by collecting data using work-issued devices like laptops and phones. To allow team members to manage the privacy of this data, the News App team has created an interface where employees can control what data the team is allowed to analyze.

User Perceptions of Privacy Interfaces in the Workplace

131

1.3 Manager-Oriented (Positive) You are an engineer on a small eight-person team, and you have a very positive, strong relationship with your manager, Jamie. You’ve worked with Jamie for 2 years already, and you trust that they make great decisions for the team. Jamie has recently started a pilot program to improve team members’ general well-being by collecting data using work-issued devices like laptops and phones. To allow team members to manage the privacy of this data, Jamie has created an interface where employees can control what data can be analyzed.

1.4 Manager-Oriented (Negative) You are an engineer on a small eight-person team, and you have a very negative, strained relationship with your manager, Jamie. You’ve worked with Jamie for 2 years already, and you’ve witnessed them making questionable decisions for the team. Jamie has recently started a pilot program to improve team members’ general well-being by collecting data using work-issued devices like laptops and phones. To allow team members to manage the privacy of this data, Jamie has created an interface where employees can control what data can be analyzed.

1.5 Teammate-Oriented (Positive) You are an engineer on a small eight-person team, and you have a very positive, strong relationship with your manager and teammates. You’ve worked on your team for 2 years already, and you trust that your manager and teammates all make great decisions. One of your teammates, Jamie, has recently started a pilot program to improve team members’ general well-being by collecting data using work-issued devices like laptops and phones. To allow team members to manage the privacy of this data, Jamie has created an interface where employees can control what data can be analyzed.

1.6 Teammate-Oriented (Negative) You are an engineer on a small eight-person team, and although you have a very positive, strong relationship with your manager and most teammates, you have a very negative, strained relationship with one of your teammates, Jamie. You’ve worked with this teammate for 2 years already, and you’ve witnessed them making questionable decisions for the team.

132

M. S. Lam et al.

Jamie has recently started a pilot program to improve team members’ general well-being by collecting data using work-issued devices like laptops and phones. To allow team members to manage the privacy of this data, Jamie has created an interface where employees can control what data can be analyzed.

References Adler, D. A., Tseng, E., Moon, K. C., Young, J. Q., Kane, J. M., Moss, E., Mohr, D. C., & Choudhury, T. (2022). Burnout and the quantified workplace: Tensions around personal sensing interventions for stress in resident physicians. Proceedings of the ACM on Human-Computer Interaction, 6(CSCW2), 1–48. Barth, S., de Jong, M. D., Junger, M., Hartel, P. H., & Roppelt, J. C. (2019). Putting the privacy paradox to the test: Online privacy and security behaviors among users with technical knowledge, privacy awareness, and financial resources. Telematics and Informatics, 41, 55–69. Conn, V. S., Hafdahl, A. R., Cooper, P. S., Brown, L. M., & Lusk, S. L. (2009). Meta-analysis of workplace physical activity interventions. American Journal of Preventive Medicine, 37(4), 330–339. Corbyn, Z. (2022). ‘Bossware is coming for almost every worker’: The software you might not realize is watching you. https://www.theguardian.com/technology/2022/apr/27/remote-worksoftware-home-surveillance-computer-monitoring-pandemic Costa, J., Guimbretière, F., Jung, M.F., Choudhury, T. (2019). Boostmeup: Improving cognitive performance in the moment by unobtrusively regulating emotions with a smartwatch. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 3(2), 1–23. Dechand, S., Naiakshina, A., Danilova, A., & Smith, M. (2019). In encryption we don’t trust: The effect of end-to-end encryption to the masses on user perception. In 2019 IEEE European Symposium on Security and Privacy (EuroS&P) (pp. 401–415). IEEE. Diel, S., Gutheil, N., Richter, F., & Buck, C. (2022). My data, my choice?! the difference between fitness and stress data monitoring on employees’ perception of privacy. In Proceedings of the 55th Hawaii International Conference on System Sciences. Dwork, C. (2008). Differential privacy: A survey of results. In International conference on theory and applications of models of computation (pp. 1–19). Springer. Fong, P. W. (2011). Relationship-based access control: protection model and policy language. In: Proceedings of the first ACM conference on Data and application security and privacy (pp. 191–202). Harwell, D. (2019). Is your pregnancy app sharing your intimate data with your boss? https://www. washingtonpost.com/technology/2019/04/10/tracking-your-pregnancy-an- app-may-be-morepublic-than-you-think/ Howe, E., Suh, J., Bin Morshed, M., McDuff, D., Rowan, K., Hernandez, J., Abdin, M. I., Ramos, G., Tran, T., & Czerwinski, M. P. (2022). Design of digital workplace stress-reduction intervention systems: Effects of intervention type and timing. In CHI Conference on Human Factors in Computing Systems (pp. 1–16). Humbert, M., Trubert, B., & Huguenin, K. (2019). A survey on interdependent privacy. ACM Computing Surveys (CSUR), 52(6), 1–40. Jun, E., McDuff, D., & Czerwinski, M. (2019). Circadian rhythms and physiological synchrony: Evidence of the impact of diversity on small group creativity. Proceedings of the ACM on Human-Computer Interaction, 3(CSCW), 1–22. Kang, R., Dabbish, L., Fruchter, N., & Kiesler, S. (2015). My data just goes Everywhere:” User mental models of the internet and implications for privacy and security. In Eleventh Symposium on Usable Privacy and Security (SOUPS 2015) (pp. 39–52).

User Perceptions of Privacy Interfaces in the Workplace

133

Kantor, J., & Sundaram, A. (2022). The rise of the worker productivity score. https://www.nytimes. com/interactive/2022/08/14/business/worker-productivity-tracking.html Kimani, E., Rowan, K., McDuff, D., Czerwinski, M., & Mark, G. (2019). A conversational agent in support of productivity and wellbeing at work. In: 2019 8th international conference on affective computing and intelligent interaction (ACII) (pp. 1–7). IEEE. Mandryk, R. L., & Inkpen, K. M. (2004). Physiological indicators for the evaluation of co-located collaborative play. In: Proceedings of the 2004 ACM conference on Computer supported cooperative work (pp. 102–111). Mark, G., Iqbal, S., & Czerwinski, M. (2017). How blocking distractions affects workplace focus and productivity. In: Proceedings of the 2017 ACM International Joint Conference on Pervasive and Ubiquitous Computing and Proceedings of the 2017 ACM International Symposium on Wearable Computers (pp. 928–934). Mendel, T., & Toch, E. (2017). Susceptibility to social influence of privacy behaviors: Peer versus authoritative sources. In: Proceedings of the 2017 ACM Conference on Computer Supported Cooperative Work and Social Computing (pp. 581–593). Nissenbaum, H. (2009). Privacy in Context. Stanford University Press. Partnership on AI. (2022). Framework for promoting workforce well-being in the ai-integrated work-place. Petronio, S. (2002). Boundaries of privacy: Dialectics of disclosure. Suny Press. Sloat, S. (2022). When my employer provides my mental health app, how much data do they have access to? https://foundation.mozilla.org/en/blog/mental-health-awareness-2022-employeraccess/ Sweeney, L. (2002). Achieving k-anonymity privacy protection using generalization and suppression. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 10(05), 571–588. Tyagi, A., Squicciarini, A., Rajtmajer, S., & Griffin, C. (2016). An in-depth study of peer influence on collective decision making for multi-party access control. In: 2016 IEEE 17th International Conference on Information Reuse and Integration (IRI) (pp. 305–314). IEEE. Wu, J., & Zappala, D. (2018). When is a tree really a truck? exploring mental models of encryption. In: Fourteenth Symposium on Usable Privacy and Security (SOUPS 2018) (pp. 395–409).

Assisting Learning and Insight in Design Using Embodied Conversational Agents Rebecca Currano and David Sirkin

Abstract Reflective practices correlate with both insight in engineering design projects (Currano. 2015 Reflective Practice in Engineering Design) and learning in engineering education (Chew et al. 2016 ASEE Annual Conference & Exposition 2016). Schön (Knowledge-Based Systems 5:3–14. 1992) observed that designers engage in a reflective conversation with design materials as they sketch and prototype their ideas. However, students can struggle when learning domain knowledge and design practices simultaneously. We hypothesized that (1) making the conversation between student designers and their prototyping materials literal and explicit, rather than metaphorical, can assist in reflection, and (2) the effect can be enhanced by embodying a conversational agent within the prototype itself. We developed an embodied conversational agent as a tool to elicit reflection and lead to greater insight and learning, during a hands-on mechatronics prototyping and design activity. In addition to guiding learners through a tutorial, the agent answers questions, offers comments, and asks deep reasoning and generative design questions, which correlate with convergent and divergent phases of the design process (Eris. DS 31: Proc. 14th Intl. Conf. on Engineering Design (ICED). 2003). We compare learners’ gained knowledge, performance, and reactions in conditions with or without a conversational agent, and using an embodied or non-embodied agent, to evaluate the impact of (1) conversational communication, and (2) physical embodiment, on learning and insight. In this chapter, we describe the background and rationale for our study, details of the study design, and preliminary insights from our pilot data.

1 Context and Human Need In The Reflective Practitioner, Donald Schön developed the concept of reflection-inaction—a process of in-the-moment reevaluation and redirection of design activity—in part through observation and interviews with student and practicing R. Currano · D. Sirkin (✉) Stanford University, Mechanical Engineering, Center for Design Research, Stanford, CA, USA e-mail: [email protected]; [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_7

135

136

R. Currano and D. Sirkin

designers engaged in open-ended architectural design tasks (Schön, 1983). Over the last decade of working with engineering design students, we have observed that student learning and attention are often split between two contexts: (1) the domain knowledge that underlies the design task and (2) the process of design itself. Being students, these novice designers are neither formally trained in nor naturally skilled at reflecting on their process in the midst of action. They learn how to reflect productively through practice and a concerted effort. Seeking help and consulting with design educators, practitioners, or domain experts can support students’ domain and process knowledge, but not necessarily their reflective practices. As a result, design students often lack direction regarding, cognizance of, and experience with reflective practice and thus may struggle in drawing broader insights during their design projects. In addition, the search for these different types of knowledge entails interrupting the flow of design activity to identify and contact the requisite expert(s), elicit information, and then synthesize learnings and how to apply them. This switching between contexts can be a further challenge to reflection, but it offers an opportunity to develop tools to elicit and guide reflection on process and content that can be applied during design activity.

2 Prior Work and Background In extending his work on reflection-in-action, Schön (1992) observed that designers learn and gather information, identify patterns, construct meaning, and assess qualities in a process of seeing-moving-seeing through reflective engagement with various media, including sketches, prototypes, and components. He called this a “conversation with the materials of a design situation,” highlighting the emergence of unintended effects of moves (e.g., changes, additions, reconfigurations), which result in the evolution of both the design itself and the designer’s intentions.

2.1

Design Is a Socio-Technical Process

While Schon’s and other works have recognized the integral role visual thinking plays in design (McKim, 1972), more recent work also shows the importance of verbal, spoken conversation in helping designers distill insights from and about their design tasks (Currano & Steinert, 2012). Sonalkar et al. (2016) developed a tool (the Interaction Dynamics Notation tool, or IDN) to help identify patterns of team interaction—which were primarily conversational—and showed that these interactions correlate with creativity measures in concept generation design outcomes. This makes sense, given the socio-technical nature of the design process (Cross & Cross, 1995), and particularly for teams actively communicating with each other and interacting with shared prototypes.

Assisting Learning and Insight in Design Using Embodied Conversational Agents

137

Cross’ seminal research on the intersection of the social process of design with the technical and cognitive processes was based heavily on observation and analysis of the team conversations during their design activity.

2.2

Question-Asking in Design

Also focusing on design processes, Dym et al. (2005) characterized design, in part, by the focus on asking and answering questions. In particular, Eris (2003) identified specific types of questioning behavior that designers engage in—deep reasoning questions (DRQs) and generative design questions (GDQs)—and found that these correlate, respectively, to convergent and divergent phases of design. Choi et al. (2005) observed that student designers found it difficult to generate quality questions. However, using a text-based scaffolding tool to support peerquestions, they found that simple adaptive questioning prompted reflection, facilitating metacognition, and enhancing learning during online discussions. Schön’s seeing-moving-seeing process suggests that design learning happens through a reflective hands-on engagement with design materials and that the materials used by designers in their work serve two purposes: (1) as components of the prototyping and sketching process and (2) as media for a reflective learning process. Additionally, the research in question-asking and the social process of design together indicate benefit in not only asking questions but doing so within the context of spoken conversation. We therefore believe that more integrally combining the two roles of design materials by literally endowing the materials with the capabilities of conversing about the design process and progress, and adaptively asking convergent and divergent questions, will enhance design and learning processes, especially for novice designers.

2.3

Physically Embodying Materials

The concept of embodying materials and endowing them with the ability to verbally engage designers has been explored in recent studies. In the context of an electronics prototyping activity, Jung et al. (2014) tested whether an agent that was embedded in or external to the physical material affected students’ perceptions and learning outcomes while also varying the agent’s expressions of social interest toward the participant and activity. They found that an embedded agent lowered perceptions of task stress, and an interested agent was more likable and socially present. The influence of either condition on learning outcomes was mixed but suggested relevance to students’ prototyping confidence and performance. Using a similar prototyping activity, Martelaro et al. (2016) tested how varying the agent’s expressivity and vulnerability influenced students’ perceptions of trust,

138

R. Currano and D. Sirkin

disclosure, and companionship. They found that increased vulnerability led to higher ratings of trust and companionship, and increased expressiveness produced higher ratings of disclosure, with trust mediating vulnerability and companionship. These studies directly inform our guiding questions, research context, and implementation. While their hypotheses focus on affective expressions of emotion and attention, and their measures emphasize social relations, our approach addresses a gap in understanding around the role of (a) interactive questioning conversation in an agent-based learning context and (b) a teaching agent’s physical embodiment, in supporting learners’ abilities to reflect and build insight during a design task.

3 Research Approach 3.1

Research Questions

Bringing these three themes together—the role of reflection, spoken conversation, and divergent and convergent questioning—inspired us to study design learning in the more interactive and physical contexts of teaching mechatronics. Our guiding research questions are therefore as follows: 1. Can conversational agents help novice designers reflect, develop insight, and improve learning by making their conversation with design materials more literal and explicit? 2. Furthermore, given the physicality of working with materials and prototyping, can embodying these conversational agents support and enhance reflection, insights, and learning for engineering design tasks?

3.2

Hypotheses

The aim of this work is to thoughtfully integrate conversation and question-asking into the prototyping process for novice designers to enhance their learning and design outcomes. Given the social nature of design and the implicit conversations embodied in seeing-moving-seeing that lead to the emergence of new insights in the reflective process, we hypothesize that. 1. Making the conversational layer explicit (via an agent that asks the learner questions and answers learners’ questions) will spark deeper/broader seeing and encourage more diverse moving. 2. The physicality afforded by embodying the conversational agent within the prototype being designed (e.g., as a robot) will assist novice designers’ reflective process and reduce shifts in attention.

Assisting Learning and Insight in Design Using Embodied Conversational Agents

3.3

139

Research Setting

We have developed a conversational agent, embodied as a robot, that responds to the moves that learners make and asks questions that elicit and guide reflection within their process, thereby helping them to learn from and discover new insights and evolve their designs. Mechatronics prototyping is an opportune context to study the use of conversational agents in the design process. It provides a natural setting to integrate an embodied conversational agent into the prototyping materials and to test its effects on insight and learning. For example, Martelaro et al. (2016) examined human-robot conversation within a mechatronics tutorial, although they focused on the social aspects of trust, disclosure, and companionship rather than learning and insight. An introductory mechatronics tutorial that we developed and have taught over the last several years at the ACM Conference on Human Factors in Computing Systems (Sirkin & Ju, 2014) serves as our starting point (see Fig. 1).

3.4

Study Design

In an experiment employing this conversational agent (implemented using a humanin-the-loop protocol), participants follow an abbreviated version of the conference tutorial to build a small paper robot and learn basic mechatronics. The study is split into four sections: a pre-activity survey to evaluate participants’ knowledge of and comfort with electronics prototyping, the learning tutorial, a sketched design task, and a post-activity survey to evaluate student learnings and perceptions of the agent. The entire study takes about 1 hour. The study has a 2 (conversational vs. non-conversational) x 2 (embodied vs. non-embodied) design, each corresponding to one of our two hypotheses. In the conversational condition, the agent speaks in an engaging and interactive style and asks questions, while in the non-conversational condition, it communicates the same background and instructions but speaks in a noninteractive style without asking questions. In the embodied condition, the agent refers to itself,

Fig. 1 Paper robot project from our conference course in electronics prototyping using Arduino

140

R. Currano and D. Sirkin

and its voice projects from the paper robot, while in the non-embodied condition, it refers to the robot as a separate entity and its voice comes from the laptop that participants use to program the robot.

3.4.1

Conversational Vs. Non-conversational

During the tutorial, both conversational and non-conversational agents offer guidance on how to assemble the robot. They provide information such as component names and uses (e.g., “A potentiometer is a variable resistor. When you turn the knob, the resistance changes between the middle pin and the outer pins.”), and they also offer suggestions about how to correctly connect components together (e.g., “You will need to add a resistor first, to limit the current in my/the LED to a safe value. Otherwise, it might burn out.”). The conversational agent uses a slightly less formal speaking style appropriate to spoken conversation, while the nonconventional agent uses a style appropriate to written instructions. In addition, the conversational agent interjects DRQs (e.g., “A photocell is an ambient light sensor. What do you think makes its resistance change?”) and GDQs (e.g., “Can you tell me a few different things you could use a servo to do?”) to inspire convergent and divergent oriented reflection. In both conditions, participants can ask the agent questions and receive answers about components or instructions or ask for statements to be repeated, to ensure that they have adequate access to the information presented. However, in the conversational condition, the agent tells participants “You can ask me questions at any time.”, whereas in the non-conversational condition, it instructs them to direct their questions to the study administrator.

3.4.2

Embodied Vs. Non-embodied

During the tutorial, both embodied and non-embodied agents provide the same information but in a different manner. The embodied agent speaks in a selfreferential style (e.g., “Hello, my name is Robbie. I’m a robot and today you are going to build me.” and “You can plug my potentiometer into the blue sockets of my breadboard.”), in contrast to the non-embodied agent, which refers to the robot as a separate entity (e.g., “Today you will be building a small paper robot.” and “Look for the 10K Ohm resistor on the robot’s breadboard.”). In both cases, the agent speaks with the same synthetic male voice, raised in pitch and slowed in rate 15% from its default setting, to better suggest it as being a rough prototype—more youthful and less authoritative. At the same time, the embodied agent’s voice emanates from a small Bluetooth speaker mounted to the same physical platform as the robot’s other components (see Fig. 2), while the non-embodied agent’s voice originates from the speakers of the laptop computer that participants use to edit Arduino sketches and upload them to the robot.

Assisting Learning and Insight in Design Using Embodied Conversational Agents

141

Fig. 2 Paper robot project from the current study, showing the mechatronics platform in three states: (left) as participants first encounter it, (center) assembled to complete its functionality, and (right) with components moved off-board and ready to be inserted into the cardboard box in Fig. 1

3.4.3

Designing a New Robot

After participants complete the tutorial, they complete a task to design one or more different robots using the same types of components, thinking about what they would want their robot to do and how it might do those things. Participants are encouraged to think about components at different scales, to incorporate as many components as they like, and to create additional robot designs if they want. They are also asked to make sketches and notes as they work out their ideas. To encourage novelty, our instructions state that we will evaluate all participants’ designs and rank them based on creativity, but we will not evaluate their drawing skills, just how they reimagine or reapply the tools they learned in the tutorial. The design task is timed and takes 15 minutes.

3.5

Data, Measures, and Analysis

Data are collected from three sources: (1) physical actions and conversation by participants, (2) sketches from the robot design task, and (3) pre- and post-activity questionnaire and post-activity interview responses. Measures include the depth and breadth of design ideation, instances of reflective practice (manifest as physical or verbal conversations with materials), locus of attention, and learning, across study conditions.

3.5.1

Physical Actions and Conversation

During the tutorial and robot redesign task, we video-record participants’ physical actions and spoken interactions with the agent for coding and analysis (Heyman et al., 2014) to identify behavioral patterns and understand levels of engagement across conversational and embodiment conditions. For coding, two to three

142

R. Currano and D. Sirkin

researchers codevelop an evaluation framework based on a subset of the data, using Cohen’s kappa (Lombard et al., 2002) to check inter-rater reliability, and then they apply it to the remaining participant interactions. An initial analysis of pilot participant data will focus on two alternatives as proxies for reflection and learning: 1) actions such as asking follow-up questions and manipulating materials while ideating or answering questions and 2) performance measures such as task completion time or number of mistakes.

3.5.2

Robot Designs

We also analyze the notes, descriptions, and sketches resulting from the robot design task, focusing on (1) flexibility, through changes made to the robot’s form, functions, context of use, or the means through which these new functions are accomplished, and (2) fluency, through participants’ ability to apply this knowledge to create a breath of distinct designs.

3.5.3

Questionnaire

We query participants through pre- and post-task surveys and a brief interview, first about their prior domain knowledge and design process experience and then about their design intentions, learnings, insights, and reflections. Questions include novel and standardized measures, such as hedonic and pragmatic user experience scales (UEQ-S) (Schrepp et al., 2017). Questionnaire data will be fit using generalized linear or mixed effects models, which offer flexibility in data form and conditions and support random effects for participants and conditions, potentially lowering the chance of type 1 error (Boisgontier & Cheval, 2016). We also include questions to confirm whether participants recognize and respond to the manipulations within their study condition, including direct questions and measures of social presence such as copresence, involvement, and engagement (Biocca et al., 2003).

4 Preliminary Findings 4.1 4.1.1

Revisions Made while Piloting Clarifying Conditions

In pilot testing, we found that having the agent ask intermittent check-in questions between tasks in the instructions such as “Are you following along so far?” encouraged participants to remain engaged in the conversation and speak up at other times. We also added small interjections like “Ok, great!” and “Good work!” to acknowledge participants’ statements and answers and comply with basic principles of

Assisting Learning and Insight in Design Using Embodied Conversational Agents

143

politeness in conversational interaction. These additions help separate conversational and non-conversational conditions. Also, to be sure participants correctly perceive the embodied and non-embodied conditions, we added three checks. These were (1) a manipulation check asking participants where the agent’s voice seemed to come from (robot vs. computer vs. somewhere else), (2) verification of how extensively they talked with the agent (not at all, somewhat, a lot), and (3) identification of the entity with whom it felt like they were speaking (the robot they were building, the computer, or something else).

4.1.2

Clarifying Instructions

During the tutorial task, participants were confused by the presence of lettering and numbering on both the sockets of the breadboard and the pins of the Arduino. Updating the early portion of the tutorial to note these distinctions resolved their uncertainty. Participants also missed hearing or sometimes forgot component names. Labeling the components in addition to describing them helped participants identify them more easily. Finally, participants often got distracted when finding, examining, and assembling components. By better separating and reordering operations, they could more easily keep focused on one task at a time.

4.2

Observations of Reflective Behavior

In the conversational conditions, we encourage reflection by asking questions about components just after participants pick them up or plug them in. Similarly we ask questions about programming component behavior just after they look at the Arduino code. In the non-conversational conditions, we encourage reflection by telling participants to notice or observe what happens after they manipulate components or upload the Arduino code. Pilot data in the conversational embodied condition demonstrates that when participants are asked what alternative things a servo might be used for, they often hold the components up and test how they work, squeezing the force-sensing resistor and watching the servo rotate a few times, which then provides inspiration for their answers. Likewise, when they are asked how else they might modify the ambient light coming from a photocell, they often wave their hands over the sensor and observe its effect. These examples show that asking questions during the tutorial can help promote learners engaging in reflective conversations with the materials of a design situation and that we can observe the process of seeing-moving-seeing within these conversations. We also have preliminary evidence that speaking reflections out loud, rather than reflecting silently, can help learners think about components and connections and think through their answers to questions, more effectively. We observed some

144

R. Currano and D. Sirkin

instances of participants reflecting aloud about various instructions, wondering for example, how they can know whether the Arduino (or some other source) is powering the LEDs or if it would be safe to plug in a second LED without adding a second resistor. Likewise, arm and hand gestures may help support the ideation process, as we observed participants gesticulating as they thought (and spoke) through their answers to GDQs, which sometimes inspired additional ideas.

4.3

Observations of Robot Redesign

Participants took very different approaches to creating and describing their designs. One sketched several robots, with minimal text, keeping the tutorial robot’s form but changing which components were used for the eyes, nose, and arms and adding other features such as a mouth or ears (see Fig. 3). Another used almost exclusively text, focusing on a single component in each design, stating a task that a robot would use it to do (e.g., kick a ball or erase a whiteboard) without describing other aspects of the robot (see Fig. 4). Others took a more needs-oriented approach and used a combination of sketches and accompanying text to describe how a new device might use one or two components to solve an everyday problem (e.g., call a waiter in a restaurant or keep a laptop from going to sleep) (see Fig. 5).

Fig. 3 One pilot participant’s robot redesign sketches focused on alternative configurations, keeping the overall design but changing or adding components and features

Fig. 4 One pilot participant’s robot redesign designs were primarily text-based, with each focusing on one function the robot could perform

Assisting Learning and Insight in Design Using Embodied Conversational Agents

145

Fig. 5 One pilot participant’s robot redesign sketches combined text and images, with each focusing on solving a human need in a different context

The most creative of the designs were the ones that focused on needs. These explored how components might be used in a new way, and a specific setting, to solve a real-life problem participants had experienced.

4.4

Questionnaire Responses

Pre- and post-activity questionnaires and interviews included several notable responses. Referencing her experience building the robot after the tutorial, one participant said that she “didn’t know what was happening until I closed the lid”

146

R. Currano and D. Sirkin

of the cardboard box. Another said her favorite part of the activity was “just seeing the robot at the very end.” Participants knew they were building a robot, given the tutorial’s multiple references to “the robot.” However, these comments suggest that as they were working through the tutorial, participants did not have a sense of the robot’s final state or that it would be noticeably different from the assembled components. This raises the question of whether to display an image or physical copy of the completed robot when participants begin the activity. In the end, we decided against this, as doing so would have (a) focused participants on reaching an end goal rather than participating in a learning process and (b) diminished their sense of discovery upon completing the tutorial. This also raises the question of how reflecting on steps of the tutorial or functions of the components differs from reflecting on the overall experience and whether embodiment or asking questions can prompt the latter during different stages of system integration. To avoid interrupting the learning process, we decided against asking participants to reflect on the overall experience during the tutorial. Regarding the difficulty of the tutorial, participants noted that it was “easier than I imagined,” “wasn’t as hard as I imagined,” and “very clear for the most part, for someone without any experience.” They also agreed that they had learned from the two activities, describing their learnings as “how to put together a robot, also what some components are, and how Arduinos work.” When asked what was most challenging about the redesign task, participants focused on ideation, including “I wasn’t sure I would have any ideas,” “just figuring out my ideas,” and to “trying to think about something to make,” suggesting that they may yet lack confidence in their new knowledge. But even so, all participants were able to produce multiple redesign ideas. Participants also completed Likert-style questions rating the pace, length, and difficulty of the tutorial. Pace and length ratings fell at around 4, generally corresponding to the written comments and indicating that each was appropriate. Difficulty ratings ranged from 1 to 3, which is the lower end of the scale, but we consider this appropriate as the tutorial is designed to make electronics accessible to inexperienced learners, and being slightly easy is preferable to being slightly difficult. Responses also may not perfectly match experiences, as we observed participants experiencing challenges, such as remembering component names, discerning which sockets to plug components into, and responding to “how else might you. . .” questions. Altogether, the responses and ratings suggest that the study is well matched to our participant population and timeframe without being trivial and that it prompts learning.

5 Moving Forward Building on the discussion of Data, Measures, and Analysis above, the corpus of study data will comprise the following:

Assisting Learning and Insight in Design Using Embodied Conversational Agents

147

• Behavioral coding of participants’ physical actions and conversations, which will now focus on confirming reflective behaviors observed during pilot testing, such as lifting, holding, and testing components when asked DRQs. • Sketches of robot designs, which we will evaluate using measures such as idea fluency (depth), flexibility (breadth), detail and completeness, focus of the robot’s task (such as solving users’ needs vs. featuring robots’ functions or form), and use of the provided components in different ways. • Analysis of questionnaire and interview responses, including a comparison of pre- and post-activity responses to measure self-assessment of learning. The pilot’s outcomes thus far inform experiment design for future studies of agents in support of ideation and learning, as well as technology-supported education. We expect the full study’s results to be relevant to the human-computer (HCI) and human-robot (HRI) interaction research communities.

References Biocca, F., Harms, C., & Burgoon, J. (2003). Toward a more robust theory and measure of social presence: Review and suggested criteria. Presence: Teleoperators & Virtual Environments, 12(5), 456–480. Boisgontier, M., & Cheval, B. (2016). The ANOVA to mixed model transition. Neuroscience & Biobehavioral Reviews, 68, 1004–1005. Choi, I., Land, S., & Turgeon, A. (2005). Scaffolding peer-questioning strategies to facilitate metacognition during online small group discussion. Instructional Science, 33(5), 483–511. Cross, N., & Cross, A. (1995). Observations of teamwork and social processes in design. Design Studies, 16(2), 143–170. Currano, R., & Steinert, M. (2012). Characterizing activities that promote ideation: Survey construction targeting reflective practices. In DS 73–1 proc. 2nd Intl. Conf. On design creativity (Vol. 1, pp. 297–306). Dym, C., Agogino, A., Eris, O., Frey, D., & Leifer, L. (2005). Engineering design thinking, teaching, and learning. Journal of Engineering Education, 94(1), 103–120. Eris, Ö. (2003). Asking generative design questions: A fundamental cognitive mechanism in design thinking. In DS 31: Proc. 14th Intl. Conf. On engineering design (ICED) (pp. 587–588). Heyman, R., Lorber, M., Eddy, J., West, T., Reis, E., & Judd, C. (2014). Behavioral observation and coding. In Handbook of research methods in social and personality psychology. Cambridge University Press. Jung, M., Martelaro, N., Hoster, H., & Nass, C. (2014). Participatory materials: Having a reflective conversation with an artifact in the making. In Proc. 2014 conference on designing interactive systems (DIS) (pp. 25–34). Lombard, M., Snyder-Duch, J., & Bracken, C. (2002). Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research, 28(4), 587–604. Martelaro, N., Nneji, V., Ju, W., & Hinds, P. (2016). Tell me more: Designing HRI to encourage more trust, disclosure, and companionship. In Proc. 11th ACM/IEEE Intl. Conf. On humanrobot interaction (HRI) (pp. 181–188). McKim, R. (1972). Experiences in visual thinking. Brooks/Cole Publishing. Schön, D. (1983). The reflective practitioner: How professionals think in action. Basic Books.

148

R. Currano and D. Sirkin

Schön, D. (1992). Designing as reflective conversation with the materials of a design situation. Knowledge-Based Systems, 5(1), 3–14. Schrepp, M., Hinderks, A., & Thomaschewski, J. (2017). Design and evaluation of a short version of the user experience questionnaire (UEQ-S). Intl. Jnl. Interactive Multimedia and AI, 4(6), 103–108. Sirkin, D., & Ju, W. (2014). Make this!: Introduction to electronics prototyping using Arduino. In Ext. abstracts ACM CHI conf. On human factors in computing systems (CHI) (pp. 1017–1018). Sonalkar, N., Mabogunje, A., Pai, G., Krishnan, A., & Roth, B. (2016). Diagnostics for design thinking teams. In Design thinking research (pp. 35–51). Springer.

How to Tame an Unpredictable Emergence? Design Strategies for a Live-Programming System Marcel Taeumel, Patrick Rein, Jens Lincke, and Robert Hirschfeld

Abstract Programming environments that provide a feeling of liveness help professionals and amateurs alike to approach unfamiliar domains with ease through short feedback loops. Exploration and experimentation are promoted because any change to the program under construction can be observed immediately. However, live-programming systems such as Squeak/Smalltalk struggle with the predictable emergence of adapted program behavior as object communication can be unconstrained and diverse. While programmers wish for immediate effects, it would be helpful to at least know whether anything will happen after some time. In this chapter, we take a closer look at the means available in Squeak to explore and adjust object state and object behavior so that programmers can ensure the system’s responsiveness and hence observe gradual or even induce eventual emergence. We argue that these design strategies are sufficient to architect communication patterns that reward changes with immediate effects. We believe that our work can help programmers to better understand their leverage toward a predictable emergence in systems whose liveness stems from objects and messaging in a space where tools and applications live side by side.

1 Introduction Programming feels empowering. We express our creative thought through digital media to experience and learn from new perspectives on our favorite domains. Artistic crafts, scientific simulations, and engineering works: People can leverage the human intellect through computers. There are programming concepts, languages, and systems that can match—or challenge—our cognitive abilities. It can be intriguing (and fun) to explore a system’s responses to our (creative) questions. But isn’t

M. Taeumel (✉) · P. Rein · J. Lincke · R. Hirschfeld Hasso Platter Institute, University of Potsdam, Potsdam, Germany e-mail: [email protected]; [email protected]; [email protected]; [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_8

149

150

M. Taeumel et al.

Fig. 1 The exploratory feedback loop: Programmers translate their goal-oriented intents into actions that let the tool-driven system adapt the execution to then yield (observable) effects, which are to be interpreted and evaluated when planning for the next steps. A feeling of liveness occurs when the loop’s frequency is high; direct manipulation (Hutchins et al., 1985) helps; continual responsiveness (Shneiderman & Plaisant, 2010) is crucial. However, a timely emergence can be challenging to achieve (Rein et al., 2016)

programming hard to master? Maybe. Fortunately, the idea of liveness in a programming system makes programming more accessible—and exploration more efficient. Exploratory programming is expedient for professionals and amateurs alike. Iteratively, any problem domain unfolds through experimentation; trial-and-error is encouraged for the sake of learning (Sheil, 1998; Trenouth, 1991). That is, expert programmers are typically most curious toward unfamiliar challenges (Gabriel, 2014), initially walking in the darkness like beginners, fumbling for light. Along the way, exploring the problem and solution space entails multiple conversations (Taeumel et al., 2022) with the programming system (and environment) at hand. Questions and responses. Input and output. Mouse clicks and animated graphics. Immediacy (Ungar et al., 1997) plays an important role in this feedback loop (Hutchins et al., 1985). If the system can answer in a timely fashion, programmers can safely follow their train of thought. Therefore, a feeling of liveness can boost the efficiency of exploratory practices (Taeumel et al., 2022; Taeumel et al., 2021). We consider a programming experience as live when the system takes only a few milliseconds to adapt its behavior and respond in a meaningful way. See Fig. 1 for such an idealized feedback loop. In our setting, the act of live programming means continuously modifying a running system so that its observable effects immediately yield added value. Unlike other programming flavors, we focus on systems that allow for updating executable software artifacts in situ while avoiding restarts and loss of volatile information. Programmers still have to come up with an actionable plan, but the programming system at hand will not disrupt their train of thought with unexpected pauses or unclear effects. Instead, all the tools and methods in the system will be like an unobtrusive extension of the mind. The programming system of our choice is made of objects that communicate via messages (Taeumel & Hirschfeld, 2016). This communication yields observable effects forming all kinds of object-oriented applications such as productivity tools, multimedia games, and educational simulations. Most notably, programmer tools and user applications share a single object space, which can expedite exploration (and experimentation) through short feedback loops. That is, tools allow for atomic

How to Tame an Unpredictable Emergence? Design Strategies for. . .

151

updates of object identity, state, or behavior; the system will adapt immediately. Unfortunately, programmers might not notice (and understand) effects immediately as objects communicate concurrently and not always in an observable way. The objects in question might not even participate in any conversation at the moment and maybe they never will. Fortunately, our programming system—Squeak/ Smalltalk1—provides several means to cope with such unpredictable emergence. What does “emergence” mean in other domains? How can this concept help us understand the challenge of achieving liveness in programming (systems)? We ground our explanations on the following definitions: In philosophy, systems theory, science, and art, emergence occurs when an entity is observed to have properties its parts do not have on their own, properties or behaviors that emerge only when the parts interact in a wider whole. [. . .] —Wikipedia on “emergence” (2022-11-10) Emergence: the act or an instance of emerging. Emerging: newly formed or prominent. —Merriam-Webster on “emergence” (2022-11-10) The next phase comprises the emergence of an observable change in the behavior of the application from the adapted executable form. An observable change can be for example a changed textual output on the console, a different color of a graphical element, or a changed way of moving of a graphical element. [. . .] —Rein et al. on “emergence” (Rein et al., 2016)

As illustrated in Fig. 1, we define “emergence” as the time between after a system’s adapted execution and before a programmer comes up with the next idea on how to continue exploration. That is, we see liveness as primarily user-centered because observation, interpretation, and evaluation (of effects) all lead toward a better understanding. The system keeps on doing what it is currently programmed to do; the user (or programmer) sits nearby and perceives (visual or audible) output. A timely emergence can then lead to a feeling of liveness,2 which makes exploration of problem and solution space more efficient. Now, emergence becomes unpredictable not because a certain effect is supposed to show but because something different happens. Instead, any effect is supposed to happen and be seen immediately, or at least after some time (Tanimoto, 2013). That is the challenge: The Squeak/Smalltalk system (Fig. 2) cannot guarantee a timely emergence because the objects of interest might not currently contribute to (observable) system behavior. The emergence is unpredictable in the general sense. Yet, there are predictable, domain-specific live-coding systems such as for audio or visual performance art (Rein et al., 2018), which leads us to our research question: 1

The Squeak/Smalltalk programming system, https://squeak.org Note that a feeling of liveness needs more than a timely emergence. A direct-manipulation interface (Hutchins et al., 1985) helps users to directly map their intents to actions (e.g., tangible object commands) as well as quickly figure out the meaning of what they can see (e.g., visual object representations)0.2. Note that a feeling of liveness needs more than a timely emergence. A directmanipulation interface (Hutchins et al., 1985) helps users to directly map their intents to actions (e.g., tangible object commands) as well as quickly figure out the meaning of what they can see (e.g., visual object representations). 2

152

M. Taeumel et al.

Fig. 2 The Squeak/Smalltalk programming system: tools and applications are made of objects, which communicate via messages to yield the desired behavior. Here, stars (★) denote such objects by example. Especially code objects have no visual representation other than (structured) text

How can we tame an unpredictable emergence in a general-purpose, objectoriented programming system to provide a feeling of liveness? For simplicity, we assume that all other aspects of the feedback loop (Fig. 1) are already optimized: (1) object-based design plan (Meyer, 1998), (2) direct mapping of intent to action (Hutchins et al., 1985; Maloney, 2002), and (3) very fast adaptation (Rein et al., 2016). What remains is the challenge after the system has adapted its execution: the unpredictable emergence. Continual responsiveness3 is crucial for uncovering gradual emergence in systems that struggle with predictability (and immediacy) for generic, unoptimized cases. That is, programmers can only use exploration tools within the system if one group of objects does not prohibit the communication in other important groups of objects. For example, mouse clicks should work or event timers fire even if a costly algorithm consumes precious computing time. Remember that objects represent tools and applications, and programmers use tools to inspect or modify these objects. Such means for exploratory programming allow for mitigating unpredictable emergence by manually “poking around” the objects that should be affected, maybe giving them an explicit “push” to see what happens. However, the tools will not work if the system becomes unresponsive. This goal of continual object communication leads us to our second research question: 3

Desirable response times depend on the particular use cases (Shneiderman & Plaisant, 2010, p. 445). Typing and cursor movement lies somewhere between 50 and 150 milliseconds. Simple frequent tasks such as button clicks should be no longer than 1 s until something happens. If it takes too long, users cannot make the connection between their action and the system’s response. They might lose track of cause and effect.

How to Tame an Unpredictable Emergence? Design Strategies for. . .

153

How can we design object communication to keep the system responsive to uncover gradual emergence or induce eventual emergence? Note that our programming system offers general-purpose tools to create domainspecific applications. If it can remain responsive during divergent experiments, insights can converge into specific forms that offer immediate emergence during onward modification (or use). We are interested in laying out this design space, which enables a transition from gradual (or eventual) to immediate feedback. Programmers can then deliberately benefit from actual4liveness at some point when working on their project. There is one more building block in Squeak to better fathom the potential of continual object communication: processes. Each process represents a communication thread where a chain of messages is to be resolved; objects in that chain are waiting for a response; the one object at the end might just provide a response; or it needs to talk to another object first to do so. In programming terms, each process has a stack of activated methods, which grows and shrinks all the time. There are many processes in the system, but only a single one can run at a time.5 To ensure fairness, scheduling depends on processes to give up computing time in a cooperative fashion. Then, the next process in line will get its turn; the current one goes back to the end. Additionally, each process has a priority and means to suspend its execution until a certain event (or semaphore) gets signaled. Processes with higher priority will always interrupt the ones with lower priority. Consequently, the system’s responsiveness will be impaired if the processes responsible for user input and graphics output do not get their fair share of computing time. In this chapter, we document design strategies that already exist in the Squeak/ Smalltalk system or are provided by selected research extending that system. We try to generalize our findings toward continual object communication so that other object-based systems might benefit from these ideas to improve their liveness as well. However, we are aware that our thoughts are biased—or too idealized—which could easily lead us omitting challenges (or hard constraints) specific to other ecosystems. Nevertheless, we examine the following strategies:

4 To this day, the most promising systems for live programming offer level-4 liveness (Tanimoto, 2013). These systems are “informative, significant, responsive, and live.” Program descriptions are human-readable and human-executable; modifying these descriptions changes system execution automatically without any noticeable delay. Systems with level 5 or 6 would be tactically and/or strategically predictive, but we do not know how such cleverness could be implemented. Squeak/ Smalltalk provides level-4 liveness without immediate emergence (Rein et al., 2016), which, however, programmers can deliberately induce in their projects. 5 Note that such green threading makes the virtual machine more robust because programmers cannot corrupt the object memory inadvertently. In an informal way, they can explore and experiment unless their experiments botch the processes (or means) responsible for continual responsiveness.

154

M. Taeumel et al.

Section2 Programmers can write down and evaluate snippets of source code in almost any text field to explore the objects at hand. Section3 An extensible hierarchy of exceptions helps programmers understand and fix erroneous object communication. Section4 Code simulation enables a step-by-step observation of objects exchanging messages; time stands still to be explored and continued on demand. Section5 Users can hit a special key combination to interrupt an apparently unresponsive system to then explore it step by step. Section6 A graceful retreat from one interactive—yet unresponsive—framework to another, still responsive, one allows users to fix one “kind of liveness” with the help of another one. Section7 Watchdog processes can perform all kinds of process-monitoring tasks such as progress supervision, preemptive scheduling, and recursion detection. Finally, in Sect.8, we conclude our thoughts on how to tame the unpredictable emergence in the Squeak/Smalltalk live-programming system.

2 Exploratory Workspace Everywhere Programmers can type and evaluate code in any text field. The most prominent tool, which is just a text field plus window decorations, is the workspace. We documented several patterns around this conversational style with the system (Taeumel et al., 2022). Yet, considering liveness and emergence, we focus on the basic mechanism that allows for talking to objects explicitly. Programmers might want to understand why a group of objects is or is not exhibiting a changed behavior. That is, they might want to observe gradual or induce eventual emergence. Text (and source code) is a convenient and direct way of expressing one’s thoughts. Users can type and easily revise their textual expressions; following an executable language helps the system understand the user. In Smalltalk, the syntax is straightforward almost like natural language: anObject whatIsYourName. anObject doSomething. anObject do: #homework within: 60 minutes.

Here, we talk to anObject through three different messages. The first one, whatIsYourName, might only answer some text that helps us recognize a specific instance behind this rather generic “an object” label. Maybe it actually calls itself “a chess game.” This would greatly improve our understanding and how to further interact with it. The second one, doSomething, might modify something in the system, as maybe the object is “exhausted” afterward, meaning its own state has changed as well. The third one, do:within:, appends arguments, which are objects themselves. The first argument, #homework, is called (literal) symbol and

How to Tame an Unpredictable Emergence? Design Strategies for. . .

155

hopefully means something to the receiving object. The second argument, 60 min, is actually an example of another—albeit tiny—conversation: We ask the 60 to convert itself into a duration of 60 min. Both arguments hopefully mean something to the receiving object. All in all, programmers use this form of textual and direct communication for exploration in all kinds of programming tools. Names are the tangible things programmers need to acquire before they can talk to objects. In technical terms, these names are bindings in a workspace. There is always the name “self” bound to an object in tools that allow for looking into that object itself. From the outside, programmers can come up and use any name that feels memorable and fitting: | game pawn | game := ChessMorph new. pawn := game newPiece: 1 white: true.

Here, we declare, define, and use the names game and pawn for objects that represent a chess game. If we inspect the game to open a workspace (or inspector) that allows for looking into the game, we could then choose a more “intimate” way of talking to that object: | pawn | pawn := self newPiece: 1 white: true.

There can be all kinds of programming tools that bind the name self to any object of interest. Programmers must learn a tool’s scope, which is usually indicated through visual cues. Multiple tool windows can each represent a scope (and namespace) of their own. Such a multi-tool usage supports programmers in following multiple ideas (or hunches) side by side, which reduces cognitive load and thus expedites exploration. Still, finding a good name can be quite a challenge. If you do not know anything specific about an object, there are messages that all understand: • • • •

Tell me your kind (or class). Describe yourself in brief text. Show me who is referring to you. Show me who you refer to.

There are objects that have a visual representation that might be better suited than (abstract) text. In recent Squeak, most graphical objects are morphs (Maloney,2002), which can be rendered per request to collect visual snapshots for pixel-wise examination. Note that the act of message passing can be observed (i.e., via “debug it”) if a response is not satisfactory (i.e., via “print it” or “inspect it”). Through these simple means of direct communication, programmers can explore all objects in the system. Such an exploration might start with a visual handle or binding in a tool. Then, a simple “Tell me your kind” will reveal the object’s class. A class can be used to browse the object’s vocabulary. The correct vocabulary helps form appropriate messages. Message arguments can still be challenging to find.

156

M. Taeumel et al.

However, object communication is often diverse and deep, and so is the exploratory journey toward improving the system’s feeling of liveness.

3 Exceptions: Error, Notification, Halt Exceptions bypass normal object communication to implement out-of-the-ordinary behavior. When an exception gets signaled, it takes a shortcut along the active chain of messages in a process. That is, the exception ignores the objects waiting for their responses and instead looks for an exception handler to talk to. That handler can then try to resolve the issue, abort the communication, or let the user intervene. Both exception and handler are objects themselves, which means that other objects must request their assistance in the first place. Whenever a problematic conversation might occur, one object requests a handler to watch out. Whenever a problem in a conversation does occur, another object requests an exception to go find the next handler. This mechanism, which is complementary to normal messaging, can help ensure responsiveness in the system. The following example illustrates an ordinary payment process in a supermarket where a customer realizes that there is not enough cash left in the wallet: OutOfCashException class >> signal Customer >> lookIntoWallet | Customer >> doPayment | Cashier >> showReceipt | Cashier >> askForPayment v OutOfCashExceptionHandler >> watch Cashier >> treatCustomer ...

Whatever the full chain of messages may be at this moment—maybe the customer’s task is to buy milk and put it into the fridge at home—the top messages are the important ones. The cashier is prepared for this situation because it has happened before. Maybe the shopping cart can be parked until the customer returns with more cash later. The situation can be handled. The customer signals this exception as soon as the empty wallet is noticed. As a result, all communication down to askForPayment will be canceled. If the customer cannot otherwise proceed in this situation, the entire shopping process might need to be canceled. But what to put into the granola if not milk? In technical terms, objects can be in all kinds of unforeseen, troublesome situations, which then endanger the system’s responsiveness. There are three kinds of exceptions: errors, notifications, and halts. First, errors indicate that something unexpected has happened and that normal messaging cannot continue. An exception handler has to clean up if necessary; errors are usually not resumable. For example, ZeroDivision means that at some point the math has failed. While programmers can override and pretend that some other result is meaningful, this might lead to even more errors along the way. Again, exploration

How to Tame an Unpredictable Emergence? Design Strategies for. . .

157

and experimentation is encouraged, yet risky when facing errors. Second, notifications (or warnings) indicate that something less serious has happened and can be ignored. An exception handler is not necessary but may be beneficial to better understand the cause of unexpected behavior. Notifications can be resumed. For example, UnknownSelector triggers an interactive dialog where the user can choose from known messages (or selectors) to let the source code compile. Third, halts are an exploratory tool that lets the communication stop at a well-defined point so that users can investigate the circumstances of unexpected system behavior. When Halt is signaled, the process will stop, an interactive debugger will open, and programmers can freely explore the state of affected objects. While time stands still for that process, programmers can communicate to the objects of interest and even cause changes. A crucial instrument for reliable object communication (and a responsive system) is the error MessageNotUnderstood. In combination with an object’s response to the message doesNotUnderstand:, programmers can interactively explore and repair erroneous processes. If the receiver is wrong, they can try to figure out which object should actually receive that message. If the message is wrong, they can change it to something more appropriate. However, if both receiver and message are basically correct, the programmer can choose to only implement the behavior and thus extend the object. The process will then resume, and the system hopefully exhibit the desired behavior. That is, MessageNotUnderstood and doesNotUnderstand: will promote short feedback loops. Programmers can thus experiment and send what they think is correct to the objects at hand. There is a system-wide safety net in the form default exception handlers. That is, normal object communication does not have to use handlers at all, which can be tricky to foresee in most cases. The default effect of errors or warnings that are not handled is to signal an UnhandledError or UnhandledWarning exception. Now, the situation is codified as objects (or errors) again. If those special exceptions are not handled, they will simply stop the process and open a debugger. Then, programmers must intervene to fix the issue if possible and necessary. This safety net allows experiments such as evaluating 7/0 without being punished too seriously. Exploration is encouraged; short feedback loops can be established; an eventual emergence (of observable behavior) can be induced.

4 Stop and Step: Code Simulation Message sends are very fast6 thanks to the OpenSmalltalk VM7 (Miranda et al., 2018), which a promising baseline for message-based liveness and a responsive system. So, there is plenty of room for extra checks and safety nets that support

6 On a Microsoft Surface Pro 6 with an Intel Core i7-8650U on Windows 10 21H2 in Squeak 6.0, the VM can process about 190,000,000 sends per second. 7 The OpenSmalltalk VM, https://opensmalltalk.org/

158

M. Taeumel et al.

carefree experimentation. As mentioned above, unhandled errors will suspend the process they were signaled in. A debugger window will appear, representing a tangible handle for that suspended process. Programmers can then explore the process’ active methods8 and continue message passing step by step. They might even be able to go back in time, which is not a feature of vanilla Squeak, but available as a research project.9 Code simulation is the mechanism that allows for understanding and fixing erroneous object communication as it happens. Live (Ingalls, 2020). In contrast, code browsers merely show abstract pieces of text without a path to the running system. It is not obvious that any fast VM also offers a means to look into its execution details. For exploring communication patterns, however, we argue that it is necessary to “stop time” for selected processes. A suspended process can then be simulated, which means that message sending can continue with the help of other objects: the (method) contexts. Contexts provide access to the message receiver, active code, argument objects, and the object that is sending the current message (or active method). Like a puppeteer being in control from the outside, contexts can trigger a step, which lets the affected process “move forward.” Programmers can then take their time to observe effects as they unfold. In Squeak, there are three kinds of objects that reify system execution so that programmers can explore (i.e., witness) gradual or induce (i.e., cause) eventual emergence: • A Message reifies an attempt of a message send and is passed as argument to doesNotUnderstand: in the case of a MessageNotUnderstood error. • A Process reifies an active communication thread where objects are waiting “in line” to get responses from their neighbors. • A Context reifies a piece of source code in execution, which is the implementation behind the message an object understands. We will now use these means to reiterate the shopping experience as described above, but we will put extra cash into the wallet: | process customer | process := [ Customer new buyMilk ] newProcess. [ process suspendedContext selector = #lookIntoWallet ] whileFalse: [ process step ]. customer := process suspendedContext receiver. customer wallet add: 500. process resume.

8

Objects send messages. An object receiving a message will entail a message lookup, which means that a symbol will be matched to a piece of source code (or byte code). That code represents a method, which is an implementation of a message. That is why a suspended process has a stack of active methods, not messages. We use the term “chain of messages” to simplify this detail. 9 Trace Debugger, a back-in-time debugger for Squeak, https://github.com/hpi-swa-lab/ squeaktracedebugger

How to Tame an Unpredictable Emergence? Design Strategies for. . .

159

This simple example illustrates a very powerful idea: Domain objects and metaobjects can interact. From the outside in a workspace, we construct a process for buying milk. Then we simulate this process up to a point of interest, which is when the customer looks into the wallet. Then, we experiment by adding extra cash to the wallet. We can directly ask the process for the current context to then get access to the customer, which is the current receiver. Yes, we can also look into that customer’s wallet; we could take things out of it, but here we add the number object representing 500. Finally, we let the (shopping) process resume. Is it enough cash? We will find out as the system execution continues. If necessary, we can always suspend the process, whose handle (or binding) we still have in our workspace. Not all communication patterns can be understood through stepwise code simulation. For example, there can be two processes communicating through a shared resource, where one is adding and one is removing items. Suspending one process might block the other and vice versa. So, there should be tools that collect and present changes for groups of objects to reveal steady frames [(Hancock, 2003), p. 57]. What seems chaotic at the lowest level might oscillate in the middle or even be steady when observed from afar, like water drops escaping a garden hose as a constant stream. Considering object state, a data-driven perspective (Taeumel, 2020) might help programmers understand the flow of information. Considering object behavior, a higher-level debugger might help programmers understand larger patterns among processes. In sum, emergence does not only affect a few objects as it might go beyond sending a couple of messages in a single process. The VM is very fast and that thousands of messages can be exchanged within seconds. Liveness also implies a certain balance and harmony between all objects in the system.

5 Let the User Interrupt Squeak’s workspaces offer many ways to get the system stuck. Programmers experiment with small scripts to explore and poke around. Eventually, some attempt will have unexpected consequences and consume too much computing time, interfering with other processes that are waiting to be scheduled. Note that the system only “moves forward.” Objects are created, communicate with each other, get modified, change behavior, and so on. There is usually no going back, which we call an immutable past (Rein et al., 2018). What to do if there is no more user input possible? What if no code snippet can be typed and evaluated? How can the system be “restored” to a responsive state without losing information? Smalltalk systems have always had a simple answer for this situation: Let the user interrupt the interfering process that is consuming all the computing time at the moment. Even the most cautious programmers can get the system stuck by accident. For example, an experiment can become problematic by simply being mistaken about the (hidden) computing effort:

160

M. Taeumel et al.

| n factorial | n := 100000. factorial := 1. [ n > 0 ] whileTrue: [ factorial := factorial * n. n = n - 1 ]. factorial explore. "... will start an object explorer ..."

Here, the problem is not necessarily in the algorithm written by the programmer: The factorial of 100,000 took only about 5 s to compute. Instead, unpredictable effects can happen when other tools get involved: factorial explore. An object explorer, for example, will construct a textual representation of factorial, which might take some time for thousands of digits. Maybe there is more than one representation for that number object. Unoptimized tools might omit caching for costly inputs. Users cannot always anticipate this. The system seems to be stuck. But why exactly? This is generally not obvious. However, even a few seconds can be enough. Users can quickly get frustrated when confronted with an unresponsive system, let alone how it affects their desire for a feeling of liveness. There is a dedicated keyboard shortcut to interrupt the active process. On macOS, for example, programmers can hit [CMD] + [.], and a debugger will appear, representing the interrupted process. They can then explore and repair object communication like they would for unhandled exceptions. This reliable gesture should be invokable through a physical key, as the (virtual) system could be in any unclear state where pixels cannot be trusted anymore. Such an interrupt-key press then simply tells the active process to suspend: | process | process := Processor activeProcess. process suspend.

Note that responsiveness will also be impaired if processes slow down the system. If basic interaction works—even if uncomfortable—programmers can use the process explorer to browse and suspend (or terminate) selected processes. Note that if tools work, programmers can also write scripts in workspaces to do the same thing. For example, they can query known process objects via Process allInstances explore. In general, given the class (object) of a do- main concept, all domain objects can be fetched this way. There is only a single object space to explore and modify, where tools and applications exist side by side. There are processes that must be running to keep the system responsive. They should not be interrupted as user input or graphics output would stop working, which are the basic means for responsiveness. A simple UI framework might look like this: | uiProcess | uiProcess := [ [ InputEventSystem handleInputEvents. ScriptingSystem evaluateScripts. GraphicsSystem drawGraphics. ] repeat ] newProcess. uiProcess resume.

How to Tame an Unpredictable Emergence? Design Strategies for. . .

161

An endless loop of input handling, script processing, and graphics drawing makes the system “look alive” and responsive. Provided that workspaces inject work into evaluateScripts, programmers might inadvertently block (or slow down) this loop, which would affect input and drawing. Consequently, the system must ensure that another uiProcess will be started as soon as the current one gets interrupted. And somehow, that interfering script should be excluded from being evaluated a second time. Also, if an interrupted uiProcess is resumed again, the current one should be terminated to keep object communication in balance and harmony—too many cooks spoil the broth. The correct timing of a user interrupt can be crucial for memory consumption. That is, too many objects might be created in a short amount of time. While the system looks unresponsive, object memory might “overflow.” Fortunately, the OpenSmalltalk VM maps large process stacks to heap memory, which means that the usually limited stack memory is guarded so that users have more time to press that interrupt key. Still, such a memory leak can slow down the system even after the interfering process gets suspended. While obsolete objects get cleaned up all the time, incremental garbage collection can be essential to ensure responsiveness in a system with a large object memory. That is, the VM should always clean up “garbage” piece by piece without the user noticing.

6 Recursive Emergency The objects that make up the debugger might stop working. That is, unexpected behavior (or erroneous communication) is not an issue limited to applications but extends to all kinds of things in the system, which includes programming tools. Recall that an unhandled error will suspend the process it was signaled in, which will then result in a debugger for this process. That debugger might as well signal another error when, for example, it tries to show itself on screen but somehow cannot. What happens when an unhandled error triggers another error that cannot be handled? We call this situation a recursive error. Recursion is the concept where a message is sent again while its response is being prepared. A recursive algorithm ends when a certain condition holds. For example, the Fibonacci sequence can be implemented as a method fib: that takes an integer argument n and then answers the result of (self fib: n - 2) + (self fib: n - 1); the results for n = 1 and n = 2 are hard-coded to finish this computation. Now, how is a recursive error handled? Is there also an exit condition? But what to do when this condition holds? A possible approach to handle recursive errors is to provide several distinctive UI frameworks as fallback (Taeumel & Hirschfeld, 2016). Each framework has programming tools and thus means to modify the code base and all other kinds of objects. If one fails, the other one takes over. We assume that frameworks share as little source code as possible to increase the chances that an error does not carry over. In vanilla Squeak,

162

M. Taeumel et al.

there is Morphic by default and MVC as fallback. Both have debuggers, code browsers, and object explorers to the rescue. While both use common means, such as the code compiler, MVC has a much simpler architecture and is less likely to be extended at this time. Consequently, if an experiment in Morphic fails, tools in MVC can take over. There can be programming tools even simpler than MVC, which we demonstrated through the keyboard-only SqueakShell (Taeumel & Hirschfeld, 2016). The system detects a recursive error by flagging the erroneous process, which indicates that a debugger is about to open. If the debugger fails to open, that recursion flag will be detected on the second attempt and then trigger a transfer to another framework. Since every “thing” is an object in Squeak, frameworks are as well. They are represented as projects (Taeumel & Hirschfeld, 2016), which are a way to organize work and describe the basic means of responsiveness. For example, MVC uses a new process per tool for handling user interaction and graphics updates; Morphic (re-)uses a single UI process for all tools. When transferring the control from one project (or framework) to another, the erroneous process will be suspended immediately to not cause further harm. Programmers can then use the tools available in that other project to restore responsiveness as desired. As switching between frameworks can be tedious, we assume that there is only one preferred one where liveness matters. If all else fails, there will be the emergency evaluator. With only a few lines of code, the system can provide a very primitive read-evaluate-print loop. Programmers can type expressions like they do in workspaces. For simplicity, there is no support for multiple lines, text selection, or bindings. What must work at this point is keyboard input, text rendering, and the code compiler. Here is the basic idea (without output) of the emergency evaluator in pseudo code: | line char | [ line := ". [ [ Sensor keyboardPressed ] whileFalse. (char := Sensor keyboard) = Character cr] whileFalse: [ line := line, char asString ]. line = ‘exit’ ] whileFalse: [ line = ‘revert’ ifTrue: [ System revertLastChange ] ifFalse: [ Compiler evaluate: line ] ].

The most important feature is to revert the latest code change. Chances are that erroneous object behavior is the culprit. Since code (or behavioral) changes are versioned, reverting those is easy. However, state changes are not versioned. So if an object has unfortunate state, programmers must figure out a way to access that object and restore it manually. Recall that allInstances is a very powerful query mechanism to access any object provided that you know the name of its class.

How to Tame an Unpredictable Emergence? Design Strategies for. . .

163

7 Reliable Watchdogs One process can watch over another process. That is, programmers can design processes that monitor responsiveness in the system and intervene if necessary. Recall that Squeak’s processes schedule cooperatively at the same priority level and preemptively between priorities. Thus, for example, a process with priority 75 could wake up every 250 ms to then monitor any process below or equal to 75. Processes under inspection will not change on their own but “stand still” until the watchdog yields control (to other watchdogs) or suspends for another 250 ms. Recall that one process can easily access other processes in the system. So, our watchdog could log current receivers, depth of message stack, or other information that might be of interest. Note that the power of watchdogs is not constrained to monitoring. All objects are free to send any message to any other object they have access to. This freedom entails all kinds of possible side effects such as process termination, stack manipulation, and actually fiddling around with domain-specific objects in your application. A common example for a watchdog is the implementation of time-outs for test runs. We will now describe four useful watchdogs, which each try to perform their analysis (and intervention) as quickly as possible to not slow down the system. Their wake-up time (or frequency) denotes the precision of their work. Consequently, watchdogs that need more time should wake up less often to not further jeopardize responsiveness in the system. Call Tracing and Profiling Squeak offers a sampling-based tool to trace object communication, which is called message tally. With a high frequency, a watchdog collects the stack of the tallied process and approximates a call tree from all samples. Programmers can then explore the most prominent communication patterns in the tally. The overall execution time can be used to estimate the duration of selected message sends. Note that sampling-based tracing trades performance for accuracy. Brief conversations between objects will likely not be traced as process scheduling does not allow for such fast sampling rates. Unanticipated Progress Indication There are cases where programmers are willing to wait in front of an unresponsive system so that maximum computing time is used to finish a workload as quickly as possible. However, they do wish for progress indication to plan their next steps. If known upfront, such indication can be requested: workItems do: [:workItem | workItem process ] displayingProgress: [:workItem | workItem label].

Unfortunately, there can be numerous places in the systems where workload can spike unexpectedly, meaning that there is likely no progress indication in place. Also, a few long-running takes need a different treatment than thousands of short-

164

M. Taeumel et al.

running tasks. That is, any overhead in progress computation and display must be minimized. After all, programmers expect work to finish as fast as possible. Now, watchdogs can be used to analyze the process stack and look for such loops where work items are enumerated: | workBlock start stop index | workBlock := [:step | ... ]. index := start. [ stop >= index ] whileTrue: [ workBlock value: index. index := index + 1 ].

Given that a process exposes its current stack, a watchdog can access bindings such as start and stop and step to reveal progress. Sampling can codify “tasks” through methods that remain active after repeated observations. Programmers are not required to prepare such progress analysis for specific cases. The watchdog can use generic patterns to find tasks. As soon as it finds a task, it can tell the user about task progress. This indication is not anticipated but welcomed. Infinite Recursion Detection Watchdogs can find patterns that indicate infinite recursion in a process, which might indicate potential out-of-memory issues. The user should be informed that the process in question will be suspended for inspection. Yet, it might be challenging to distinguish a deeply nested algorithm that uses backtracking (or dynamic programming) from actual infinite recursion. As falsepositives happen, users should always be in charge of deciding whether or not to abort a computation. Too many false-positives might disturb the programming experience up to a point where the watchdog will be disabled for good. Consequently, such background assistance should be configured to consider domain-specific workload and personal preferences. Still, generic patterns are a good starting point. A Time-Slicing Scheduler Fairness in Squeak’s process scheduling depends on the cooperation of processes running at the same priority level. However, a programmer’s experiments might not be so cooperative after all: [ anObject doExpensiveWork ] fork. "... in the background?"

Here, the programmer might think that performing an expensive task in another process keeps the system responsive. Yet, for example, in Morphic, all workspace scripts are evaluated in the same UI process. This means that the programmer either explicitly puts that task in a process with lower priority or slices it up with an occasional Processor yield in between. Now, a time-slicing scheduler could also help with this issue, introducing preemption within a priority level as well. Each process could just get some milliseconds before going back to the end of the line. Similar to progress indication, a watchdog can reconfigure process objects as required. Fairness might be such a requirement to ensure responsiveness in the system.

How to Tame an Unpredictable Emergence? Design Strategies for. . .

165

8 Conclusion Admittedly, Squeak/Smalltalk is not the most convenient live-programming system. Its building blocks, objects and messaging, are expressive and powerful. The system is for generic purposes by design; there are little constraints to object communication. An uninformed programmer might get frustrated when some experiment gets the system stuck. We argued that programmers must know their tools and learn about the means for exploration (Taeumel et al., 2022; Taeumel et al., 2021) to get the system “unstuck” again. Responsiveness is the key to observing gradual and inducing eventual emergence. Processes are the means for (de)composing elaborate communication patterns. Each process represents an active chain of messages, a topic, a discussion, or a relevant conversation to be understood. You can type and evaluate snippets of source code in workspaces everywhere in the system. Chances are that the tool in front of you—maybe another debugger—provides bindings for objects of interest. Use an object’s name and ask away; learn more about it. It might tell you why some effect is or is not showing up. Some messages will even help you fix an erroneous communication. There are tool-building frameworks that provide specific constraints for guiding adaptation and emergence. For example, Morphic (Maloney, 2002) offers the message step to codify recurrent actions for all graphical objects. Programmers (should) know that changing code in a step method will be executed at some point. For another example, Vivide (Taeumel, 2020) describes exploration tools as rules of data transformation and visual mappings. Programmers can then expect and rely on certain effects when modifying these rules. We think that frameworks on top of objects and messaging represent clear learning objectives. Once understood, programmers can express their goals in framework-specific actions to then rely on immediately observable effects. The Squeak/Smalltalk live-programming system is thus capable of providing a predictable emergence for application-specific domains. Acknowledgments Many thanks go to Dr. Sharon Nemeth for her editorial support. We gratefully acknowledge the financial support of the HPI Research School on Service-oriented Systems Engineering (www.hpi.de/en/research/research-schools) and the Hasso Plattner Design Thinking Research Program (www.hpi.de/en/dtrp).

References Gabriel, R. P. (2014). I throw itching powder at tulips. In: Proceedings of the 2014 ACM International Symposium on New Ideas, New Paradigms, and Reflections on Programming and Software, pp. 301–319. ACM. https://doi.org/10.1145/2661136.2661155. Hancock, C. M. (2003) Real-time programming and the big ideas of computational literacy. Ph.D. thesis, Massachusetts Institute of Technology. Hutchins, E. L., Hollan, J. D., & Norman, D. A. (1985). Direct manipulation interfaces. HumanComputer Interaction, 1(4), 311–338. https://doi.org/10.1207/s15327051hci0104_2

166

M. Taeumel et al.

Ingalls, D. H. H. (2020): The evolution of smalltalk: From smalltalk-72 through squeak. In: Proceedings of the 4th ACM SIGLAN History of Programming Languages Conference (HOPL IV), pp. 1–101. ACM. https://doi.org/10.1145/3386335. Maloney, J. H. (2002). An introduction to Morphic: The squeak user Interface framework, chap. 2 (pp. 39–67). Prentice Hall. Meyer, B. (1998). Object-oriented software construction (2nd ed.). Prentice Hall. Miranda, E., Béra, C., Boix, E. G., Ingalls, D. H. H. (2018). Two decades of smalltalk vm development: Live vm development through simulation tools. In: Proceedings of the 10th ACM SIGPLAN International Workshop on Virtual Machines and Intermediate Languages, pp. 57–66. ACM. https://doi.org/10.1145/3281287.3281295. Rein, P., Lehmann, S., Mattis, T., Hirschfeld, R. (2016). How live are live programming systems?: Benchmarking the response times of live programming environments. In: Proceedings of the Programming Experience 2016 (PX/16) Workshop, pp. 1–8. ACM. DOI https://doi.org/10. 1145/2984380.2984381. Rein, P., Ramson, S., Lincke, J., Hirschfeld, R., & Pape, T. (2018). Exploratory and live, programming and coding. The Art, Science, and Engineering of Programming, 3(1), 1:1–1:33. https:// doi.org/10.22152/programming-journal.org/2019/3/1 Sheil, B. (1998). Datamation®: Power tools for programmers, chap. 33 (pp. 573–580). Morgan Kaufmann. https://doi.org/10.1016/B978-0-934613-12-5.50048-3 Shneiderman, B., & Plaisant, C. (2010). Designing the user interface: Strategies for effective human-computer interaction (5th ed.). Addison-Wesley. Taeumel, M. (2020). Data-driven tool construction in exploratory programming environments. Ph. D. thesis, University of Potsdam, Digital Engineering Faculty, Hasso Plattner Institute. https:// doi.org/10.25932/publishup-44428. Taeumel, M., Hirschfeld, R. (2016). Evolving user interfaces from within self-supporting programming environments: Exploring the project concept of squeak/smalltalk to bootstrap uis. In: Proceedings of the Programming Experience 2016 (PX/16) Workshop, pp. 43–59. ACM, https://doi.org/10.1145/2984380.2984386. Taeumel, M., Lincke, J., Rein, P., Hirschfeld, R. (2022). A pattern language of an exploratory programming workspace. In: Design thinking research: Achieving real innovation, pp. 111–145. Springer https://doi.org/10.1007/978-3-031-09297-8_7. Taeumel, M., Rein, P., Hirschfeld, R. (2021). Toward patterns of exploratory programming practice. In: Design Thinking Research: Translation, Prototyping, and Measurement, pp. 127–150. Springer. https://doi.org/10.1007/978-3-030-76324-4_7. Tanimoto, S. L. (2013). A perspective on the evolution of live programming. In: 2013 1st International Workshop on Live Programming (LIVE), pp. 31–34. IEEE https://doi.org/10. 1109/LIVE.2013.6617346. Trenouth, J. (1991). A survey of exploratory software development. The Computer Journal, 34(2), 153–163. https://doi.org/10.1093/comjnl/34.2.153 Ungar, D., Lieberman, H., & Fry, C. (1997). Debugging and the experience of immediacy. Communications of the ACM, 40(4), 38–43. https://doi.org/10.1145/248448.248457

Part 3

Enhancement through Design Thinking

What Is Design Thinking? Jan Auernhammer and Bernard Roth

Abstract This chapter outlines and discusses different perspectives on design thinking. It provides a schema that illustrates three different understandings of design thinking: (1) methodology, (2) thinking of designers, and (3) practice-based design thinking (embodied thinking). We seek to clarify some of the substantial differences and nuances of alternative understandings of design thinking. Specifically, this chapter discusses significant nuances between structuralism (i.e., information-processing), formal logic (i.e., abductive reasoning), and Gestalt psychology (i.e., humanistic psychology). These differences are based on schools of thought that emerged in early experimental psychology and that then informed numerous design scholars. These perspectives have also informed various practicebased aspects of design thinking, such as variation-selection, meaning-making, and comprehensive design. By outlining these alternative understandings, this chapter presents answers to the question “What is design thinking?” from different perspectives. This discussion and outlined schema encourage researchers and practitioners to articulate their perspective when referring to “design thinking.”

1 Introduction Over the last decades, “design thinking” has attained global attention, and today many organizations and educational institutions employ a form of design thinking. However, the term design thinking is used as an umbrella term incorporating many different perspectives without a shared understanding of what design thinking is, similar to the term “design” (Auernhammer & Ford, 2022; Auernhammer & Roth, 2021). This chapter outlines and discusses different perspectives on design thinking J. Auernhammer (✉) Mechanical Engineering Design Group, Stanford University, Stanford, CA, USA University of Technology Sydney, Sydney, NSW, Australia e-mail: [email protected] B. Roth Hasso Plattner Institute of Design (dschool), Stanford University, Stanford, CA, USA © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_9

169

170

J. Auernhammer and B. Roth

Table 1 Early perspectives on design thinking Design thinking as Four areas of innovation The ability to separate forms and shape matter into an organized fashion Concept formation

Reference Arnold (1959) Alexander (1962)a

A systematic methodology

Archer (1965)

Visual (visuospatial) thinking (abstract to concrete thinking languages) Information-processing of human problem-solving Information-processing of human problem-solving A three-step process with various methods

McKim (1972, circa 1968) Simon (1969)a Eastman (1970)a Jones (1970)

Structural and productive thinking

Lawson (1972, 1980)

A universal, imaginative, and argumentative activity dependent on social context Abductive reasoning

Rittel (1987), and Rittel and Webber (1973))a March (1976)a

Sense- or meaning-making

Krippendorff (1989); Krippendorff & Butter, 1984)a Rowe (1987)

Systematic processes of designing Variation-selection learning model through experience a

Schön (1963)a

Vincenti (1990)a

Based on Various psychological perspectives Various psychological studies, including Piaget (1951) and Arnheim (1954) Various psychological perspectives, including Dewey (1938), Wittgenstein (1953), and Wertheimer (1945) Pólya’s (1957) plausible reasoning in mathematical education Gestalt psychology, including Wertheimer (1945), Arnheim (1954, 1965, 1969), and Perls et al. (1951) Structuralist psychology, including Selz (1922) and de Groot (1965) Structuralist psychology, including Selz (1922) and de Groot (1965) Various influences, such as useful techniques, such as Gordon (1961), Osborn (1957), and Asimow (1962) Various psychological perspectives, including Wertheimer (1945), Bartlett (1958), and Bruner et al. (1956) An underlying gestalt perspective without expressing it Abductive thinking by Peirce (1871, 1923) and philosophy of science by Popper (1961a, 1961b, 1962) Gestalt considerations

Various psychological perspectives, including Gestalt and structuralism (Humphrey, 1963) Epistemological ideas of Campbell (1987) and their initial application by Constant II (1980)

Did not use the term “design thinking”

starting from its early beginnings in the 1950s and 1960s (as outlined in Table 1). By doing so, we clarify some significant differences between different understandings of design thinking. An essential contribution of this chapter is a schema (as shown in Fig. 1) of different meanings of design thinking. Design thinking is understood as (1) a methodology, (2) the thinking of designers, and (3) practice-based design thinking (i.e., embodied thinking). These different understandings have emerged

What Is Design Thinking?

171

Fig. 1 Different perspectives on design thinking

from diverse schools of thought from experimental psychology (Auernhammer & Roth, 2022; Rowe, 1987). For example, the difference in thinking of designers from a structuralist and a Gestalt perspective is often distinguished as problem-solving versus sensemaking, e.g., Krippendorff (1989). We discuss these substantial nuances of the different meanings of design thinking in detail.

2 Early Perspectives on Design Thinking The term “design thinking” emerged with various understandings in the 1950s and 1960s. For example, Arnold (1959), a psychologist and engineer, described “design thinking” in terms of four areas one should think about when creating innovative products. He and others developed an entire philosophy of design based on various psychological insights (Arnold, 2016; Auernhammer et al., 2022; Clancey, forthcoming). Archer (1965), a mechanical engineer, expressed “design thinking” as systematic designing. He emphasized the shift from the sculptural to the technological, which requires incorporating knowledge from different fields (Archer, 1965). At a conference “where design educators [. . .] gathered to discuss the theme ‘Toward A Science of Design’,” (circa McKim et al., 1967), an engineer and industrial designer, discussed “the effect of language specialization upon design thinking” and expressed the importance of many different languages along the spectrum from the abstract (e.g., visual analogs) to the concrete (e.g., the real thing). Around the same time, John Christopher Jones (1970), an engineer, expressed making “design thinking” transparent as a way to allow others from diverse knowledge domains to participate. In his Ph.D. research at Aston University, architect and psychologist Bryan Lawson (1972) describes “design thinking” in architectural problem-solving as a combination of structural thinking (e.g., Bruner et al., 1956) and productive thinking (e.g., Wertheimer, 1945). Lawson (1980) further elaborates this perspective of thinking in design in How Designers Think. Several years later, Peter Rowe (1987), an architect, discusses in Design Thinking early theoretical positions from psychology applied in architecture and urban planning, such as the structuralist perspective by Oswald Külpe, Narziß Ach, and Karl

172

J. Auernhammer and B. Roth

Bühler, as discussed by Humphrey (1963), and the Gestalt perspective, as developed by Köhler (1929), Koffka (1935), and Wertheimer (1945). Table 1 outlines these several early perspectives on design thinking. Various other design scholars and practitioners have discussed the formation of concepts, intuitive design processes, and logic in design without using the term “design thinking” (e.g., Alexander, 1962; Eastman, 1970; March, 1976; Schön, 1963). For example, architect Christopher Alexander (1962) investigated how schemas emerge creatively in children. Working with psychologists Jerome S. Bruner and George A. Miller, Alexander (1962, p. 207) investigated the creative ability to shape thought and matter in an organized fashion through children’s drawings. He shows that creative schematic organization in children, as expressed in their drawings, can be “understood in terms of the child’s developing ability to force the forms apart from one another” (Alexander, 1962). Similarly, the architect and industrial consultant Donald Schön (1963) describes in Displacement of Concepts the formation of new concepts influenced by the work, such as Dewey (1938), Wittgenstein (1953), and Wertheimer (1945). The assessments of recoded design and discovery sessions in the Design Invention Group at Arthur D. Little informed Schön’s (1963) work. Parallel assessments by Bill Gordon (1961) and George Prince (1968) from the same sessions resulted in Synectics. Another contribution is by Herbert Simon (1969), a political scientist, and Charles M. Eastman (1970), an architect, who investigated and described problem-solving and intuitive design processes based on the information-processing model, promoted by Newell and Simon (1956, 1965), among others, which in turn is based on work by psychologists Otto Selz (1922) and his student Adriaan de Groot (1965). Rittel (Rittel, 1987; Rittel & Webber, 1973), a mathematician and design theorist, outlined the limitations of the rationalistic doctrine in design. With increased understanding, the problem changes as a whole and, therefore, can never be fully understood. Rittel does not cite Gestalt psychologists, such as Wertheimer (1945) or Lewin (1946); however, his work strongly resembles the same ideas. Rittel (1987) describes the reasoning in design as a universal, imaginative, and dialectic activity that is not unique or belongs solely to “the designers.” Another reasoning in design was proposed by architect Lionel J. March (1976), who describes in The Architecture of Form that the logic of design is based on abductive reasoning, as developed by Peirce (1871, 1923) and discussed in the philosophy of science by Popper (1961a, 1961b, 1962). Klaus Krippendorff (1968), with a similar perspective as Horst Rittel’s, argues for the importance of the symbolic environment in design. Krippendorff (Krippendorff, 1961, 1984, 2006; Krippendorff & Butter, 1984), a designer and communication scholar, outlines the designer’s need to make sense of things in the context of people, contrasting the product designer (Gestalter) with the constructor (Konstrukteur). His view is based on Gestalt considerations (Krippendorff, 1989). Based on these various perspectives, as summarized in Table 1, many design scholars have built on these perspectives (e.g., Brown, 2009; Dym et al., 2005; Muller, 1989; Roozenburg, 1993). Today, these diverse perspectives, as outlined in Table 1, are all referred to as “design thinking,” with often little consideration of the nuances and substantial differences between these perspectives and understandings.

What Is Design Thinking?

173

The following sections discuss these substantial nuances in more detail in order to highlight the different perspectives on design thinking.

3 Design Thinking as . . . Various scholars have discussed different notions of design thinking. For example, Jones (1970) distinguished between creative (i.e., black box), rational (i.e., glass box), and self-organizing design thinking. Others distinguish between the psychological school of thought, such as structuralism (i.e., Würzburg School) and the Gestalt Movement (i.e., holistic/humanistic) (Auernhammer & Roth, 2022; Rowe, 1987). A first design thinking perspective that can be distinct from others is the externalization of thinking as a (1) recipe or methodology. Such a perspective is independent of people and emerged in the Design Methods Movement (Alexander, 1964; Archer, 1965; Gregory, 1966; Jones & Thornley, 1963). A second perspective is the (2) thinking of designers, which occurs within the mind of people, e.g., informationprocessing, productive reasoning, or Gestalt creation (Eastman, 1970; March, 1976; McKim, circa McKim, 1968; Wertheimer, 1945). This perspective incorporates several nuances distinguishing different types of design thinking based on different psychological schools of thought. The last perspective discussed in this chapter is (3) practice-based design thinking (i.e., embodied thinking). This perspective emphasizes that psychic (mental) functions cannot be separated from somatic (bodily) ones (McKim, 1972). Thinking is not independent of embodied activities, such as gesturing (Goldin-Meadow, 2003; Tang & Leifer, 1988). These design practices are along the spectrum between engineering practices, often associated with the sciences, and sensemaking, often associated with the arts (Krippendorff, 1989; Papanek, 1973; Vincenti, 1990). However, a practice-based comprehensive design (thinking) provides a third perspective, which emphasizes the importance of designing for the whole person and environment (i.e., for physical, intellectual, and profound needs) (Arnold, 1959; McKim, 1959). Figure 1 provides a schema of the different perspectives on design thinking.

3.1

Design Thinking as a Methodology

Design thinking as a methodology (i.e., systematic method) emerged from research and developments on “organized” creativity (e.g., Arnold, 1954, 1959; Guilford, 1950; Osborn, 1957; Pólya, 1945; Zwicky, 1948), which informed designers and resulted in the Design Methods Movement in the 1960s (Archer, 1965; Gregory, 1966; Jones, 1959; Jones & Thornley, 1963). One of the first designers to express “design thinking” as a systematic method was Leonard Bruce Archer (1965) from

174

J. Auernhammer and B. Roth

the Royal College of Art in London. Archer (1965) emphasized a rational approach to design and described design thinking as follows: In the face of this situation there has been a world-wide shift in emphasis from the sculptural to the technological. Ways have had to be found to incorporate knowledge of ergonomics, cybernetics, marketing, and management science into design thinking. As with most technology, there has been a trend towards the adoption of a systems approach as distinct from an artifact approach. [emphasis in bold added]

Archer (1965) provides various checklists and arrow diagrams that bring together knowledge from different disciplines into a structure representing the design act. Underlying Archer’s (1965) systematic method is George Pólya’s (1945) work on heuristics and plausible reasoning in mathematical education. Pólya (1945) outlines a set of steps: (1) Understand the problem. (2) Make a plan. (3) Carry out the plan. (4) Look back on your work. He also outlines various heuristic strategies, including those titled Analogy, Draw a Picture, Generalization, Subconscious Work, Test by Dimension, Variation of the Problem, and Working Backwards. Pólya (1945, p. 122) developed his plausible reasoning on the work of philosopher Ernst Mach, mathematician Jacques Hadamard and F. Krauss, and psychologists William James, Wolfgang Köhler, and Karl Duncker. For example, Duncker (1935, 1945) described in On Problem Solving (Psychologie des Produktiven Denkens) the discovery of various strategies to determine new solution means, identified in his think-aloud protocol research. The various heuristic strategies, identified in psychological research, expanded into design methods.

3.1.1

Design Methods (Heuristic Strategies)

Various creative, useful techniques in engineering, design, and industrial consultancy emerged based on the findings and insights from experiential psychology research on heuristic and creative strategies (e.g., Arnold, 1962b; Crawford, 1954; Gordon, 1961; Osborn, 1957; Zwicky, 1948). These heuristic and creative strategies further developed into design methods. John Christopher Jones (1970) wrote one of the first comprehensive methods books, Design Methods: Seeds of Human Futures. He argues that designers needed new methods to deal with the complexity inherent in problems and wanted to bring ergonomics and other disciplines to industrial design (Jones, 1970). Design methods are the “attempt to isolate the essence of designing and to write it down as a standard method, or recipe, that can be relied upon in all situations” (Jones, 1970). Jones (1970, p. 45) expressed the value of the methodological approach as follows: A major advantage of bringing design thinking into the open is that other people, such as users, can see what is going on and contribute to it information and insights that are outside the designer’s knowledge and experience. [emphasis in bold added]

To make design thinking transparent, Jones (1970) states three stages: (1) divergence (i.e., extending the boundary of a design situation), (2) transformation (i.e., creative patternmaking), and (3) convergence (i.e., reducing uncertainties progressively).

What Is Design Thinking?

175

Each phase contains several heuristic strategies and useful techniques outlined as design methods. In the 1980s, the systematic method developed further in Europe in engineering design (Hubka & Eder, 1987; Pahl et al., 1996). Similar to Jones (1970) attempting to bring design thinking into the open for others to contribute, David Kelley Design (DKD) developed a methodology handbook in 1989 “to provide a living record of the process to which any DKD designer can contribute” (Curtis, 1990). The essence of this methodology included the following activities: Brainstorming, Quick prototypes, Size the product, Get everybody in the same room, Look wide, Get narrower and narrower, Iterate, Smart people, User chooser, Working with the client, Working with vendors, and Working with the industrial design firm (Curtis, 1990). This methodology then developed further into IDEO’s Designkit (Ideo, 2015). The DKD handbook included a “Floss talk” by David Kelley from 1989 that included concepts such as “Fail sometimes, be Left-handed, get Out there, be Sloppy and be Stupid, and Off the wall? Maybe” (Curtis, 1990). These concepts became the various mindsets in IDEO’s (Ideo, 2015) Designkit. The term “mindset” emerged with the introduction of the “growth mindset” (Dweck, 2006), which has been part of the human potential movement that was deeply ingrained in Stanford’s Design Division (Arnold, 1959; Fadiman, 1986; McKim et al., 1967). In general, design processes and methods are abstractions from people’s thinking, psychological principles, and heuristic strategies as identified in psychological research. Many design scholars outlined their own version of processes that are logical sequencing of heuristic strategies. For example, the double diamond visualization is based on Guilford (1950) productive thinking of divergent and convergent thinking. As outlined by Jansson and Smith (1991), design fixation is a recontextualization of Duncker’s (1935, 1945) functional fixedness. Similarly, reframing is the heuristic strategy that allows overcoming the block of “functional fixedness” by rephrasing the problem, identifying problem variations, and determining new purpose-means relationships (Duncker, 1935, 1945; Pólya, 1945; Selz, 1922).

3.1.2

Today’s Design Thinking as a Methodology

Today, there are many recipes or “cookbooks” on design thinking that include stepby-step processes and related tools and methods that all resemble the same process of exploring problem situations, creating solutions, evaluating the problem-solution fit, and implementing (Dubberly, 2005; Ideo, 2015; Kumar, 2012; Liedtka & Ogilvie, 2011). Each phase has particular heuristic strategies and associated techniques. For example, Dorst (2015) outlines in Frame Creation several heuristic strategies, such as understanding the problem (Archeology) and conflict analysis (Paradoxes), as identified by Duncker (1935, 1945), in a logical sequence. Such recipes are generally employed in short design thinking or participatory design workshops/sprints. However, this recipe approach is not without criticism for various reasons.

176

3.1.3

J. Auernhammer and B. Roth

Critique of the Methodology Perspective

Various scholars have questioned and criticized design thinking as a methodology (Alexander, 1971a, 1971b; Jones, 1977, 1991; McKim, 1972, 1980c). Jones (1977) fundamentally changed his mind on the notion of design methods and criticized the methodology perspective as “[he] dislike[ed] the machine language, the behaviourism, [and] the continual attempt to fix the whole of life into a logical framework.” Jones (1991) repeats his criticisms in Designing Designing as follows: We sought to be open minded, to make design processes that would be more sensitive to life than were the professional practices of the time. But the result was rigidity: a fixing of aims and methods to produce designs that everyone now feels to be insensitive to human needs. Another result was that design methods became more theoretical and many of those drawn to the subject turned it into the academic study of methods (methodology) instead of trying to design things better. The language used to describe designing, and to describe the aims and purposes of things designed, became more and more abstract. The words lost touch with how it feels to be a designer and how it feels to inhabit the systems being designed.

In a similar manner, Christopher Alexander (1971a) expresses the absurdity of the idea of design methods in the Preface to the Paperback edition of the Notes to the Synthesis of Form as follows: [. . .] so many readers have focused on the method which leads to the creation of the diagrams, not on the diagrams themselves, and have even made a cult of following this method. Indeed, since the book was published, a whole academic field has grown up around the idea of ‘design methods’ - and I have been hailed as one of the leading exponents of these so-called design methods. I am very sorry that this has happened, and want to state, publicly, that I reject the whole idea of design methods as a subject of study, since I think it is absurd to separate the study of designing from the practice of design. In fact, people who study design methods without also practicing design are almost always frustrated designers who have no sap in them, who have lost, or never had, the urge to shape things. Such a person will never be able to say anything sensible about ‘how’ to shape things either.

In response to a question about future work in design methodology, Alexander (1971b) replied, “I would say forget it, forget the whole thing. Period.” Similarly, McKim (1972, p. 26) criticized the step-by-step methodological approach as a “cookbook pattern,” as follows: There is no single way to experience the material presented in this book. Indeed, a step-bystep procedure is not recommended. Visual thinking in relation to a challenging problem does not proceed in cookbook fashion (for example, (1) recenter your viewpoint, (2) add a dab of inner imagery, (3) now make six sketches, (4) cook in the brain overnight.) Visual thinking that is equipped with a rich repertoire of visual responses moves far more flexibly. Consequently, learning to think visually should occur in a flexible way.

In 1980, McKim (1980c) repeated his criticisms with stronger phrasing, that “[t] he strategy approach presented in this book is emphatically not intended to be used as a step-by-step problem-solving method” [emphasis added in italic]. In his writings, he highlighted that seeing, imagining, and visually expressing, including prototyping, are not independent to the embodied activities of people and their abilities, values, needs, and emotions (McKim, 1972, 1980a, 1980c, circa McKim,

What Is Design Thinking?

177

1968). The emphasis is not on following a process but on developing human capacities, such as awareness, attitudes, and abilities (Auernhammer & Roth, 2021). Similarly, Arnold (1959) emphasized that questioning, observing, and associating are not a step-by-step process but attitudes of the inventor that “[. . .] should be going on all the time, simultaneously or in almost any kind of combination or sequence.” The criticism highlights that design thinking as a methodology separates methods from actual practice (Alexander, 1971a; McKim, 1972). It reduces the effortful, complex, and time-consuming act of invention into a highly abstract set of steps and frameworks, ignoring essential human abilities and motivation (Arnold, 1959; Jones, 1991). This perspective dehumanizes the designers’ sensitivity and ability, such as cognitive flexibility, into a process and methods (Jay, 1962; McKim, 1972, 1980c). A universal recipe is not applicable in every context, as each situational circumstance requires different responses (Auernhammer & Leifer, 2023; Auernhammer & Roth, 2022; Koffka, 1927; McKim, 1980c). Methodology ignores the fact that not the method, such as brainstorming, produces creative ideas. The psychological safety and freedom enacted by participants allow open conversation that is conducive to creativity (Arnold, 1959; Arnold et al., 1960; Rogers, 1954). In general, design thinking as a methodology is a recipe or cookbook representing the essence of designing in an abstract process coupled with various methods (e.g., heuristic strategies and useful techniques). The benefit of making design thinking open and transparent is that it allows various persons from different backgrounds to contribute with their knowledge. However, it dehumanizes design into abstract processes separate from situational circumstances and human qualities, such as abilities, experiences, values, needs, and emotions. Another perspective that focuses on internal thought activities is design thinking as the thinking of designers.

3.2

Design Thinking as the Thinking of Designers

The research on the thinking of designers emerged from development in experimental psychology, providing research approaches, such as think-aloud protocols (Arnheim, 1954, 1965; Bartlett, 1958; de Groot, 1965; Duncker, 1926, 1935; Koffka, 1912; Köhler, 1925; Selz, 1922; Wertheimer, 1945). Early experimental psychology research informed various design scholars and sparked research on the thinking in design. In the 1960s, design scholars investigated creative problemsolving and schema creation (Alexander, 1962; Gordon, 1961; Harman et al., 1966; Schön, 1963). At Harvard University, Alexander (1962) investigated schema creation in children through their drawings. In the San Francisco Bay Area, Bob McKim and Jim Fadiman studied the effects of psychedelic agents (LSD-25, mescaline) on creative experiences (Harman et al., 1966). The study resulted in various creative solutions and patents (Fadiman, 2011). Notably, Doug Engelbart was part of this creativity study (Markoff, 2005). He introduced the first personal computer in the Mother of All Demos around the same time. This study also revealed the importance

178

J. Auernhammer and B. Roth

of visual thinking and imagery in creative discovery, design, and invention (McKim, 1972, 1980a). Gordon (1961) and Schön (1963) investigated creative discovery in design by investigating and reflecting on their creative group sessions. In the 1970s, various scholars examined creative and intuitive thinking in design (e.g., Akin, 1978; Eastman, 1969, 1970; Lawson, 1972, 1979). In these early developments in research on thinking, different schools of thought emerged that influenced the various design scholars (Auernhammer & Roth, 2022; Cross, 1982; Humphrey, 1963; Rowe, 1987). These schools of thought are (1) structuralism (e.g., information-processing), (2) functionalism (i.e., pragmatism), (3) Gestalt psychology, as well as other schools including constructivism. Research from a structuralist perspective, as developed by Selz (1922), de Groot (1965), Paul M. Fitts, and George Miller, informed design scholars, such as Eastman (1969, 1970), Simon (1969), Akin (1978), Maher and Poon (1996), and many others. The functionalist, as developed by Dewey (1933, 1938) in combination with Gestalt psychology, in particular, Wertheimer (1945), informed the work by Schön (Bamberger & Schön, 1983; Schön, 1963, 1983, 1992). Similarly, Maslow (1943, 1954, 1956) combined the dynamism from functionalism (e.g., Dewey, 1933; Dewey, 1938; James, 1909) and holisticness from Gestalt psychology (e.g., Goldstein, 1934; Koffka, 1935; Wertheimer, 1945) into humanistic psychology, which influenced the work at Stanford’s Design Division (Arnold, 1959; Auernhammer & Roth, 2021; McKim, 1959, 1972; Wilde, 1972). The combination of functionalism from Peirce (1871, 1923) and ideas from structuralism, as discussed by Popper (1961a, 1961b, 1962), informed March (1976) on abductive reasoning in architectural design. These different schools of thought provide different nuances in thinking in design.

3.2.1

Design Cognition as Information-Processing

At Carnegie Mellon University, Allen Newell, Herbert Simon, and John C. Shaw (Newell et al., 1958; Newell & Simon, 1965, 1972) and Charles Eastman (1969, 1970) examined human problem-solving and intuitive design processes through an information-processing perspective. The perspective is based on research by Selz (1922), who outlined productive thinking as creating complexes (i.e., schemata) of purpose-means relationships (Zweckmittelverbinding). Interestingly, this theory was criticized as a “machine theory” by Benary (1923), which informed the research that sparked the field of artificial intelligence in the 1950s. The information-processing theory became a dominant perspective in design cognition, resulting in various descriptive design processes and models (e.g., Akin, 1978; Dorst & Cross, 2001; Eastman, 1970; Maher & Poon, 1996; Simon, 1969; Ullman et al., 1988). A new solution is determined by creating a purpose-means (i.e., problemsolution) relationship complex through direct, reproductive, or anticipative productive thinking (Selz, 1922). Otto Selz’s student Adriaan de Groot (1965) applied these principles of productive thinking to chess players, which in turn informed the work by Herb Simon (1981) and colleagues. This thinking in design developed further into

What Is Design Thinking?

179

the problem space search and solution space, means-end analysis, and problemsolution coevolution (Dorst & Cross, 2001; Maher & Poon, 1996; Simon, 1969). Such design thinking is predominantly based on computational models. Informationprocessing that incorporate specific heuristic strategies (i.e., design methods) are made transparent, reducing thinking into a systematic methodology, as discussed above. Another perspective on the thinking of designers is grounded in formal logic and the philosophy of science.

3.2.2

Abduction as Reasoning in Design

In The Architecture of Form, March (1976) discusses the logic of design. March (1976) separates the goal of science from the objective of design. In science, the major goal is to establish a general law or theory, which requires deduction and inductive reasoning to generalize, as discussed by Popper (1961a, 1961b, 1962). The objective in design is to realize a particular case or design which requires deduction and productive inference as to particularize (i.e., a design hypothesis of a particular instance produced from a general notion and specific data) (March, 1976). The logic of design relates to the work by Peirce (1871, 1923) abductive reasoning. March (1976) develops the logic of design in the context of Alexander (1964) design patterns, trying to develop a logic that resolves the subjective value determination of patterns. In doing so, March (1976) separates deductive or analytical reasoning (i.e., prediction of effects; proof that something must be), inductive or synthetic reasoning (i.e., the discovery of laws; shows that something actually is operative), and abductive or productive reasoning (i.e., the discovery of causes; merely suggests that something may be). Productive reasoning is the inference of a case (i.e., a design) from rule and result, which is the only logic operation that introduces a new idea or injects new values (March, 1976). These three types of reasoning provide the iterative design process of production (i.e., create), deduction (i.e., predict), and induction (i.e., evaluate) (PDI). Each productive inference is value-laden. A substantial challenge is that designers’ underlying values determine the new design, requiring ethical considerations (March, 1976; McKim, 1959; Papanek, 1973). Furthermore, March (1976) outlines that the logic of design in relation to Bayesian probability and decision theory resolves a whole range of issues Christopher Alexander has neglected in the value judgment of patterns. March (1976) concludes that the important aspect of his argument is that there are two types of synthetic reasoning, inductive, related to evaluation, and productive, related to analogy. However, March (1976, p. 28) emphasizes that the chapter was written “in the spirit of opening up discussion rather than laying down a definite statement. Consequently, there are many ragged ends to the argument.” Nevertheless, the abductive reasoning argument has been put forward as thinking or reasoning in design (Dorst, 2011; Roozenburg, 1993).

J. Auernhammer and B. Roth

180

3.2.3

Design Thinking as Gestalt Creation

A more elaborated perspective from psychology is the thinking of designers as Gestalt creations. The Gestalt1 perspective influenced many design scholars (Alexander, 1962, 1964; Arnold, 1954, 1959; Goldschmidt, 1991; Krippendorff, 1989; McKim, 1959, 1972, 1980a, 1980c, circa McKim, 1968; Rittel, 1987; Schön, 1963, 1992; Tversky, 2006; Wilde, 1972). Unlike the information-processing perspective, the Gestalt perspective emphasizes the direct experience of situational circumstances in perception, imagination, thinking, and social interactions (Arnheim, 1954, 1965, 1969; Duncker, 1935, 1945; Koffka, 1935; Köhler, 1925, 1929; Lewin, 1936, 1946, 1947, 1951; Maier, 1930; Perls, 1947; Perls et al., 1951; Wertheimer, 1922, 1923, 1945). This perspective influenced the developments in humanistic psychology that sparked the human potential movement (Hartman, 1959a, 1959b; Maslow, 1943, 1954, 1956, 1959, 1962, 1971; Rogers, 1954). At the center of the human potential movement was the Esalen Institute, where Gestalt therapy was taught through experimental exercises (Callahan, 2009; Perls et al., 1951). Bob McKim and other faculties established the Stanford-Esalen program, bringing these principles into experiential exercises in design education (McKim et al., 1967). For example, Doug Wilde and Bernie Roth established the Peopledynamics Lab that taught the various Gestalt learnings in design education (Wilde, 1972). Similarly, Rittel (1987), in his outline of the evolution of the problem as a whole through new learnings and Krippendorff (1989) product semantics, essential in making sense of things, is grounded in Gestalt considerations (Arnheim, 1967, 1969; Koffka, 1935; Lewin, 1936, 1946; Wertheimer, 1922, 1923, 1945). A new Gestalt (i.e., schema) is created by separating a Gestalt from its environment (i.e., figure-ground) and by recognizing the situation as a new whole in seeing, imagining, and expressing, which is different from the sum of its parts (Alexander, 1962; Koffka, 1912, 1935; Köhler, 1947; McKim, 1972; Wertheimer, 1922, 1923, 1945). Wertheimer (1945) described productive thinking from a Gestalt perspective as experiencing tension or gap (i.e., need and ambiguous situation), which resolves with perceiving, imagining, and creating/enacting a harmonious or clear Gestalt. Such productive thinking incorporates recentering by perceiving the situation as a new whole. Perceiving the whole situation from various perspectives, such as others (i.e., empathy2), overcomes stereotype vision (Arnold, 1954; McKim, 1972; Wertheimer, 1945). However, designers too often have premature closure, seeing only stereotypes, not seeing the obvious, and not perceiving concepts in new ways (Arnold, 1959; McKim, 1972; Schön, 1963; Wertheimer, 1945). Such productive thinking occurs in music compositions, visual art, mathematics, science, conceptual design in architecture, visual design, product semantics, social interactions,

The German term for design is “gestaltung” and for designing is “gestalten.” The psychologist Edward Titchener, who studied in Wilhelm Wundt’s laboratory, introduced the term “empathy” in 1909 into the English language as the translation of the German term “einfühlung.” 1 2

What Is Design Thinking?

181

everyday problem-solving, and sensemaking (Arnheim, 1954, 1965, 1969, 1993; Bamberger & Schön, 1983; Duncker, 1935, 1945; Goldschmidt, 1991; Krippendorff, 1989; Köhler, 1925; Lewin, 1946, 1947; Schön, 1992; von Ehrenfels, 1890; Wertheimer, 1945). It is not just in the domain of “the designers” (Rittel, 1987). This embodied thinking in design is not interdependent of direct experience and situational circumstances. People’s abilities, practices, and circumstances are essential for design thinking.

3.3

Practice-Based Design Thinking (Embodied Thinking)

Practice-based design thinking is grounded in somatic practice and situational circumstances (i.e., embodied thinking). Embodied thinking emphasizes that the psyche is inseparable from acting with a physical body within the total environment in which that body is immersed (Arnold, 1959; Gibson, 2014; Goldstein, 1934). In Chap. 1 of Experiences in Visual Thinking, McKim (1972, 1980a) emphasizes that “[p]sychic (mental) functions cannot be readily separated from somatic (bodily) ones.” Design thinking (i.e., productive thinking in design) is inseparable from design practice and vice versa. Thinking, seeing, imagining, and expressing incorporate human needs, abilities, values, emotions, communication, and collaboration which emerge within interactions with the natural (i.e., materials and living organisms) and cultural environment (i.e., people) (Auernhammer & Roth, 2021; Goldstein, 1934; Hartman, 1959b; Maslow, 1954, 1962; McKim, 1972). For example, McKim (1972, p. 63) expressed that seeing depends on the whole being as follows: Seeing is encountering reality with all of your being. To encounter reality deeply, you cannot leave part of yourself behind. All of your senses, your emotions, your intellect, your language-making abilities - each contributes to seeing fully.

Design cannot be simply “used” or “implemented” by following a recipe or cookbook of various methods. For example, need-sensitivity or empathy, the capability to “feel into,” is not a design or qualitative research method. It is a human capability that is dependent on situational circumstances. There are various practicebased design thinking perspectives that emphasize the importance of experiential learning, including variation-selection (i.e., technological development), meaningmaking (i.e., making sense through things), and comprehensive design (i.e., many thinking languages).

3.3.1

Practice-Based Design Thinking as Variation-Selection

The first practice-based design thinking is creating technology that is grounded in engineering practice (Vincenti, 1990, 1994a). Engineers deal with the physical and technical world in which practice-based design thinking follows a variation-selection learning model (Vincenti, 1990). This model is based on the evolutionary

182

J. Auernhammer and B. Roth

“epistemological ideas of Campbell (1987) and their initial application to technical design and development by Constant II (1980).” Vincenti (1994b, p. 575) describes this model as follows: This model views novelty both in artifacts and in the knowledge underlying them as coming about through [the] introduction of candidate variants, for which the outcome cannot help but be unforeseeable to some degree, followed by selective retention of those that work best in some appropriate sense.

This variation-selection learning model requires testing prototypes through various strategies, such as in sequence and parallel prototyping within the circumstances of operation (Camburn et al., 2017; Vincenti, 1994b). Each variation is examined against various criteria, such as performance, weight, initial cost, reliability, and maintenance, to consider various trade-offs (Vincenti, 1994b). Vincenti (1994b) explains, in the case of the “Northrop anomaly,” that the performance advantages of retraction could be learned only through the (direct) experience of the performance of the prototype in context. Such practice-based design thinking focuses on technical aspects (e.g., functionality) in context.

3.3.2

Practice-Based Design Thinking as Sensemaking

Another practice-based design thinking is making sense of products by making them understandable or meaningful to people based on product semantics (Krippendorff, 1989; Rittel, 1987). This perspective is grounded in Gestalt psychology, which is sharply differentiated from the information-processing perspective. Krippendorff (1989) expressed this design thinking as follows: The criteria that govern choices of this kind [i.e., sense making] show little resemblance with those of problem solving or representational uses. Means and ends are indistinguishable here, and objects and what they mean become one. The criteria are based more on gestalt considerations and are concerned chiefly with how users weave their own identity into the symbolic fabric of society. [emphasis in bold added]

This practice-based design thinking focuses on the symbolic qualities of humanmade artifacts, including shapes and forms, colors, and the organization of interfaces, to make sense of artifacts for people in their cognitive, social, and cultural contexts of use and application (Krippendorff, 1989, 2006; Krippendorff & Butter, 1984). The study of product semantics is the study of the meanings that emerge in human interactions with objects (Krippendorff, 1989). This practice-based design thinking requires direct experience with artifacts in context. For example, designers create the artifact and observe the product semantics of form (i.e., seen as) and meaning (i.e., sensemaking) by the user in a specific context, which informs the designer in the creation of the artifact (Krippendorff & Butter, 1984). This practicebased design thinking emphasizes that designers are making an “argument” (e.g., the presentation of a specific form) that evolves through the various conversations with diverse stakeholders (Krippendorff, 2006).

What Is Design Thinking?

183

Victor Papanek (1973), a student of John Arnold, expresses a fundamental challenge of this design thinking in Designing for the Real-World: Designers too often make sense of things from an egocentric position with auto-therapeutic (unconscious) intents at the expense of spectators and/or consumers. Similarly, Buckminster Fuller (1963) criticized the overemphasis on aspects such as forms and colors in industrial design practices in Ideas and Integrities and emphasized the need for a comprehensive design practice.

3.3.3

Practice-Based Comprehensive Design Thinking

The last practice-based design thinking is based on the idea that “design is the response to a human need” (McKim, 1959). Design as a response to a human need is not focused on technology per se or merely reinterpreting existing artifacts for different identities and lifestyles. It focuses on resolving the tensions (i.e., needs) of people and other living organisms that arise from the natural and cultural environment through designed artifacts, interactions, and environments (Arnold, 1959; Goldstein, 1934; Maslow, 1954; McKim, 1959; Wertheimer, 1945). It follows Wertheimer’s (1945) productive thinking from tensions to harmony. Importantly, it is not about harmonizing the ego needs of designers (i.e., self-expression needs or recognition needs) (McKim, 1959). Therefore, designers are required to develop the sensibility for human needs and communication and collaboration abilities (McKim, 1972, 1980a, 1980b; Wilde, 1972). They must communicate the artifact (often as a prototype) in context to people in need to observe if it resolves the tension. This direct experience is similar to the necessary observations in the variation-selection and sensemaking practice (Krippendorff, 1989; Vincenti, 1994a). McKim (1959) outlined three different types of people’s needs for which the designer needs to design: (1) physical needs, (2) intellectual needs, and (3) emotional (i.e., profound) needs. While physical needs (i.e., functionality and useability) are satisfied by engineering and ergonomics, intellectual needs require making (visually) sense of things (i.e., product semantics for meaning) (McKim, 1959, 1972). Designing for emotional needs involves inventing new designs (i.e., artifacts) that resolve people’s profound tension or needs caused by, e.g., new technology or changing social values (i.e., trends) (McKim, 1980b). Resolving such needs is about making a difference in the world. For example, the introduction of Apple’s iTunes in 2001 resolved the tension for artists (i.e., the profound need of financial well-being) and executives of music production companies (i.e., the profound need of company survival) caused by the human behavior of downloading illegally songs enabled by the platform Napster. Designing for the whole person and various groups (i.e., physical, intellectual, and emotional needs) requires comprehensive design.3 Comprehensive design brings

3

The term comprehensive design was introduced by Buckminster Fuller and adapted by John Arnold (1954).

184

J. Auernhammer and B. Roth

together capabilities from science, engineering, fine arts, psychology, and management and therefore abilities in all thinking languages (e.g., visual, mathematical, emotional, kinesthetic, and verbal languages) (Adams, 1974; McKim, 1972). Other areas of comprehensive design include values (e.g., need-sensitivity awareness), attitudes (e.g., questioning and observing), and creative confidence, which is achieved by overcoming various blocks (e.g., perceptual, emotional, and cultural blocks) (Adams, 1974; Arnold, 1954, 1959; Auernhammer & Roth, 2021; Kelley & Kelley, 2013; McKim, 1959, 1972, 1980a, 1980b, 1980c). For example, McKim (circa 1968, p. 3) discussed the effects of diverse design languages on design thinking as follows: Before giving examples of the effect of language specialization upon design thinking, let me first develop a larger context for the discussion. There are, of course, many languages of design: Each language has its own informational content. Design languages can be arranged in a hierarchy, from the abstract to the real, as shown in Figure 2 [see Figure 2 in this article]. Each design language, on the hierarchy from abstract to real, can play a special role in furthering the design process [i.e., internal process through imagination and rapid visualization]. The designer[s] who limit [themselves] to the one or two design languages [they] use for communication, thereby limits [their] thinking to the specialized information contained in these languages. [emphasis in bold added]

Each language and related abilities provide an essential design or visual thinking aspect, as illustrated in Fig. 2. For example, abstract visual analogs make abstract ideas visible, while orthographic projection makes spatial relationships visible and measurable. The importance of mockups, working prototypes, and appearance models to design thinking is often underestimated. Quickly executed cardboard mockups can enormously stimulate the imagination because direct, multisensory inputs are literally “food for thought” (McKim, circa 1968). McKim (circa 1968) emphasizes the importance of fluency in different languages as follows: [T]he designer who lacks fluency in the most concrete, three-dimensional languages of design often suffers a kind of imaginative malnutrition which I call ‘abstractionitis’. The designer's predecessor, the craftsman, fed [her/]his senses in the pursuit of [her/]his craft. The ‘paper designer’ lacks this essential sustenance, and the quality of [her/]his design thinking often shows it.

To facilitate the development of students’ ability of imagination (mind’s eye), Bob McKim, Nelson van Judah, and several students developed the Imaginarium, as shown in Fig. 3. The Imaginarium is a 16-foot (4.88 m) diameter geodesic dome “designed to enable you to experience the content and infinite inner space of your imagination” (McKim). Inside the Imaginarium, underneath the platform on which people lied down, was an array of audiovisual equipment: slide projects, a dissolver unit, and a 16 mm movie projector, which sends imagery to a 45° mirror, reflecting it to the spherical lens, where it is in turn expanded to the ceiling of the dome (McKim, n.d.). An audiorecorded narration and other objects (e.g., an apple) facilitated the experience of imagining. In addition to training the abstract to concrete thinking languages and related abilities, Arnold (1959, 1962a), McKim (1959, 1972, 1980a, 1980b, 1980c, circa

What Is Design Thinking?

Fig. 2 Design languages: abstract to real—adapted from the original by McKim (circa 1968) Fig. 3 Bob McKim in front of the Imaginarium—a geometric dome to experience and develop the ability of inner image (imagination, mind’s eye). The image was taken by Jose Mercado

185

186

J. Auernhammer and B. Roth

Fig. 4 The many languages and human qualities of comprehensive design thinking

1968), Adams (1974, 2019), Roth (1973, 2015), Wilde (1972), Kahn (2005), (Molenkamp & Center, 1989), Faste (1987, 1994), (Sheppard et al., 2009), (Dym et al., 2005; Leifer & Steinert, 2011), (Kelley & Kelley, 2013), and many others developed and taught various experiential exercises for students to experience comprehensive design thinking from many diverse perspectives. For example, encounter groups provide experiential learning of group dynamics, developing creative collaborative abilities. These experiential practices were led by Bernie Roth and Doug Wilde in a class called Peopledynamics Lab (Wilde, 1972), which was heavily influenced by Gestalt therapy by Perls (Perls, 1947; Perls et al., 1951). Such experimental enabled group dynamics to creatively collaborate with many diverse people. These experiential exercises aim to develop the necessary human qualities in design, such as attitudes, abilities, sensitivities, values, confidence, thinking languages, practices, and various techniques required for this comprehensive design (Auernhammer & Roth, 2021). This practice-based comprehensive design thinking focuses on designers as people and their direct experiences and embodied thinking. It requires fluency in approach (e.g., different thinking languages, creative collaboration, and diverse perspectives/experiences), diversity in abilities (e.g., seeing, imagining, and expressing), and human values (e.g., attitudes and sensitivities) to respond to the specific situational circumstances (Adams, 2019; Arnold, 1959; Auernhammer & Roth, 2021; McKim, 1959, 1980a, 1980b). Figure 4 illustrates the many domains, languages, and human qualities of the practice-based comprehensive design thinking. The practice-based comprehensive design thinking aims to resolve the tensions (i.e., needs) inherent in the natural and sociocultural environment (i.e., performance, meaning, and profound tensions), which is an ongoing, never-ending activity of evolving Gestalt (Arnold, 1959; Auernhammer & Roth, 2021; McKim, 1959; Rittel & Webber, 1973; Wertheimer, 1945).

What Is Design Thinking?

187

4 Implication and Contribution In this chapter, we show that the widely used term “design thinking” is understood and used from diverse perspectives. Many scholars and practitioners refer to the umbrella construct “design thinking” without clarifying which perspective of design thinking they are referring to. This chapter elucidates some of the substantial differences and nuances of these different perspectives on design thinking. Methodology views design thinking as independent of people’s abilities and situational circumstances. Similarly, perspectives of thinking of designers view design thinking as mental activities, often independent of embodied activities and practices. Different schools of thought provide substantial nuances in the understanding of thinking in design. These considerable nuances view design thinking as either determining purpose-means relationships (i.e., information-processor), productive reasoning (i.e., the logic of design), or creating a harmonious Gestalt (i.e., productive thinking through recentering) (Peirce, 1923; Selz, 1922; Wertheimer, 1922, 1923, 1945). The practice-based, embodied thinking in design emphasizes direct experience that is dependent upon human potentialities (e.g., abilities, attitudes, and values) and situational circumstances (e.g., performance, meaning, profound needs, and emerging impact) (Krippendorff, 2006; McKim, 1980a; Vincenti, 1990). Different practices, such as technical prototyping for physical performance, detailed representation for social identity and meaning, and comprehensive practices to respond to people’s profound needs, require different thinking languages. These differences in the understanding of design thinking have several implications for research, education, and practice.

4.1

Implication for Research

The schema on the different perspectives on design thinking (Fig. 1) allows scholars and researchers to refer to specific types of design thinking, going beyond referring to the all-encompassing umbrella term. Design thinking as a methodology (e.g., a universal recipe) is a common perspective discussed in nontraditional design fields, such as management (Elsbach & Stigliani, 2018; Micheli et al., 2019). The designers’ thinking is in the domain of psychology, which design researchers embraced and contextualized (Auernhammer & Roth, 2022; Rowe, 1987). Practice-based design thinking interlinks thinking with design qualities, such as specific practices and situational circumstances through direct experiences (Auernhammer & Roth, 2021; Krippendorff, 2006; McKim, 1972, 1980b, circa 1968; Vincenti, 1990). Each perspective incorporates different research questions. Research on design thinking as methodology examines the “implementation” of a set of methods, often used in workshop formats, and its potential outcomes (Magistretti et al., 2020). Research on design thinking as the thinking of designers focuses on the cognitive activities of designers, which is often examined through experimental

188

J. Auernhammer and B. Roth

designs in laboratory settings (Bamberger & Schön, 1983; Lawson, 1979). The results are interpreted through different schools of thought, such as an information-processing and Gestalt psychology perspective, producing different theories, which inform different practice. Research in design thinking as a practice investigates the various human qualities, such as values, attitudes, abilities, and practices necessary for specific design practices, e.g., variation-selection model, meaning-making, or comprehensive design. From this perspective, many unanswered questions could provide further explanations of the fundamental human activity of designing inventions that resolve the profound tensions in nature and society. Research should investigate human aspects, such as the influence of designers’ values on design outcomes, designers’ sensitivity to needs in relation to new inventions, and abilities to collaborate and the resolution of need-tensions of different groups. Such research provides further insights into enabling people’s experiences related to the human qualities of comprehensive design.

4.2

Implications for Practice and Education

Each of the different perspectives on design thinking emphasizes different practices. The methodology perspective emphasizes the facilitation and enactment of methods. The methodological perspective makes the design process transparent for diverse people to contribute to specific activities. However, such cookbook approaches are not without substantial limitation and criticism (Alexander, 1971a, 1971b; Jones, 1977, 1991; McKim, 1972, 1980c). A cookbook approach may or may not allow an adequate response to the situation and may not include adequate abilities and attitudes. The thinking of designers focuses on developing specific heuristic strategies to make design decisions and solve problems, develop knowledge systems to judge the value of a design, and develop the seeing, imagining, and expressing abilities for creating Gestalts (Lawson, 1979; March, 1976; McKim, 1980a). However, thinking is not independent of human qualities, such as practice and abilities (Arnold, 1959; Auernhammer & Roth, 2021; McKim, 1980a). The practice-based design thinking emphasizes the development of these human qualities, such as prototyping practices to select the superior variation, meaning-making through visual abilities, different thinking languages, need-sensitivity, and collaborative abilities to respond to profound tensions in nature and society (Arnold, 1959; Krippendorff, 2006; McKim, 1959, 1980a, 1980b, circa 1968; Vincenti, 1990). As John Arnold (1959) and Buckminster Fuller (1969, 1981) highlighted, developing the human potentialities essential for the practice-based comprehensive design thinking is imperative for resolving the tension inherent in today’s global challenges. Acknowledgments In loving memory of Bob McKim, Jim Adams, and Doug Wilde, who recently passed away. In memory of some of the most important design practitioners, educators, and researchers (Table 2) that profoundly influenced our understanding of design and its related thinking.

What Is Design Thinking?

189

Table 2 In memory of some of the pioneers in design and its related thinking Name Klaus Krippendorff John Chris Jones Robert H. McKim Christopher Alexander James L. Adams Douglas Wilde Ömer Akin Walter G. Vincenti Tomás Maldonado

Mar. 21, 1932 Oct. 7, 1927 Sept. 24, 1926 Oct. 4, 1936 Mar. 6, 1934 Aug. 1, 1929 June 30, 1946 April 20, 1917 April 25, 1922

Oct, 2022

University University of Pennsylvania

Aug. 13, 2022 Jul. 17, 2022

University of Manchester Open University Stanford University

Mar. 17, 2022 Jan. 15, 2022 Oct. 28, 2021 Mar. 13, 2020 Oct. 11, 2019

University of California, Berkeley

Nov. 26, 2018

Lionel John March

Jan. 26, 1934

Feb. 20, 2018

Sara little Turnbull

Sept. 21, 1917 May 29, 1928 June 25, 1943 Nov. 22, 1922 Sept. 6, 1943

Sept. 42,015

Matthew Kahn William Moggridge Bruce L. Archer Rolf Faste Herbert Simon Victor J. Papanek

June 15, 1916 Nov. 22, 1923

June 24, 2013 Sept. 8, 2012 May 16, 2005 Mar. 6, 2003 Feb. 9, 2001 Jan. 10, 1998

Donald Schön

Sept 19, 1930

Horst Rittel

July 14, 1930

Sept. 13, 1997 July 9, 1990

Richard Buckminster Fuller John E. Arnold

July 12, 1895

July 1, 1983

Mar. 14, 1913

Sept. 28, 1963

Stanford University Stanford University Carnegie Mellon University Stanford University Politecnico di Milano Hochschule für Gestaltung Ulm Princeton’s School of Architecture University of Cambridge University of California, Los Angeles Stanford grad. School of Business Stanford University Royal College of Art Stanford University Royal College of Art Syracuse University Stanford University Carnegie Mellon University California Institute of the Arts Kansas City art Institute University of Kansas Massachusetts Institute of Technology University of California, Berkeley Hochschule für Gestaltung Ulm Harvard University Massachusetts Institute of Technology Stanford University

. . . and so many more (as this list only reflects people who published in English)

190

J. Auernhammer and B. Roth

Additionally, we are grateful for the support by the Hasso Plattner Design Thinking Research program, in particular Larry Leifer, Christoph Meinel, and Hasso Plattner. We also thank the School of Engineering for the permission to reprint the picture of Bob McKim in front of the Imaginarium. We would like to thank the wider design and design research community for the support and many conversations with countless individuals. For the many insights on the developments of the comprehensive design education and practices at Stanford University, we want to thank Dave Beach, David Kelley, Larry Leifer, Sheri Sheppard, Bill Verplank, Jim Fadiman, Bill Scott, Craig Milroy, Gayle Curtis, Steven McCarthy, Michael Barry, Ade Mabogunje, Carissa Carter, Bill Burnett, Dennis and Brendan Boyle, Louis Hsiao, Kathy Davies, George Kembel, Perry Klebahn, Haakon Faste, Mike Nuttall, Jim Yurchenko, Pam Greene, Linda Kuo, Bill Potts, Peter Dreissigacker, Jerry Manock, and many more. For input on John E. Arnold, we would like to thank William Clancey and Subarna Basnet. Many thanks go to Jill Grinager and Sharon Nemeth for their support.

References Adams, J. L. (1974). Conceptual blockbusting: A guide to better ideas. Stanford Alumni Association. Adams, J. L. (2019). Conceptual blockbusting: A guide to better ideas (5th ed.). Basic Books. Akin, Ö. (1978). How do architects design? In IFIP working conference - artificial intelligence and pattern recognition in computer aided design, North-Holland, New York. Alexander, C. (1962). The ORIGIN of creative power in children*. The British Journal of Aesthetics, 2(3), 207–226. https://doi.org/10.1093/bjaesthetics/2.3.207 Alexander, C. (1964). Notes on the synthesis of form. Harvard University Press. Alexander, C. (1971a). Notes of the synthesis of form. Harvard University Press. Alexander, C. (1971b). The state of the art in design methods. DMG Newsletter, 5(3). Archer, B. L. (1965). Systematic method for designers. Council of Industrial Design. Arnheim, R. (1954). Art and visual perception. University of California. Arnheim, R. (1965). Visual thinking. In G. Kepes (Ed.), Education of vision (Vol. Vision + value series). George Braziller. Arnheim, R. (1967). Art and visual perception: A psychology of the creative eye. Faber. Arnheim, R. (1969). Visual thinking. University of California Press. Arnheim, R. (1993). Sketching and the psychology of design. Design Issues, 9(2), 15–19. https:// doi.org/10.2307/1511669 Arnold, J. E. (1954). Selected summer session notes 1954. Massachusetts Institute of Technology. Arnold, J. E. (1959). Creative engineering seminar, 1959. Stanford, University. Arnold, J. E. (1962a). Education for innovation. In S. J. Parnes & H. F. Harding (Eds.), A source book for creative thinking. Charles Scribner’s Sons. Arnold, J. E. (1962b). Useful Creative Techniques. In S. J. Parnes & H. F. Harding (Eds.), A source book for creative thinking. Charles Scribner’s Sons. Arnold, J. E. (2016). Creative Engineering promoting innovation by thinking differently. Edited with an introduction and biographical essay by W. J. Clancey. Stanford Digital Repository. Arnold, J. E., Blake, S. P., & Jones, S. (1960). The generalist-specialist dichotomy in the Management of Creative Personnel. Stanford University Libraries. Asimow, M. (1962). Introduction to design. Prentice-Hall. Auernhammer, J., & Leifer, L. (2023). Innovation for the real-world through knowing why. In Y. Ohsawa (Ed.), Living beyond data: Toward sustainable value creation (pp. 29–38). Springer International Publishing. https://doi.org/10.1007/978-3-031-11593-6_2 Auernhammer, J. M., & Ford, C. (2022). What is design? The semantic Core and periphery of design definitions design thinking research symposium 13. Haifa.

What Is Design Thinking?

191

Auernhammer, J. M., Leifer, L., Meinel, C., & Roth, B. (2022). A humanistic and creative philosophy of design. In C. Meinel & L. Leifer (Eds.), Design thinking research: Achieving real innovation. Springer Nature. Auernhammer, J. M., & Roth, B. (2021). The Origin and evolution of Stanford University’s design thinking: From product design to design thinking in innovation management. Journal of Product Innovation Management, 38, 623–644. https://doi.org/10.1111/jpim.12594 Auernhammer, J. M., & Roth, B. (2022). Different types of productive thinking in design: From rational to social design thinking. In C. Meinel & L. Leifer (Eds.), Design thinking research: Achieving real innovation (Vol. 13). Springer Nature. Bamberger, J., & Schön, D. A. (1983). Learning as reflective conversation with materials: Notes from work in Progress. Art Education, 36(2), 68–73. https://doi.org/10.2307/3192667 Bartlett, F. C. (1958). Thinking an experimental and social study. Basic Books. Benary, W. (1923). Besprechung von II. Psychologische Forschung, 3, 417–425. Brown, T. (2009). Change by design: How design thinking transforms organizations and inspires innovation. HarperCollins. Bruner, J. S., Goodnow, J. J., & Austin, G. A. (1956). A study of thinking. John Wiley and Sons. Callahan, J. F. (2009). Manual of gestalt practice in the tradition of dick Price. The Gestalt Legacy Project. Camburn, B., Viswanathan, V., Linsey, J., Anderson, D., Jensen, D., Crawford, R., et al. (2017). Design prototyping methods: State of the art in strategies, techniques, and guidelines. Design. Science, 3, e13., Article e13. https://doi.org/10.1017/dsj.2017.10 Campbell, D. T. (1987). Evolutionary epistemology. In G. Radnitzky & W. W. Bartley III (Eds.), Evolutionary epistemology. Rationality and the sociology of knowledge (pp. 47–89). Open Court. Clancey, W. J. (forthcoming). The philosophy of comprehensive design: The collected articles and speeches of John E. Arnold. Manuscript in preparation. (Original work 1954–1961). Constant, E. W., II. (1980). The origins of the turbojet revolution (pp. 6–8). Johns Hopkins. Crawford, R. P. (1954). The techniques of creative thinking: How to use your ideas to achieve success. Hawthorn Books. Cross, N. (1982). Designerly ways of knowing. Design Studies, 3(4), 221–227. https://doi.org/10. 1016/0142-694X(82)90040-0 Curtis, G. (1990). Methodology handbook David Kelley Design. David Kelley Design. de Groot, A. D. (1965). Thought and choice in chess. Mouton. Dewey, J. (1933). How we think. Wyatt North Publishing, LLC. Dewey, J. (1938). Logic the theory of inquiry. Henry Holt and Company. Dorst, K. (2011). The core of ‘design thinking’ and its application. Design Studies, 32(6), 521–532. https://doi.org/10.1016/j.destud.2011.07.006 Dorst, K. (2015). Frame innovation: Create new thinking by design. MIT Press. Dorst, K., & Cross, N. (2001). Creativity in the design process: Co-evolution of problem–solution. Design Studies, 22, 425–437. https://doi.org/10.1016/S0142-694X(01)00009-6 Dubberly, H. (2005). How do you design? In A compendium of models. Dubberly Design Office. Duncker, K. (1926). A qualitative (experimental and theoretical) study of productive thinking (solving of comprehensible problems). The Pedagogical Seminary and Journal of Genetic Psychology, 33(4), 642–708. https://doi.org/10.1080/08856559.1926.10533052 Duncker, K. (1935). Zur Psychologie des produktiven Denkens [The psychology of productive thought.]. Springer. Duncker, K. (1945). On Problem-solving (Vol. 58, p. i). American Psychological Association. Dweck, C. S. (2006). Mindset: The new psychology of success. Random House Publishing Group. Dym, C. L., Agogino, A. M., Eris, O., Frey, D. D., & Leifer, L. J. (2005). Engineering design thinking, teaching, and learning [Review]. Journal of Engineering Education, 94(1), 103–120. https://doi.org/10.1002/j.2168-9830.2005.tb00832.x Eastman, C. (1969). Cognitive processes and ill-defined problems: A case study from design. International Joint Conference on Artificial Intelligence: IJCAI.

192

J. Auernhammer and B. Roth

Eastman, C. (1970). On the analysis of intuitive design processes. In G. T. Moore (Ed.), Emerging methods in environmental design and planning. MIT Press. Elsbach, K. D., & Stigliani, I. (2018). Design thinking and organizational culture: A review and framework for future research. Journal of Management, 44(6), 2274–2306. https://doi.org/10. 1177/0149206317744252 Fadiman, J. (1986). Omega: Be all that you are. Westlake Press. Fadiman, J. (2011). The psychedelic Explorer's guide: Safe, therapeutic, and sacred journeys. Park Street Press. Faste, R. A. (1987). Perceiving needs. SAE Transactions, 96, 419–423. https://doi.org/10.2307/ 44472796 Faste, R. A. (1994). Ambidextrous thinking. In Innovations in mechanical engineering curricula for the 1990s. American Society of Mechanical Engineers. Fuller, R. B. (1963). Ideas and integrities. Prentice Hall. Fuller, R. B. (1969). Operating manual for spaceship earth. E.P. Dutton & Co. Fuller, R. B. (1981). Critical path. St. Martin’s Publishing Group. Gibson, J. J. (2014). The ecological approach to visual perception (Classic ed.). Taylor & Francis. Goldin-Meadow, S. (2003). Hearing gesture how our hands help us think. Harvard University Press. Goldschmidt, G. (1991). The dialectics of sketching. Creativity Research Journal, 4(2), 123–143. https://doi.org/10.1080/10400419109534381 Goldstein, K. (1934). Der Aufbau des Organismus. Einführung in die Biologie unter besonderer Berücksichtigung der Erfahrungen am kranken Menschen. Nijhoff. Gordon, W. J. J. (1961). Synectics: The development of creative capacity. Harper & Row. Gregory, S. A. (1966). The design method. Butterworth Press. Guilford, J. P. (1950). Creativity. American Psychologist, 5(9), 444–454. https://doi.org/10.1037/ h0063487 Harman, W. W., McKim, R. H., Mogar, R. E., Fadiman, J., & Stolaroff, M. J. (1966). Psychedelic agents in creative problem-solving: A pilot study. Psychological Reports, 19(1), 211–227. https://doi.org/10.2466/pr0.1966.19.1.211 Hartman, R. S. (1959a). The science of value. In A. H. Maslow (Ed.), New knowledge in human values (pp. 13–37). Regency/Gateway Inc. Hartman, R. S. (1959b). The value structure of creativity. In J. E. Arnold (Ed.), Creative engineering seminar, 1959. Stanford University. Hubka, V., & Eder, E. W. (1987). A scientific approach to engineering design. Design Studies, 8(3), 123–137. https://doi.org/10.1016/0142-694X(87)90035-4 Humphrey, G. (1963). Thinking an introduction to its experimental psychology. John Wiley and Sons. Ideo. (2015). The field guide to human-centered design: Design kit. IDEO. James, W. (1909). The meaning of truth: A sequel to ‘Pragmatism’. Longmans, Green. Jansson, D. G., & Smith, S. M. (1991). Design fixation. Design Studies, 12(1), 3–11. https://doi.org/ 10.1016/0142-694X(91)90003-F Jay, L. S. (1962). A systematic approach to the problems of town and regional planning. In Conference on Systematic and Intuitive Methods in Engineering, Industrial Design, Architecture and Communications, London, UK. Jones, J. C. (1959). A Systematic Design Method. Design, 124, 48–53. Jones, J. C. (1970). Design methods: Seeds of human futures. John Wiley & Sons. Jones, J. C. (1977). How my thoughts about design methods have changed during the years. Design Methods and Theories, 11(1), 48–62. Jones, J. C. (1991). designing designing. Architecture Design and Technology Press. Jones, J. C., & Thornley, D. G. (1963). Conference on design methods. Pergamon Press. Kahn, M. (2005). Design/Soul & Body. Stanford University. https://exhibits.stanford.edu/Kahn/ feature/design-soul-body

What Is Design Thinking?

193

Kelley, T., & Kelley, D. (2013). Creative confidence: Unleashing the creative potential within us all. Crown Publishing Group. Koffka, K. (1912). Zur Analyse der Vorstellungen und ihrer Gesetze: eine experimentelle Untersuchung. Quelle & Meyer. Koffka, K. (1927). Bemerkungen zur Denk-Psychologie. Psychologische Forschung, 9(1), 163–183. https://doi.org/10.1007/BF02409758 Koffka, K. (1935). Principles of gestalt psychology. Harcourt. Krippendorff, K. (1961). Produktgestalter kontra Konstrukteur. Output, 4(5), 18–21. Krippendorff, K. (1968). A perspective for the design department: Preliminary suggestions for reorienting the Department of Design at the University of California, Berkeley. https:// repository.upenn.edu/asc_papers/806. Krippendorff, K. (1984). Die Produkt-Semantik öffent die Türen. Zeitschrift für Gestaltung, 108(109), 14–16. Krippendorff, K. (1989). On the essential contexts of artifacts or on the proposition that "design is making sense (of things)". Design Issues, 5(2), 9–39. https://doi.org/10.2307/1511512 Krippendorff, K. (2006). The semantic turn: A new Foundation for Design. Taylor & Francis. Krippendorff, K., & Butter, R. (1984). Product semantics - exploring the symbolic qualities of form. Departmental Papers, ASC), 3. Kumar, V. (2012). 101 design methods: A structured approach for driving innovation in your organization. Wiley. Köhler, W. (1925). The mentality of apes. In. Harcourt Brace & Co. Köhler, W. (1929). Gestalt psychology. Liveright. Köhler, W. (1947). Gestalt psychology: An introduction to new concepts in modern psychology. New American Library. Lawson, B. (1972). Problem solving in architectural design. Aston University. Lawson, B. (1979). Cognitive strategies in architectural design. Ergonomics, 22(1), 59–68. https:// doi.org/10.1080/00140137908924589 Lawson, B. (1980). How designers think. Architectural Press. Leifer, L. J., & Steinert, M. (2011). Dancing with ambiguity: Causality behavior, design thinking, and triple-loop-learning. Information.Knowledge.Systems Management, 10(1–4), 151–173. Lewin, K. (1936). Principles of topological psychology. McGraw-Hill book Company. Lewin, K. (1946). Action research and minority problems. Journal of Social Issues, 2(4), 34–46. https://doi.org/10.1111/j.1540-4560.1946.tb02295.x Lewin, K. (1947). Frontiers in group dynamics: Concept, method and reality in social science; social equilibria and social change. Human Relations, 1(1), 5–41. https://doi.org/10.1177/ 001872674700100103 Lewin, K. (1951). Field theory in social science: Selected theoretical papers (edited by Dorwin cartwright.). The Harp. Liedtka, J., & Ogilvie, T. (2011). Designing for growth: A design thinking tool kit for managers. Columbia Business School Pub., Columbia University Press. Magistretti, S., Dell'Era, C., & Doppio, N. (2020). Design sprint for SMEs: An organizational taxonomy based on configuration theory. Management Decision, 58(9), 1803–1817. https://doi. org/10.1108/MD-10-2019-1501 Maher, M. L., & Poon, J. (1996). Modeling design exploration as co-evolution [10.1111/ j.1467-8667.1996.tb00323.X]. Computer-Aided Civil and Infrastructure Engineering, 11(3), 195–209. https://doi.org/10.1111/j.1467-8667.1996.tb00323.x Maier, N. R. F. (1930). Reasoning in humans. I. on direction. Journal of Comparative Psychology, 10(2), 115–143. https://doi.org/10.1037/h0073232 March, L. (1976). The logic of design and the question of value. In L. March (Ed.), The architecture of form. Cambridge University Press. Markoff, J. (2005). What the dormouse said: How the sixties counterculture shaped the personal computer industry. Penguin Publishing Group.

194

J. Auernhammer and B. Roth

Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50(4), 370–396. https://doi.org/10.1037/h0054346 Maslow, A. H. (1954). Motivation and personality. Harper & Brothers. Maslow, A. H. (1956). Toward a humanistic psychology. ETC: A Review of General Semantics, 14(1), 10–22. Maslow, A. H. (1959). New knowledge in human values. HarperCollins Publishers. Maslow, A. H. (1962). Emotional blocks to creativity. In S. J. Parnes & H. F. Harding (Eds.), A source book for creative thinking (pp. 93–103). Charles Scribner's Sons. Maslow, A. H. (1971). The farther reaches of human nature. Viking Press. McKim, R. H. (n.d.). IMAGINARIUM one/script. Stanford University. McKim, R. H. (1959). Designing for the whole man. In J. E. Arnold (Ed.), Creative engineering seminar, 1959. Stanford University. McKim, R. H. (1972). Experiences in visual thinking. Belmont, CA. McKim, R. H. (1980a). Experiences in visual thinking. Brooks/Cole Publishing Company. McKim, R. H. (1980b). Need-finding chapters. Stanford University. McKim, R. H. (1980c). Thinking visually: A strategy manual for problem solving. Dale Seymour Publications. McKim, R. H. (circa 1968). Visual thinking and the design process. Stanford University. McKim, R. H., Smith, J., Sanford, N., Hastings, A., Eisner, E., Roth, B., et al. (1967). Winter 1967 experiential education - an introduction to the Esalen program at Stanford University. Stanford University. Micheli, P., Wilner, S. J. S., Bhatti, S. H., Mura, M., & Beverland, M. B. (2019). Doing design thinking: Conceptual review, synthesis, and research agenda. Journal of Product Innovation Management, 36(2), 124–148. https://doi.org/10.1111/jpim.12466 Molenkamp, J., & Center, S. (1989). Type terms and other useful information on type. In Typography and Printing Technology. HP Book Series. Muller, W. (1989). Design discipline and the significance of visuo-spatial thinking. Design Studies, 10(1), 12–23. https://doi.org/10.1016/0142-694X(89)90021-5 Newell, A., Shaw, J. C., & Simon, H. A. (1958). Elements of a theory of human problem solving. Psychological Review, 65(3), 151–166. https://doi.org/10.1037/h0048495 Newell, A., & Simon, H. (1956). The logic theory machine--a complex information processing system. IRE Transactions on Information Theory, 2(3), 61–79. https://doi.org/10.1109/TIT. 1956.1056797 Newell, A., & Simon, H. A. (1965). Simulation of human processing of information. The American Mathematical Monthly, 72(2), 111–118. https://doi.org/10.2307/2313317 Newell, A., & Simon, H. A. (1972). Human Problem Solving. Prentice-Hall. Osborn, A. F. (1957). Applied imagination principles and procedures of creative problem-solving. Charles Scribner's Sons. Pahl, G., Beitz, W., & Wallace, K. (1996). Engineering design: A systematic approach. Design Council. Papanek, V. (1973). Design for the Real World. Peirce, C. S. (1871). The works of George Berkeley. The North American Review, 113, 449–472. Peirce, C. S. (1923). Chance, love and logic. In Philosophical Essays. Kegan Paul, Trench, Trubner & Company. Perls, F. S. (1947). Ego, hunger, and aggression. In The Beginning of Gestalt Therapy. The Gestalt Journal Press. Perls, F. S., Hefferline, R., & Goodman, P. (1951). Gestalt therapy: Excitement and growth in the human personality. Dell Publishing. Piaget, J. (1951). Play, dreams and imitation in childhood. . Popper, K. (1961a). The logic of scientific discovery, first published as Logik der Forschung in 1934. Science Editions. Popper, K. (1961b). The poverty of historicism, first published in 1957. Routledge and Kegan Paul.

What Is Design Thinking?

195

Popper, K. (1962). The open society and its enemies. In First published in 1945 (Vol. 2 vols). Routledge and Kegan Paul. Prince, G. M. (1968). The operational mechanism of synectics. The Journal of Creative Behavior, 2(1), 1–13. https://doi.org/10.1002/j.2162-6057.1968.tb00077.x Pólya, G. (1945). How to solve it: A new aspect of mathematical method. Princeton University Press. Pólya, G. (1957). How to solve it: A new aspect of mathematical method. Doubleday. Rittel, H. (1987). The reasoning of designers: Delivered at the international congress on planning and design theory. IGP. Rittel, H., & Webber, M. (1973). Dilemmas in a general theory of planning. Policy Sciences, 4(2), 155–169. https://doi.org/10.1007/BF01405730 Rogers, C. R. (1954). Toward a theory of creativity. ETC: A Review of General Semantics, 11(4), 249–260. Roozenburg, N. F. M. (1993). On the pattern of reasoning in innovative design. Design Studies, 14(1), 4–18. https://doi.org/10.1016/S0142-694X(05)80002-X Roth, B. (1973). Design process and creativity. Beer Sheva. Roth, B. (2015). The achievement habit: Stop wishing, start doing, and take command of your life. Harper Business. Rowe, P. G. (1987). Design thinking. MIT Press. Schön, D. A. (1963). Displacement of concepts. Tavistock Publications. Schön, D. A. (1983). The reflective practitioner: How professionals think in action. Basic Books. Schön, D. A. (1992). Designing as reflective conversation with the materials of a design situation. Knowledge-Based Systems, 5(1), 3–14. https://doi.org/10.1016/0950-7051(92)90020-G Selz, O. (1922). Über die Gesetze des geordneten Denkverlaufs: Zur Psychologie des Produktiven Denkens und des Irrtums. F. Cohen. Sheppard, S. D., Macatangay, K., Colby, A., Sullivan, W. M., & Shulman, L. S. (2009). Educating engineers designing for the future of the field. Wiley. Simon, H. A. (1969). The sciences of the artificial. MIT Press. Simon, H. A. (1981). Otto Selz and information-processing psychology. In N. H. Frijda & A. D. d’ Groot (Eds.), Otto Selz: His contribution to psychology (pp. 147–163). De Gruyter Mouton. https://doi.org/10.1515/9783110816877-006 Tang, J. C., & Leifer, L. J. (1988). A framework for understanding the workspace activity of design teams. In Proceedings of the 1988 ACM conference on computer-supported cooperative work, Portland, Oregon, USA. Tversky, B. (2006). Gestalts of thought. In Visual thought (pp. 155–163). John Benjamins Publishing Company. https://doi.org/10.1075/aicr.67.12tve Ullman, D. G., Dietterich, T. G., & Stauffer, L. A. (1988). A model of the mechanical design process based on empirical data. Artificial Intelligence for Engineering Design, Analysis and Manufacturing, 2(1), 33–52. https://doi.org/10.1017/S0890060400000536 Vincenti, W. G. (1990). What engineers know and how they know it: Analytical studies from aeronautical history. Johns Hopkins University Press. Vincenti, W. G. (1994a). The retractable airplane landing gear and the Northrop “anomaly”: Variation-selection and the shaping of technology. Technology and Culture, 35(1), 1–33. https://doi.org/10.2307/3106747 Vincenti, W. G. (1994b). Variation—Selection in the innovation of the retractable airplane landing gear: The Northrop ‘anomaly’. Research Policy, 23(5), 575–582. https://doi.org/10. 1016/0048-7333(94)01007-2 von Ehrenfels, C. (1890). Über Gestaltqualitäten. Vierteljahrsschrift für wissenschaftliche Philosophie, 14(3), 249–292. Wertheimer, M. (1922). Untersuchungen zur Lehre von der Gestalt. Psychologische Forschung, 1(1), 47–58. https://doi.org/10.1007/BF00410385 Wertheimer, M. (1923). Untersuchungen zur Lehre von der Gestalt II. Psycologische Forschung, 4, 301–350.

196

J. Auernhammer and B. Roth

Wertheimer, M. (1945). Productive thinking. Harper. Wilde, D. (1972). Gestalt learning in an engineering class. In American Society of Engineering Education, Educational Research Methods. Wittgenstein, L. (1953). Philosophical investigations. Basil Blackwell. Zwicky, F. (1948). The morphological method of analysis and construction. In Studies and essays. (Vol. Courant Anniversary Volume.). Interscience.

NeuroDesign: Greater than the Sum of Its Parts Jan Auernhammer, Jennifer Bruno, Alexa Booras, Claire McIntyre, Daniel Hasegan, and Manish Saggar

Abstract This chapter outlines the recent developments, such as neuroscience on design, design neurocognition, and NeuroDesign, in the intersection of neuroscience and design. This intersection of diverse disciplines, including psychology, neurophysiology, engineering, interaction design, and architecture, provides various opportunities and challenges to advance areas, such as design thinking, neurotechnology, embodied artificial intelligence (AI), and human-centered AI. We outline some of the opportunities and challenges with several examples, such as methodological and technological developments, necessary to develop this promising pan-disciplinary field. We emphasize the importance of educating researchers (i.e., NeuroDesign Researchers) and practitioners (neurodesigner/engineers) to advance this intersection toward a new area that could be greater than the sum of its parts.

1 Introduction In recent years, neuroscientists examined brain activity in the context of design, and at the same time, design researchers employed neuroscience instruments to examine the thinking of designers (Balters et al., 2023; Ohashi et al., 2022; Pidgeon et al., 2016). For neuroscientists, design provides an exciting context to examine various psychological and cognitive phenomena, such as figural or visual creativity. From a design perspective, neuroscience instruments and methodologies provide new opportunities to investigate design thinking/cognition. Adapting research J. Auernhammer (✉) Center for Design Research, Department of Mechanical Engineering, Stanford University, Stanford, CA, USA University of Technology Sydney, Sydney, NSW, Australia e-mail: [email protected] J. Bruno · A. Booras · C. McIntyre · D. Hasegan · M. Saggar Center for Interdisciplinary Brain Sciences Research, Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, USA © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_10

197

198

J. Auernhammer et al.

methodology from psychology, such as the think-aloud protocol and protocol analysis, allowed advancing the research on design thinking over the last halfcentury (Duncker, 1945; Eastman, 1970; Ericsson & Simon, 1998; Lawson, 1972). Neuroscientific approaches and instruments provide a similar opportunity to further advance research and an understanding of design thinking. However, there are several challenges to bringing these two cultures to meet productively (Auernhammer et al., 2021). The highly reductionist approach in neuroscience contrasts with the holistic or situational approach in design. Researchers need to make tradeoffs in their designs to either increase the spatial resolution of brain scans or to increase ecological validity, i.e., the natural environment in which designers design. However, a pan-disciplinary intersection of NeuroDesign provides the opportunity to go beyond current tradeoffs and approaches. Designing new technologies, such as nonintrusive near-infrared optical technology, and developing new analysis techniques provide opportunities to investigate design thinking and advance neuroscience approaches (Jöbsis, 1977; Maki et al., 1995; Saggar et al., 2018). Such technological and methodological developments need both the “neuro” (e.g., neuroscience) and “design” (e.g., engineering design) perspectives, creating a field that is greater than the sum of its parts. In this chapter, we outline some developments, challenges, and opportunities at the intersection of neuroscience and design. We believe that by fostering researchers that are at the intersection of both neuro (e.g., psychology, cognitive science, and anthropology) and design (e.g., engineering, product design, and computer science) disciplines, we can advance and contribute to both fields in new ways. For example, a better understanding of design thinking through neuroscience on design and design neurocognition provides the opportunity to advance design education. In this way, education advances design practices, which, in turn, results in more advanced products and services. More advanced, designed, and developed neuroscientific instruments allow new, meaningful investigations of the human brain. Therefore, this innovative advancement of NeuroDesign is autopoietic (Auernhammer, 2012; Auernhammer & Hall, 2014). A better understanding of design thinking and its practice leads to better neurotechnological products, which in turn helps investigate design thinking.

2 What Is NeuroDesign Research? NeuroDesign research is the unique combination of neuroscience and design research and practice. Similar to developments in neuroeconomics (Glimcher & Rustichini, 2004), NeuroDesign bridges disciplines of “neuro,” such as biology, neurophysiology, neurology, phenomenology, and psychology, and “design,” such as engineering, computer science, interaction, product, and visual design, as illustrated in Fig. 1. The intersection between “neuro” and “design” has existed for several decades. Early research on creativity and design thinking helped translate many psychological

NeuroDesign: Greater than the Sum of Its Parts

199

Fig. 1 NeuroDesign: The pan-disciplinary intersection of neuroscience and design

theories and principles into design practices (Arnold, 1959; Harman et al., 1966; Lawson, 1972; McKim, 1972). Sparked by the work of John Arnold (1954, 1959, 1962a, 1962b), Stanford’s Design Division (today Design Group1) has a long tradition of integrating psychological insights, principles, and theories into design education and practices to develop students’ potential (Adams, 2019; Arnold, 1959; Auernhammer & Roth, 2021; Fadiman, 1986; McKim, 1972; Wilde, 1972). Courses such as the ME101 Visual Thinking, Peopledynamics Lab, or ME211 Psychology of Design take an experimental approach to bring neuroscience and psychological insights into design education and practices (Bulletin, 2022; McKim, 1980; Wilde, 1972). One of the first scholars who bridged the two disciplines through an engineering and neurological approach was Larry Leifer at Stanford University in the 1960s. In his Ph.D. research on the Characterization of single muscle fiber discharge during voluntary isometric contraction of the biceps brachii muscle in man, supervised by Leon Cohen (neurology), James Bliss (electrical engineering), and Donald Wilson (biological sciences), Leifer (1969) investigated neurophysiological questions in the intersection of “neuro” and “design.” More recent developments in NeuroDesign at Stanford University have emerged in the Hasso Plattner Design Thinking Research (HPDTR) program from neuroscientific research on creative thinking in design. The collaborative efforts at Stanford between Manish Saggar and Alan Reiss from the neurosciences and Grace Hawthorne from the Hasso Plattner Institute of Design (dschool) produced work investigating figural creativity and the effects of design thinking training on creative capacity (Saggar et al., 2016; Saggar et al., 2015). Over the last two decades, research that utilized neuroscience instruments to examine various cognitive tasks and activities in design has emerged at various institutes (Alexiou et al., 2009; Jenkins et al., 2009; Petkar et al., 2009; Steinert & Jablokow, 2013; Sun et al., 2013). Notably, Steinert and Jablokow (2013) aimed to understand the relationships between engineering design behavior in situ, problem1

About the design group https://me.stanford.edu/groups/about-design-group.

200

J. Auernhammer et al.

Table 1 Opportunities of the pan-disciplinary intersection of NeuroDesign Neuroscience Neuroscience

Neuroscience Neuroscience Biology, neuroscience, phenomenology Neuroscience, psychology “Neuro”-disciplines

NeuroDesign Neuroscience on design and design neurocognition (Balters et al., 2023; Gero, 2019) Creative design education (e.g., Saggar et al., 2016) Neurotechnology (e.g., Maki et al., 1995; Zhang et al., 2020) Embodied artificial intelligence (e.g., Pfeifer & Iida, 2004) Augmenting human capabilities (e.g., Flesher et al., 2021) Further interesting intersections

Design Design thinking research

Design education Engineering Engineering (software, electronic, and mechanical), robotics, biomechanics, material sciences Human-centered design, product design, software engineering “Design”-disciplines

solving preference, and real-time physiological data of engineers measured through electroencephalogram (EEG). The study was executed in the design observatory at Stanford’s Center for Design Research. The neuroscience research on creativity in design within the HPDTR program sparked in 2018 the establishment of Stanford’s NeuroDesign Research.2 NeuroDesign emerged as a global community that included the Hasso Plattner Institute at the University of Potsdam (Germany), Tokyo Tech (Japan), Beijing Normal University (China), and several other institutions. The NeuroDesign symposia in Potsdam and California, and online seminars, advanced the conversations in this intersection further (von Thienen et al., 2021). In 2019, the special issue on design neurocognition, by John Gero, Kosa GoucherLambert, Tripp Shealy, and Yong Zeng, in the Design Science Journal provided space for stimulating publications (Fu et al., 2019; Hay et al., 2019; Hu et al., 2021; Shealy et al., 2020; Vieira et al., 2020; Zhao et al., 2020). Most of the research in this intersection focuses on using neuroscience instruments to examine cognitive tasks associated with the human activity of design (Balters et al., 2023; Ohashi et al., 2022). This area of the neuroscience on design and design neurocognition is one of the emerging fruitful spaces within the intersection of neuro and design. However, we emphasize that this pan-disciplinary intersection of NeuroDesign contributes to various other developments. NeuroDesign provides fruitful areas to advance our understanding of creativity, design thinking, and creative design education, by developing technologies to advance research and design products to augment human capabilities, as exemplified in Table 1. For the development of this pan-disciplinary intersection, we need to educate researchers, e.g., through Ph.D. programs, who will develop new research methodologies and enable practitioners (e.g., neurodesigners/engineers) to create new neurotechnologies (e.g., functional near-infrared spectroscopy (fNIRS) and 2

Stanford’s NeuroDesign Research https://neurodesign.stanford.edu.

NeuroDesign: Greater than the Sum of Its Parts

201

hyperscanning) and neuroproducts (e.g., mind-controlled robot arm). However, we must first tackle several challenges to take advantage of these opportunities. This means going beyond individual perspectives in the context of the other.

3 Opportunities and Challenges Developing the intersection of NeuroDesign beyond the application of neuroscience instruments to the design or the investigation of cognitive phenomena in the design context provides new opportunities and incorporates several challenges.

3.1

Neuroscience and Design Research on Cognition in Design

The two paradigms of “neuro” and “design” can be integrated in three different ways: (1) neuroscientists investigating cognitive tasks in the context of design (neuroscience of design), (2) design researchers using neuroscience instruments to investigate design thinking/cognition (design neurocognition), and (3) the development of new techniques and methodologies that advance both fields (NeuroDesign).

3.1.1

Neuroscience of Design

The neuroscience of design applies a neuroscience approach to the investigation of cognitive tasks, such as creativity, related to the fundamental human activity of design. Various neuroscience scholars examined figural creativity, brain synchronicity, and the generation and evaluation of ideas in the context of design (Ellamil et al., 2012; Jia & Zeng, 2021; Mayseless et al., 2019; Saggar et al., 2015). Depending on the neuroscientific instrument, these cognitive tasks are investigated through specific experimental designs. For example, functional magnetic resonance imaging (fMRI) studies primarily use experimental paradigms of two categories, namely, (1) block design (BD) or (2) event-related design (Chee et al., 2003). Figure 2 illustrates the investigation of creativity through a block design experimental paradigm. While such block designs allow superior statistical power relative to event-related designs, they reduce ecological validity. This is especially problematic when studying design thinking, which cannot be broken into discrete cognitive tasks. For example, sketching ideas and concepts is a fluent task in which one move informs the next (Bamberger & Schön, 1983; Goldschmidt, 1991, 2014; McKim, 1972; Schön, 1992). In such cases, event-related paradigms can be useful.

202

J. Auernhammer et al.

Fig. 2 Prototypical block design

The main idea of an event-related design (e.g., stimuli and response) is the separation of cognitive processes into discrete points in time (i.e., events), allowing differentiation of their associated fMRI signals (Huettel, 2012). However, eventrelated paradigms also present challenges. Designers often create their own stimuli that occur naturally rather than act on provided stimuli. These created stimuli (e.g., sketches) are seen in new ways as the “situation talks back” (Bamberger & Schön, 1983; Goldschmidt, 1991; Schön, 1992; Wertheimer, 1922, 1923). Thus, the intended experimental control and statistical power are diminished. NeuroDesign provides an opportunity for the development of new research paradigms that allow researchers to observe the neurocognitive activities of designers when they freely engage in the design task.

3.1.2

Design Neurocognition

In contrast to neuroscience on design, design neurocognition is the application of neuroscience instruments to investigate questions related to thinking/cognition associated with design activities. Design neurocognition emerged from earlier developments in research on creativity and design cognition (Gero, 2019; von Thienen et al., 2021). Design researchers examined neurocognitive activities in design through instruments such as EEG, fMRI, and fNIRS (Alexiou et al., 2009; Balters et al., 2023; Goucher-Lambert et al., 2018; Hay et al., 2019; Shealy et al., 2020; Vieira et al., 2020). For example, Goucher-Lambert et al. (2018, 2019) used an fMRI block design to investigate neural activity during successful and unsuccessful design solution generation. Other researchers studied brain activation in the prefrontal cortex of engineering students, while they utilized different design concept generation techniques such as TRIZ, brainstorming, and morphological analysis (Shealy et al., 2020). Many of the studies by design researchers utilized generalized linear modeling in fMRI and fNIRS studies to identify brain activation within specific regions and networks (Balters et al., 2023). However, such data analysis techniques model small portions of an expected hemodynamic response. Integrating more advanced

NeuroDesign: Greater than the Sum of Its Parts

203

techniques from neuroscience would advance research in design neurocognition networks (Balters et al., 2023). /.

3.1.3

Pan-Disciplinary NeuroDesign

Our novel approach is the development of new research techniques and methods by integrating perspectives and approaches from “neuro” and “design.” NeuroDesign takes a pan-disciplinary standpoint to overcome some challenges inherent in approaching the question from a neuroscience or design research perspective. We are exemplifying this third perspective based on our current study. We approached the investigation of creative and design thinking through an fMRI study in free flow, overcoming the limitations of the block design and event-related design paradigms to yield higher ecological validity. We also utilized the opportunity to investigate brain activation underlying conceptual design activities (i.e., concept sketching) in free flow with high spatial and temporal resolution through a multiband, multi-echo fMRI sequence. To study designers in a naturalistic setting (i.e., free flow) a major challenge is the collection and analysis of the data. We combined the perspectives, research methods, and analysis techniques from both neuroscience and design fields, and specifically measured brain activity during a free-flow design activities using sketching and screen capture video recording. Next, we applied a post-scan Think Aloud Protocol and video analysis of the sketching activities commonly employed in design thinking research (Eastman, 1970; Goldschmidt, 1991, 2014; Lawson, 1972, 1979; Lloyd et al., 1995). Then, we matched the fMRI-based brain activation time course with the time course of design activities. Finally, we used a recently developed data-driven method the topological data analysis to examine (1) the underlying manifold (or shape) of brain’s dynamical organization and (2) the transitions between states over time at the level of individual samples (or time frames) (Saggar et al., 2022; Saggar et al., 2018). We briefly present our approach in Fig. 3 and preliminary results from one participant in Fig. 4. Figure 4 shows how we can project the Mapper-generated manifold back to the time domain, to extract and match moment-to-moment transitions in design activity as well as activation in brain networks. This analysis could allow us to identify which brain networks are associated with specific heuristic design activities, such as a moment of insight or change in problem perspective. Overall, using this fMRI study design, we aim to investigate transition states between specific events that occurred naturally in the free flow, making it a novel approach in both fields. This pan-disciplinary development of new research designs, through such integrated data collection and analysis methods from diverse disciplines, is greater than the sum of its parts.

204

a

J. Auernhammer et al.

Data matrix

Preprocessed fMRI data

Projection/Filtering

Binning/Partial clustering

Graph construction

3D

bra ov in vo er tim lume s e

Time frames (TRs)

Features (ROIs/voxels)

Cluster represents three similar time frames

Auditory

Default mode

CinguloOperuclar

Retrospenial

Frontopariental

Dorsal attention

Salience network

Visual

Somatomotor

Mapper graph annotated by brain network activations

Ventral attention

Situation perspective Situation outline Selection of perspective

Inspiration

Mapper graph annotated by design thinking activity

Idea creation

Mapper graph for one participant

Idea sketching

b

Fig. 3 (a) Topological data analysis (TDA)-based approach, Mapper, which allows extracting the underlying low-dimensional manifold from spatiotemporally rich high-dimensional fMRI data at the single participant level, without averaging or collapsing data at the outset (Saggar et al., 2022; Saggar et al., 2018). (b) Mapper-generated manifold graph from one participant while engaged in the free-flow design task. We also present annotations of the manifold graph based on the design thinking activity and brain network activations

3.2

NeuroDesign to Advance Neurotechnology

Another fruitful area of NeuroDesign is the integration of neuroscience and product design/engineering. The technological development of noninvasive infrared and near-infrared topography allows researchers to examine changes in tissue blood volume and the average hemoglobin-oxyhemoglobin equilibrium (Jöbsis, 1977; Maki et al., 1995). For example, recording brain activity using near-infrared light to measure and visualize the pattern of hemodynamic changes in the cerebral cortex at the brain surface allowed neuroscientific research on infants in situ (Taga et al., 2003). Designers and engineers need to design from a neuroscience perspective to develop neurotechnologies and products (Maki, 2021). Designing from a neuroscience perspective is greater than the sum of bringing an engineering perspective to neuroscience and vice versa. These technological advancements allowed various neuroscientists and design researchers to investigate creative collaboration and sketching in design. For example, the fNIRS allowed Kato et al. (2018) to examine blood oxygenation level

NeuroDesign: Greater than the Sum of Its Parts

205

Fig. 4 Projecting annotated Mapper-generated graphs back to the time domain could allow us to identify which brain networks are associated with different design activity states

dependent (BOLD) in the prefrontal cortex during different sketching tasks. Others investigated the differences in prefrontal cortex activation in situ (Shealy et al., 2020; Shealy & Gero, 2019). These technological advancements sparked new research areas, such as interaction neuroscience, the neuroscience that investigates collaboration (Baker et al., 2016; Cui et al., 2012; Liu et al., 2016; Mayseless et al., 2019; Miller et al., 2019; Xie et al., 2020). For example, Mayseless et al. (2019) observed neural synchronicity in creative design tasks using fNIRS and hyperscanning. This pan-disciplinary approach to neuroproduct design and neurotechnology development is a great opportunity to advance neuroscience research and product design. NeuroDesign is greater than the sum of its parts.

3.3

Further Opportunities for NeuroDesign

The pan-disciplinary integration of the “neuro” and “design” disciplines incorporates several other productive areas, such as embodied and human-centered AI and augmenting human capabilities through neurotechnology. Artificial Intelligence NeuroDesign provides new opportunities in embodied AI, such as visual and haptic perception (Ullman, 1986, 2019). For example, neuromorphic engineering and resulting technologies represent a promising approach for the creation of robots that can seamlessly integrate into society

206

J. Auernhammer et al.

(Bartolozzi et al., 2022). NeuroDesign research also adds an additional productive perspective to other human-centered AI research approaches (Auernhammer, 2020). Intelligent systems have the potential to profoundly impact people and society. For example, a neurophysiological investigation through eye-tracking identified that individuals’ selective information search rather than algorithmic curation of search results might result in filter bubbles (Ekström et al., 2022). NeuroDesign research has the potential to evaluate various aspects of impact on the human psyche, behavior, and well-being, advancing practices in human-centered AI. Augmentation of Human Capabilities Developing new neurotechnology through a human-centered design approach provides opportunities to augment human capabilities (Engelbart, 1962). For example, brain-computer interfaces allow voluntary motor output from brain activity and the mimicry of sensory input from the skin of a hand (Flesher et al., 2021; Gerven et al., 2009; Green & Kalaska, 2011). Such NeuroDesign/engineering augments human capabilities, particularly for people with physical disabilities and injuries. In doing so, the pan-disciplinary intersection advances the field of design, such as the augmentation of human capabilities through robotics (Burgar et al., 2000; Engelbart, 1962). The pan-disciplinary intersection of NeuroDesign provides opportunities to integrate the cognitive disciplines of “neuro” and engineering and making disciplines of “design” to develop intelligent systems and advance human capabilities.

4 Developing the Pan-Disciplinary Field of NeuroDesign In this chapter, we illustrated that neurodisciplines and design-disciplines together form a prolific pan-disciplinary field that is greater than the sum of its parts. Developing this pan-disciplinary field of NeuroDesign provides many opportunities to advance design thinking, creativity, design education and practice, neuroscience research, artificially intelligent systems, and technologies to augment human capabilities. For such progress to happen, one of the main challenges is funding. For sustainable development and advancement of NeuroDesign research, education, and practice, large-scale funding (that spans several typical NSF/NIH grants) is required. Philanthropic and foundation grants and the national institutes need to come together to provide such critical funding.

4.1

NeuroDesign Research

To expand and develop NeuroDesign research, we envision pan-disciplinary research centers incorporating Ph.D. research programs. These programs would integrate classes from the neuro- and design-disciplines to develop the next generation of NeuroDesign researchers and engineers. Such efforts would incorporate

NeuroDesign: Greater than the Sum of Its Parts

207

various creative design and research practices, inspiring the development of new techniques, methodologies, and technologies. Bringing together practices and individuals in this way creates a pan-disciplinary research intersection greater than the sum of its parts.

4.2

NeuroDesign Education and Practice

The NeuroDesign centers and programs bring together individuals from the humanities, sciences, and the arts to develop design practices for the next century. Over the last century, psychology theories and principles advanced the ways we design today (Adams, 2019; Auernhammer & Roth, 2021, 2022; Card et al., 1983; Chapanis et al., 1949; Lawson, 2006; Norman, 1988). Similarly, advancements in NeuroDesign provide new educational principles, allowing the development of new creative design practices. Classes taught in collaborative teaching teams made up of neuroscientists, computer scientists, engineers, and creative designers allow for the development of experts across different fields. Individuals such as Larry Leifer (engineering design and neurology) from the late 1960s and young rising stars that combine multiple fields bridge the domains of “neuro” and “design” in new and meaningful ways. Training NeuroDesign researchers, designers, and engineers, who creatively approach methodological and technological challenges, is an essential part of advancing the pan-disciplinary intersection of NeuroDesign, creating a field that is something other than the sum of its parts. Acknowledgments We would like to thank the wider design and neuroscience community at Stanford University, in particular Professor Emeritus Larry Leifer and Professor Allan Reiss for their pioneering efforts in neurophysiology, neuroscience, engineering, and design.

References Adams, J. L. (2019). Conceptual blockbusting: A guide to better ideas (5th ed.). Basic Books. Alexiou, K., Zamenopoulos, T., Johnson, J. H., & Gilbert, S. J. (2009). Exploring the neurological basis of design cognition using brain imaging: Some preliminary results. Design Studies, 30(6), 623–647. https://doi.org/10.1016/j.destud.2009.05.002 Arnold, J. E. (1954). Selected summer session notes 1954. Massachusetts Institute of Technology. Arnold, J. E. (1959). Creative engineering seminar, 1959. Stanford, University. Arnold, J. E. (1962a). Education for innovation. In S. J. Parnes & H. F. Harding (Eds.), A source book for creative thinking. Charles Scribner’s Sons. Arnold, J. E. (1962b). Useful Creative Techniques. In S. J. Parnes & H. F. Harding (Eds.), A source book for creative thinking. Charles Scribner’s Sons. Auernhammer, J. M. (2012). Autopoietic organisation of knowledge, creativity and innovation : A case study of the automotive manufacturer Daimler AG (publication number 0000 0004 2742 4145) Edinburgh Napier, university. British Library. https://ethos.bl.uk/OrderDetails.do? did=1&uin=uk.bl.ethos.580699

208

J. Auernhammer et al.

Auernhammer, J. M. (2020). Human-centered AI: The role of Human-centered Design Research in the development of AI Synergy - DRS International Conference 2020, Held online. Auernhammer, J. M., & Hall, H. (2014). Organizational culture in knowledge creation, creativity and innovation: Towards the Freiraum model. Journal of Information Science, 40(2), 154–166. Auernhammer, J. M., & Roth, B. (2021). The origin and evolution of Stanford University’s design thinking: From product design to design thinking in innovation management. Journal of Product Innovation Management, 38, 623–644. https://doi.org/10.1111/jpim.12594 Auernhammer, J. M., & Roth, B. (2022). Different types of productive thinking in design: From rational to social design thinking. In C. Meinel & L. Leifer (Eds.), Design thinking research: Achieving real innovation (Vol. 13). Springer Nature. Auernhammer, J. M., Sonalkar, N., & Saggar, M. (2021). NeuroDesign: From neuroscience research to design thinking practice. In C. Meinel & L. Leifer (Eds.), Design thinking research : Interrogating the doing (pp. 347–355). Springer International Publishing. https://doi.org/10. 1007/978-3-030-62037-0_16 Baker, J. M., Liu, N., Cui, X., Vrticka, P., Saggar, M., Hosseini, S. M. H., & Reiss, A. L. (2016). Sex differences in neural and behavioral signatures of cooperation revealed by fNIRS hyperscanning. Scientific Reports, 6(1), 26492. https://doi.org/10.1038/srep26492 Balters, S., Weinstein, T., Mayseless, N., Auernhammer, J., Hawthorne, G., Steinert, M., et al. (2023). Design science and neuroscience: A systematic review of the emergent field of design neurocognition. Design Studies, 84, 101148. https://doi.org/10.1016/j.destud.2022.101148 Bamberger, J., & Schön, D. A. (1983). Learning as reflective conversation with materials: Notes from work in Progress. Art Education, 36(2), 68–73. https://doi.org/10.2307/3192667 Bartolozzi, C., Indiveri, G., & Donati, E. (2022). Embodied neuromorphic intelligence. Nature. Communications, 13(1), 1024. https://doi.org/10.1038/s41467-022-28487-2 Bulletin, S. (2022). ME 211: Psychology of design: Experience and thinking about thinking in design. Stanford University. Retrieved 25. November from. Burgar, C. G., Lum, P. S., Shor, P. C., & Van der Loos, H. F. M. (2000). Development of robots for rehabilitation therapy: The Palo Alto VA/Stanford experience. Journal of Rehabilitation Research and Development, 37(6), 663–674. Card, S. K., Moran, T. P., & Newell, A. (1983). The psychology of human-computer interaction. L. Erlbaum Associates. Chapanis, A., Garner, W. R., & Morgan, C. T. (1949). Applied experimental psychology: Human factors in engineering design. John Wiley & Sons Inc. https://doi.org/10.1037/11152-000 Chee, M. W., Venkatraman, V., Westphal, C., & Siong, S. C. (2003). Comparison of block and event-related fMRI designs in evaluating the word-frequency effect. Human Brain Mapping, 18(3), 186–193. https://doi.org/10.1002/hbm.10092 Cui, X., Bryant, D. M., & Reiss, A. L. (2012). NIRS-based hyperscanning reveals increased interpersonal coherence in superior frontal cortex during cooperation. NeuroImage, 59(3), 2430–2437. https://doi.org/10.1016/j.neuroimage.2011.09.003 Duncker, K. (1945). On Problem-solving. American Psychological Association. Eastman, C. (1970). On the analysis of intuitive design processes. In G. T. Moore (Ed.), Emerging methods in environmental design and planning. MIT Press. Ekström, A. G., Niehorster, D. C., & Olsson, E. J. (2022). Self-imposed filter bubbles: Selective attention and exposure in online search. Computers in Human Behavior Reports, 7, 100226. https://doi.org/10.1016/j.chbr.2022.100226 Ellamil, M., Dobson, C., Beeman, M., & Christoff, K. (2012). Evaluative and generative modes of thought during the creative process. NeuroImage, 59(2), 1783–1794. https://doi.org/10.1016/j. neuroimage.2011.08.008 Engelbart, D. C. (1962). Augmenting human intellect: A conceptual framework. Menlo Park. Ericsson, K. A., & Simon, H. A. (1998). How to study thinking in everyday life: Contrasting thinkaloud protocols with descriptions and explanations of thinking. Mind, Culture, and Activity, 5(3), 178–186. https://doi.org/10.1207/s15327884mca0503_3 Fadiman, J. (1986). Omega: Be all that you are. Westlake Press.

NeuroDesign: Greater than the Sum of Its Parts

209

Flesher, S. N., Downey, J. E., Weiss, J. M., Hughes, C. L., Herrera, A. J., Tyler-Kabara, E. C., et al. (2021). A brain-computer interface that evokes tactile sensations improves robotic arm control. Science, 372(6544), 831–836. https://doi.org/10.1126/science.abd0380 Fu, K. K., Sylcott, B., & Das, K. (2019). Using fMRI to deepen our understanding of design fixation. Design. Science, 5, e22. Article e22. https://doi.org/10.1017/dsj.2019.21 Gero, J. S. (2019). 2019. From Design Cognition to Design Neurocognition. Gerven, M. V., Farquhar, J., Schaefer, R., Vlek, R., Geuze, J., Nijholt, A., et al. (2009). The brain– computer interface cycle. Journal of Neural Engineering, 6(4), 041001–041001. https://doi.org/ 10.1088/1741-2560/6/4/041001 Glimcher, P. W., & Rustichini, A. (2004). Neuroeconomics: The consilience of brain and decision. Science, 306(5695), 447–452. https://doi.org/10.1126/science.1102566 Goldschmidt, G. (1991). The dialectics of sketching. Creativity Research Journal, 4(2), 123–143. https://doi.org/10.1080/10400419109534381 Goldschmidt, G. (2014). Linkography unfolding the design process. The MIT Press. https://doi.org/ 10.2307/j.ctt9qf7d9 Goucher-Lambert, K., Moss, J., & Cagan, J. (2018). Inspired internal search: Using neuroimaging to understand design ideation and concept generation with inspirational stimuli. https://doi.org/ 10.1115/DETC2018-85690 Goucher-Lambert, K., Moss, J., & Cagan, J. (2019). Unsuccessful external search: Using neuroimaging to understand fruitless periods of design ideation involving inspirational stimuli. Design Computing and Cognition ’18, Cham. Green, A. M., & Kalaska, J. F. (2011). Learning to move machines with the mind. Trends in Neurosciences, 34(2), 61–75. https://doi.org/10.1016/j.tins.2010.11.003 Harman, W. W., McKim, R. H., Mogar, R. E., Fadiman, J., & Stolaroff, M. J. (1966). Psychedelic agents in creative problem-solving: A pilot study. Psychological Reports, 19(1), 211–227. https://doi.org/10.2466/pr0.1966.19.1.211 Hay, L., Duffy, A. H. B., Gilbert, S. J., Lyall, L., Campbell, G., Coyle, D., & Grealy, M. A. (2019). The neural correlates of ideation in product design engineering practitioners. Design Science, 5, e29., Article e29. https://doi.org/10.1017/dsj.2019.27 Hu, M., Shealy, T., & Milovanovic, J. (2021). Cognitive differences among first-year and senior engineering students when generating design solutions with and without additional dimensions of sustainability. Design Science, 7, e1., Article e1. https://doi.org/10.1017/dsj.2021.3 Huettel, S. A. (2012). Event-related fMRI in cognition. NeuroImage, 62(2), 1152–1156. https://doi. org/10.1016/j.neuroimage.2011.08.113 Jenkins, S., Brown, R., & Rutterford, N. (2009). Comparing thermographic, EEG, and subjective measures of affective experience during simulated product interactions. International Journal of Design, 3(2), 2009. Jia, W., & Zeng, Y. (2021). EEG signals respond differently to idea generation, idea evolution and evaluation in a loosely controlled creativity experiment. Scientific Reports, 11(1), 2119. https:// doi.org/10.1038/s41598-021-81655-0 Jöbsis, F. F. (1977). Noninvasive, infrared monitoring of cerebral and myocardial oxygen sufficiency and circulatory parameters. Science, 198(4323), 1264–1267. https://doi.org/10.1126/ science.929199 Kato, T., Okada, H., & Izu, Y. (2018). Measurement of brain activities of idea generation (sketch) DESIGN 2018 15th international design conference. Dubrovnik. Lawson, B. (1972). Problem solving in architectural design Aston University]. . Lawson, B. (1979). Cognitive strategies in architectural design. Ergonomics, 22(1), 59–68. https:// doi.org/10.1080/00140137908924589 Lawson, B. (2006). How designers think: The design process demystified. Elsevier/Architectural. Leifer, L. J. (1969). Characterization of single muscle fiber discharge during voluntary isometric contraction of the biceps brachii muscle in man. Stanford University.

210

J. Auernhammer et al.

Liu, N., Mok, C., Witt, E. E., Pradhan, A. H., Chen, J. E., & Reiss, A. L. (2016). NIRS-based Hyperscanning reveals inter-brain neural synchronization during cooperative Jenga game with face-to-face communication [original research]. Frontiers in Human Neuroscience, 10. Lloyd, P., Lawson, B., & Scott, P. (1995). Can concurrent verbalization reveal design cognition? Design Studies, 16(2), 237–259. https://doi.org/10.1016/0142-694X(94)00011-2 Maki, A. (2021). Designing from a neuroscience perspective. Stanford NeuroDesign Research. Retrieved November 20th from https://www.youtube.com/watch?v=cyq7zQRt6C4 Maki, A., Yamashita, Y., Ito, Y., Watanabe, E., Mayanagi, Y., & Koizumi, H. (1995). Spatial and temporal analysis of human motor activity using noninvasive NIR topography. Medical Physics, 22(12), 1997–2005. https://doi.org/10.1118/1.597496 Mayseless, N., Hawthorne, G., & Reiss, A. L. (2019). Real-life creative problem solving in teams: fNIRS based hyperscanning study. NeuroImage, 203, 116161. https://doi.org/10.1016/j. neuroimage.2019.116161 McKim, R. H. (1972). Experiences in visual thinking. Harvard Business School. McKim, R. H. (1980). Experiences in visual thinking. Brooks/Cole Publishing Company. Miller, J. G., Vrtička, P., Cui, X., Shrestha, S., Hosseini, S. M. H., Baker, J. M., & Reiss, A. L. (2019). Inter-brain synchrony in mother-child dyads during cooperation: An fNIRS hyperscanning study. Neuropsychologia, 124, 117–124. https://doi.org/10.1016/j. neuropsychologia.2018.12.021 Norman, D. A. (1988). The psychology of everyday things. Basic Books. Ohashi, T., Auernhammer, J., Liu, W., Pan, W., & Leifer, L. (2022). NeuroDesignScience: Systematic literature review of current research on design using neuroscience techniques. Design Computing and Cognition’20. Petkar, H., Dande, S., Yadav, R., Zeng, Y., & Nguyen, T. A. (2009). A pilot study to assess Designer’s mental stress using eye gaze system and electroencephalogram. https://doi.org/10. 1115/DETC2009-86542 Pfeifer, R., & Iida, F. (2004). Embodied artificial intelligence: Trends and challenges. In F. Iida, R. Pfeifer, L. Steels, & Y. Kuniyoshi (Eds.), Embodied artificial intelligence: International seminar, Dagstuhl Castle, Germany, July 7–11, 2003. Revised papers (pp. 1–26). Springer. https://doi.org/10.1007/978-3-540-27833-7_1 Pidgeon, L. M., Grealy, M., Duffy, A. H. B., Hay, L., McTeague, C., Vuletic, T., et al. (2016). Functional neuroimaging of visual creativity: A systematic review and meta-analysis. Brain and behavior, 6(10), e00540. https://doi.org/10.1002/brb3.540 Saggar, M., Quintin, E.-M., Bott, N. T., Kienitz, E., Chien, Y.-H., Hong, D. W. C., et al. (2016). Changes in brain activation associated with spontaneous improvization and figural creativity after design-thinking-based training: A longitudinal fMRI study. Cerebral Cortex, 27(7), 3542–3552. https://doi.org/10.1093/cercor/bhw171 Saggar, M., Quintin, E.-M., Kienitz, E., Bott, N. T., Sun, Z., Hong, W.-C., et al. (2015). Pictionarybased fMRI paradigm to study the neural correlates of spontaneous improvisation and figural creativity. Scientific Reports, 5(1), 10894. https://doi.org/10.1038/srep10894 Saggar, M., Shine, J. M., Liégeois, R., Dosenbach, N. U. F., & Fair, D. (2022). Precision dynamical mapping using topological data analysis reveals a hub-like transition state at rest. Nature Communications, 13(1), 4791. https://doi.org/10.1038/s41467-022-32381-2 Saggar, M., Sporns, O., Gonzalez-Castillo, J., Bandettini, P. A., Carlsson, G., Glover, G., & Reiss, A. L. (2018). Towards a new approach to reveal dynamical organization of the brain using topological data analysis. Nature Communications, 9(1), 1399. https://doi.org/10.1038/s41467018-03664-4 Schön, D. A. (1992). Designing as reflective conversation with the materials of a design situation. Knowledge-Based Systems, 5(1), 3–14. https://doi.org/10.1016/0950-7051(92)90020-G Shealy, T., Gero, J., Hu, M., & Milovanovic, J. (2020). Concept generation techniques change patterns of brain activation during engineering design. Design Science, 6, e31., Article e31. https://doi.org/10.1017/dsj.2020.30

NeuroDesign: Greater than the Sum of Its Parts

211

Shealy, T., & Gero, J. S. (2019). The neurocognition of three engineering concept generation techniques. Proceedings of the Design Society: International Conference on Engineering Design, 1(1), 1833–1842. https://doi.org/10.1017/dsi.2019.189 Steinert, M., & Jablokow, K. (2013). Triangulating front end engineering design activities with physiology data and psychological preferences. In DS 75–7: Proceedings of the 19th international conference on engineering design (ICED13), Seoul, Keorea. Sun, L., Xiang, W., Chai, C., Wang, C., & Liu, Z. (2013). Impact of text on idea generation: An electroencephalography study. International Journal of Technology and Design Education, 23(4), 1047–1062. https://doi.org/10.1007/s10798-013-9237-9 Taga, G., Asakawa, K., Maki, A., Konishi, Y., & Koizumi, H. (2003). Brain imaging in awake infants by near-infrared optical topography. Proceedings of the National Academy of Sciences, 100(19), 10722–10727. https://doi.org/10.1073/pnas.1932552100 Ullman, S. (1986). Artificial intelligence and the brain: Computational studies of the visual system. Annual Review of Neuroscience, 9(1), 1–26. https://doi.org/10.1146/annurev.ne.09.030186. 000245 Ullman, S. (2019). Using neuroscience to develop artificial intelligence. Science, 363(6428), 692–693. https://doi.org/10.1126/science.aau6595 Vieira, S., Gero, J. S., Delmoral, J., Gattol, V., Fernandes, C., Parente, M., & Fernandes, A. A. (2020). The neurophysiological activations of mechanical engineers and industrial designers while designing and problem-solving. Design. Science, 6, e26., Article e26. https://doi.org/10. 1017/dsj.2020.26 von Thienen, J. P. A., Szymanski, C., Santuber, J., Plank, I. S., Rahman, S., Weinstein, T., et al. (2021). Neurodesign Live. In C. Meinel & L. Leifer (Eds.), Design thinking research: Interrogating the doing (pp. 357–425). Springer International Publishing. https://doi.org/10.1007/9783-030-62037-0_17 Wertheimer, M. (1922). Untersuchungen zur Lehre von der Gestalt. Psychologische Forschung, 1(1), 47–58. https://doi.org/10.1007/BF00410385 Wertheimer, M. (1923). Untersuchungen zur Lehre von der Gestalt II. Psycologische Forschung, 4, 301–350. Wilde, D. (1972). Gestalt learning in an engineering class American Society of Engineering Education. Educational Research Methods. Xie, H., Karipidis, I. I., Howell, A., Schreier, M., Sheau, K. E., Manchanda, M. K., et al. (2020). Finding the neural correlates of collaboration using a three-person fMRI hyperscanning paradigm. Proceedings of the National Academy of Sciences, 117(37), 23066–23072. https://doi. org/10.1073/pnas.1917407117 Zhang, W., Gao, B., Tang, J., Yao, P., Yu, S., Chang, M.-F., et al. (2020). Neuro-inspired computing chips. Nature Electronics, 3(7), 371–382. https://doi.org/10.1038/s41928-0200435-7 Zhao, M., Jia, W., Yang, D., Nguyen, P., Nguyen, T. A., & Zeng, Y. (2020). A tEEG framework for studying designer’s cognitive and affective states. Design. Science, 6, e29., Article e29. https:// doi.org/10.1017/dsj.2020.28

A Neuroscience Approach to Women Entrepreneurs’ Pitch Performance: Impact of Inter-Brain Synchrony on Investment Decisions Stephanie Balters, Sohvi Heaton, and Allan L. Reiss

Abstract Making a successful pitch to investors is vital to the success of startups. Improving pitch performance in women entrepreneurs might be an effective mechanism to close gender disparity in entrepreneurship. Drawing on social neuroscience studies, we present our scientific approach to shedding light on the role of “interbrain synchrony” between women entrepreneurs and investors in pitch performance. Using functional near-infrared spectroscopy (fNIRS) hyperscanning, we will scan 40 entrepreneur-investor dyads who engage in naturalistic pitch events. We will elucidate patterns of inter-brain synchrony that are associated with pitch performance. Additionally, we will assess whether the sex composition of an entrepreneurinvestor dyad affects these associations. A better understanding of the inter-brain signatures underlying successful (and unsuccessful) pitches will generate insights into the design of novel and effective interventions that can help catalyze the success of women entrepreneurs.

Potential Managerial Implications Elucidating the underlying inter-brain mechanisms associated with strong pitch performance will generate practical implications for women entrepreneurs. Our empirical findings will help craft more effective behavioral interventions (e.g., belonging [Walton et al., 2020] and interpersonal trust interventions [Balters et al., 2022, 2023]) and pitch training (e.g., Clingingsmith et al., 2022) tailored to improve pitch performance of women entrepreneurs. The results of this study can also inform the development of fNIRS neurofeedback S. Balters (✉) Department of Psychiatry and Behavioral Sciences, Stanford University, Stanford, CA, USA e-mail: [email protected] S. Heaton Santa Clara University, Leavey School of Business, Santa Clara, CA, USA A. L. Reiss Department of Psychiatry and Behavioral Sciences, Stanford University, Stanford, CA, USA Department of Pediatrics, Stanford University, Stanford, CA, USA Department of Radiology, Stanford University, Stanford, CA, USA © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_11

213

214

S. Balters et al.

paradigms (e.g., Liu et al., 2016a, b) that can provide entrepreneurs and investors with vital information to voluntarily regulate their behavior in real time. Lastly, our neuroscience-based approach may offer a new, powerful way to better match startups with investors, thereby enhancing the efficiency of the matching process.

1 Introduction Making a successful pitch to investors is vital for entrepreneurs across all business stages (Chen et al., 2009; Drover et al., 2017). However, only a small number of business ventures successfully compete for investor funding. These difficulties are magnified for women-led startups (Balachandra et al., 2021; McSweeney et al., 2022; Sanchez-Ruiz et al., 2021). Only 2.3% of venture capital investment went to women founders in 2020 (Cruncchbase, 2020), although companies founded by women deliver higher returns on investment—more than twice as much per dollar invested as those by men (Boston Consulting Group, 2018). Besides the missed opportunity of improving global gross domestic product by 2–3% of global GDP or $2 trillion (Miao, 2022), we are wasting half our genetic pool of innovative intelligence. Catalyzing pitch performance is thus a vital mechanism in fostering the success of women entrepreneurs. Prior research on pitch performance has put forward various factors that may influence pitch performance, including individual attributes (e.g., founders’ passion [Shane et al., 2020] and confidence [Sanchez-Ruiz et al., 2021]) and team characteristics (e.g., team diversity [Foo et al., 2005]). Several studies investigated entrepreneur-investor dyads and examined whether specific traits in entrepreneurs (e.g., personality, experience, social capital) influence investor decision-making (Murnieks et al., 2011; Franke et al., 2008). Sociopsychological phenomena such as similarity effect (Murnieks et al., 2011) and opinion conformity (Sanchez-Ruiz et al., 2021) have been argued to influence pitch performance. A growing number of empirical studies have further investigated the effects of gender on pitch performance and demonstrated mixed findings. Some studies found gendered differences in the level of assertiveness (McSweeney et al., 2022) and self-promotion (SanchezRuiz et al., 2021), as well as rhetorical strategies (Balachandra et al., 2021) that worked against the pitch performance of women. On the contrary, other researchers showed no gender difference in pitch performance (Balachandra et al., 2019; Hohl et al., 2021). The existing studies highlight the social complexity of pitches that involve many conscious and subconscious social variables. While emerging studies have utilized behavioral assessments to evaluate gendered pitch performance, little is known about the inter-brain mechanisms underlying entrepreneur-investor pitches. The current study fills this research gap. Specifically, we leverage the emerging technology of fNIRS hyperscanning (Fig. 1) to measure the inter-brain dynamics of an entrepreneur-investor dyad during a naturalistic pitch. Our aims are (1) to elucidate inter-brain mechanisms that are associated with high (or low) levels of pitch

A Neuroscience Approach to Women Entrepreneurs’ Pitch Performance:. . .

215

Fig. 1 Experimental setup of the fNIRS hyperscanning study. Entrepreneur-investor dyads engage in a naturalistic pitch event

performance and (2) to assess whether the sex composition of an entrepreneurinvestor dyad impacts these inter-brain mechanisms.

2 fNIRS Hyperscanning to Understand Inter-Brain Mechanisms during Entrepreneur-Investor Pitches Functional NIRS is an optical brain imaging technology that measures changes in cortical oxygenation as a proxy for neural activation (Cui et al., 2010; Strangman et al., 2002). In recent years, fNIRS systems have become increasingly portable and affordable, allowing researchers to investigate neurocognitive behavior in real-world settings (e.g., Bruno et al., 2018). In contrast to electroencephalography (EEG), another portable brain imaging technology, fNIRS is relatively robust to motion artifacts and has a relatively high spatial resolution (~1 cm; Cui et al., 2010). Researchers have extended fNIRS measurements from single-brain to inter-brain applications (i.e., “fNIRS hyperscanning”) to investigate shared brain functions underlying social interactions (Cui et al., 2012; Funane et al., 2011). A common approach has been to study when and how neural processes become synchronized and how inter-brain synchrony (“IBS,” i.e., correlation of cortical activity between brains) relates to behavioral measures of interaction (Babiloni & Astolfi, 2014; Balters et al., 2020). Numerous studies have shown the role of IBS in various behavioral and performance outcomes. Studies have found that increased IBS is associated with synchronously executed activities such as joint limb movement (Holper et al., 2012; Niu

216

S. Balters et al.

et al., 2019; Nozawa et al., 2019), coordinated button presses (Cheng et al., 2015; Cui et al., 2012; Funane et al., 2011), singing (Osaka et al., 2014, 2015), drumming (Duan et al., 2015), and marching (Ikeda et al., 2017). Beyond the mere engagement in the same activity, it appears that shared attention is essential for the occurrence of IBS. Studies have found increased IBS when pairs completed a task together as opposed to completing the identical task individually in parallel (Feng et al., 2020; Fishburn et al., 2018; Hu et al., 2017; Liu et al., 2016b; Zhou et al., 2022). In addition to joint attention, research shows that joint goal-directed intention is related to IBS (Kruse et al., 2021), and dyads demonstrate more IBS when cooperating than when competing with one another (Cui et al., 2012; T. Liu et al., 2017; Lu et al., 2019). Increased IBS has also been observed in other prosocial interaction contexts, such as in-group bonding (Yang et al., 2020), dyadic empathy (Bembich et al., 2022), and after gift exchanges (Balconi et al., 2019). The literature on IBS also provides clues as to whether the sex composition of the interacting entrepreneur-investor dyad could influence pitch performance. For example, Baker et al. (2016) and Cheng et al. (2015) conducted wavelet synchrony analyses on fNIRS hyperscanning data for dyadic cooperation tasks and found that inter-brain synchrony is highly dependent on the sex composition of the dyad. Differences in fNIRS neural signatures in association with the sex composition of a dyad also emerged during spontaneous face-to-face deception (Zhang et al., 2017b), risky decision-making during gambling games (Zhang et al., 2017a, b), and group creative idea generation (Lu et al., 2020). Given the combined findings, we expect that IBS is associated with pitch performance or its correlates (e.g., cooperative behavior) and might alter depending on the sex composition of an entrepreneur-investor dyad.

3 Dynamic Inter-Brain Biomarkers of Pitch Performance A pitch is a highly dynamic social interaction and likely requires a dynamic analysis approach to uncover associations between IBS and pitch performance. Researchers have traditionally utilized task-averaging approaches to assess mean levels of IBS (e.g., “over a task duration of 10 minutes”; Balters et al., 2020). A limitation of these approaches is that they do not capture the dynamic nature of social interaction. Our research group therefore developed a novel “dynamic IBS” approach (Li et al., 2021). This approach allows the study of inter-brain synchrony patterns (i.e., “inter-brain states”) with higher temporal resolution. Inter-brain synchrony can simultaneously occur between and across functional regions at any given time, and the dynamic IBS approach accounts for this complexity. It derives inter-brain states characterized by a “heatmap” of IBS values for all possible region of interest (ROI) pair combinations. In our recent work (Balters et al., 2022, 2023), we linked these inter-brain states for the first time to behavioral measures of cooperation (e.g., conversational turn-taking behavior). As shown in Fig. 2, we achieved the identification of inter-brain states that were significantly associated with adverse behavioral

Inter-brain state 1

Occurrence rate

1.51

217

1

0

1.51

-1.51

In-person group Virtual group

Time (8 min)

-1.51 Inter-brain state 2

Occurrence rate

Inter-brain synchrony

Inter-brain synchrony

A Neuroscience Approach to Women Entrepreneurs’ Pitch Performance:. . .

1

In-person group Virtual group

0 Time (8 min)

Fig. 2 Using fNIRS hyperscanning, we studied the differences in IBS states between in-person and virtual interaction during a problem-solving task (Balters et al., 2022, 2023). Dyads interacted either in person or over Zoom® video conferencing. Results showed that the occurrence rate of inter-brain state 1 was associated with a reduction in behavioral cooperation (r = -0.52). In contrast, the occurrence rate of inter-brain state 2 was associated with an increase in behavioral cooperation (r = 0.37). Due to the dynamic character of the analytical approach, we can observe the occurrence of each inter-brain state over time and identify temporal differences between groups (left)

cooperation (i.e., inter-brain state 1) and positive behavioral cooperation (i.e., interbrain state 2). Notably, the dynamic IBS analytical approach allows researchers to observe the dynamics within and between brain states over time. It is thus possible to study the functional characteristics of each brain state and identify “critical events” that can be linked to changes in behavior through video coding, for example. Due to dynamic characteristics, we could elucidate differences in the occurrence of interbrain states between dyads that interacted in person and over Zoom® during a problem-solving task (Fig. 2; Balters et al., 2022, 2023). We leverage this cuttingedge analytical approach in the present study to understand the impact of inter-brain states on investment decisions.

4 Methodology 4.1

Participants

A total of N = 40 female entrepreneurs and N = 40 investors (N = 20 females, N = 20 males) will participate in the study. All participants will be right-handed and healthy, with normal or corrected to normal hearing and vision. The age range will be limited to 18–50 years to avoid aging-related chronic diseases. We will recruit participants through our collaborating entrepreneurship centers and VC organizations at Stanford and the broader Silicon Valley. For this first proof-of-concept study, we will focus on informal investors and student entrepreneurs due to the cost and challenges of obtaining experts. It is common for exploratory studies to utilize naive subjects (Shane et al., 2020; Chen et al., 2009; Hsu et al., 2017). We will

218

S. Balters et al.

Table 1 Study design will follow a between-subject design with 20 dyads per group

Female entrepreneur

Female investor Group 1 (20)

Male investor Group 2 (20)

randomly match female entrepreneurs with either a female investor (group 1) or a male investor (group 2). Each group includes 20 dyads (Table 1). Age and race/ ethnicity will not be matched across participants. We will therefore utilize both variables as covariates in statistical analyses. Participants of the same dyad will be previously unacquainted. We will recruit participants through email lists and social media and obtain written consent before the study. Stanford’s Institutional Review Board has approved the experimental methodology (IRB #18160).

4.2

Experimental Procedure

The dyad partners will interact face-to-face in a presentation setting (Fig. 1). We will attach fNIRS caps while participants are in the same room and instruct participants not to talk to one another during that time. Before starting the experiment, participants will have 3 min to introduce themselves. During the pitch, participants will be alone in the room and receive task instructions via audio prompts. We will collect audio and video recordings of the participants, with four portable video cameras capturing front and side views of both participants. After the experiment, each participant will complete a post-experimental survey in a separate room to assess investment interest, pitch performance, subjective experiences during the pitch, and demographical information.

4.3

Experimental Tasks

Each entrepreneur will have 45 minutes to prepare for their pitch immediately prior to the study. Entrepreneurs will make a 10-min pitch based on their pitch script followed by a 5-min Q&A session. The pitch will be entirely oral, with no presentation tools allowed. Entrepreneurs and investors will receive the same task instruction.

4.4

Functional NIRS Hyperscanning Data Acquisition and Processing

Data Acquisition We will record the cortical hemodynamic activity of each participant using a continuous wave fNIRS system (NIRSport2 System, NIRX,

A Neuroscience Approach to Women Entrepreneurs’ Pitch Performance:. . .

219

Germany) with two wavelengths (760 and 850 mm) and a sampling frequency of 10.2 Hz. We will divide a total of 128 optodes (64 sources x 64 detectors) between the 2 participants resulting in 100 measurement channels per participant. Optodes will spread over the entire cortex according to the international 10–20 EEG placement system. Additionally, we will place 16 short channels per participant across the cortex to capture and correct background physiological noise (e.g., cardiac, respiratory, and blood pressure fluctuations). We will utilize plastic connectors between each source/detector channel pair to maintain an estimated 3 cm distance. Data Preprocessing We will analyze the raw fNIRS data using the NIRS Brain AnalyzIR Toolbox (Santosa et al., 2018) in Matlab version R2021a (MathWorks, Inc.). We will assess data quality via the scalp coupling index or “SCI” (Pollonini et al., 2016) and exclude the channels with excessive noise (i.e., SCI ≤ 0.8) from subsequent analyses to ensure good data quality. We will convert the remaining raw data to optical density data and apply motion artifacts correction using a wavelet motion correction procedure (Molavi & Dumont, 2012). Subsequently, we will transform data into concentration changes of oxygenated hemoglobin (HbO) and deoxygenated hemoglobin (HbR) according to the modified Beer-Lambert law (Wyatt et al., 1986). Since HbO and HbR data are relative values, we will convert resulting data to z-scores. We will create 32 ROIs via source localization (Huppert et al., 2017) by averaging all channels that shared a common fNIRS source (Balters et al., 2022). We will project all ROIs onto the cortical surface using an automatic anatomical labeling method (Lancaster et al., 2000; Singh et al., 2005). Because HbO measures are known to be more robust and sensitive to task-associated changes compared to HbR measures (Ferrari & Quaresima, 2012; Plichta et al., 2006), we will only use HbO data for further analyses as common in fNIRS hyperscanning research (Balters et al., 2020). Dynamic Inter-Brain State Analyses We will use wavelet transform synchrony or “WTC” (Cui et al., 2012) analysis to assess averaged inter-brain synchrony (IBS), i.e., the similarity between NIRS signals of dyad partners. For a more in-depth explanation of WTC, please see (Grinsted et al., 2004). Specifically, we will calculate IBS between each ROI and the rest of the ROIs on the converted HbO time series (a total of 1024 combinations: 32 ROIs x 32 ROIs). We will then average the IBS between the same ROI pairings resulting in 528 ROI pairings. We will calculate the average synchrony value between 0.15 and 0.02 Hz. This frequency band will allow us to exclude noise associated with cardiac pulsation (about 1 Hz) and respiration (0.2–0.3 Hz [Molavi & Dumont, 2012]). Finally, we will apply the dynamic IBS approach (Li et al., 2021) following the procedures described in Balters et al. (2022, 2023).

220

4.5

S. Balters et al.

Additional Assessments during the Experiment

Investment Interest Following Shane et al. (2020, p.8), we will measure investment interest as pitch performance. Investors rate five sub-scores on a seven-point Likert scale: (1) “I would be interested in seeing more information about this venture,” ranging from “strongly disagree” to “strongly agree”; (2) “Based on the information at hand, I would be interested in investing in this company,” ranging from “strongly disagree” to “strongly agree”; (3) “This company represents a good investment opportunity for me,” ranging from “strongly disagree” to “strongly agree”; (4) “I would expect higher financial returns from investing in this company than in other startup companies,” ranging from “strongly disagree” to “strongly agree”; and (5) “The content of this elevator pitch was,” ranging from “very poor” to “excellent.” We will calculate the final score by averaging the five sub-scores. Pitch Performance Entrepreneurs and investors will rate their subjective experience of pitch performance on a nine-point Likert scale ranging from “extremely poor” to “extremely good.” Level of Cooperation Participants will rate the overall cooperation of the dyad on a nine-point Likert scale ranging from “not at all cooperative” to “extremely cooperative.” Affective Responses Participants will complete the Affect Grid Survey (Russell, 1980) to inquire about their level of arousal and level of valence on a nine-point Likert scale ranging from “sleepy” to “energized” and “unpleasant” to “pleasant,” respectively. We will also obtain self-reported levels of stress via the Perceived Stress Scale (Cohen et al., 1994), using a nine-point Likert scale ranging from “low” to “high.” Interpersonal Closeness Index Participants will rate their subjective sense of closeness toward their dyad partner (i.e., “Interpersonal Closeness”) on five sevenpoint Likert subscales, including questions about connectedness and trust (Wiltermuth & Heath, 2009), an adapted version of the inclusion of other in selfscale (Aron et al., 1992), likeability (Hove & Risen, 2009), and similarity in personality (Valdesolo & DeSteno, 2011). We will calculate the final score by averaging the sub-scores. Coded Behavioral Metrics We will capture videos of the dyadic interaction to derive additional behavioral metrics, including the number of conversational turntaking and conversational dominance. At a sampling frequency of roughly 1 Hz, we will code who is talking at a given moment. We will count the number of turn-taking between two dyad partners and calculate conversational dominance as the ratio of the total talk duration of the less dominant partner and the total talk duration of the more dominant partner. Thus, a talk duration of one will indicate that both partners had an equal share of talking.

A Neuroscience Approach to Women Entrepreneurs’ Pitch Performance:. . .

4.6

221

Post-Experimental Assessments

Personality Traits Participants will complete the NEO-FFI-3 survey (McCrae & Costa Jr., 2007), the Adult Attachment Scale Survey (Collins & Read, 1990), and Wong and Law’s Emotional Intelligence Survey (Wong & Law, 2002) to capture personality traits. For the NEO-FFI-3 survey, we will calculate T-scores for all five subscales, including Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness. For the Adult Attachment Scale, we will calculate the Avoidant and Anxious subscales. For the Emotional Intelligence score, we will calculate the average of the four sub-scores: self-emotional appraisal, other’s emotional appraisal, use of emotion, and emotion regulation. We will utilize the personality trait scores as potential covariates in statistical analyses. Level of Professional Experience We will measure professional experience as the number of years of work experience reported by participants. Level of Self-Efficacy As a measure of self-efficacy, both entrepreneurs and investors will rate their degree of certainty in performing the role from “completely unsure” to “completely sure” (Chen et al., 1998).

5 Analytical Methods and Planned Research We are currently establishing data collection and present preliminary results at the conference. For data analysis, we will characterize the properties of inter-brain states by the occurrence rate of each inter-brain state following our previous work (Balters et al., 2022, 2023; Li et al., 2021). The occurrence rate of a state is defined as the percentage of total duration an inter-brain state occurs within the entire task period (Allen et al., 2014). We will utilize independent t-tests to assess whether there are differences in the occurrence rate between the two groups (same sex versus opposite sex dyads). We will calculate Pearson’s correlation between occurrence rate and pitch performance metrics. As an additional exploratory analysis, we will assess whether the dynamics of inter-brain states over time differ between the groups. We will also explore if we can identify other moderating factors that impact pitch performance in women entrepreneurs (e.g., personality traits, experience, selfefficacy). Building on the current study, we plan to extend the methods and measures presented in this paper to real-life pitch events. In this context, we aim to utilize fNIRS hyperscanning to develop interventions and training that can effectively improve the pitch performance of women entrepreneurs. We will also explore the opportunity of using real-time feedback to help women entrepreneurs modify IBS more successfully during the dyadic interaction. Lastly, we propose to extend two-person hyperscanning to hyperscanning of three persons simultaneously. Such a design will allow us to investigate the effects of inter-brain synchrony in a more

222

S. Balters et al.

realistic pitch setting that usually involves multiple actors. Related findings might allow us to derive and understand inter-brain patterns that are associated with the efficiency of the entrepreneur-investor matching process. We hope this work will provide novel and valuable insights into key factors of successful pitches and foster the success of women entrepreneurs. Funding This work is supported by a Hasso Plattner Design Thinking Research Program grant.

References Allen, E. A., Damaraju, E., Plis, S. M., Erhardt, E. B., Eichele, T., & Calhoun, V. D. (2014). Tracking whole-brain connectivity dynamics in the resting state. Cerebral Cortex, 24(3), 663–676. Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of Personality and Social Psychology, 63(4), 596. Babiloni, F., & Astolfi, L. (2014). Social neuroscience and hyperscanning techniques: Past, present and future. Neuroscience & Biobehavioral Reviews, 44, 76–93. Baker, J. M., Liu, N., Cui, X., Vrticka, P., Saggar, M., Hosseini, S. H., & Reiss, A. L. (2016). Sex differences in neural and behavioral signatures of cooperation revealed by fNIRS hyperscanning. Scientific Reports, 6(1), 1–11. Balachandra, L., Briggs, T., Eddleston, K., & Brush, C. (2019). Don’t pitch like a girl!: How gender stereotypes influence investor decisions. Entrepreneurship Theory and Practice, 43(1), 116–137. Balachandra, L., Fischer, K., & Brush, C. (2021). Do (women’s) words matter? The influence of gendered language in entrepreneurial pitching. Journal of Business Venturing Insights, 15, e00224. Balconi, M., Fronda, G., & Vanutelli, M. E. (2019). Donate or receive? Social hyperscanning application with fNIRS. Current Psychology, 38(4), 991–1002. Balters, S., Baker, J. M., Hawthorne, G., & Reiss, A. L. (2020). Capturing human interaction in the virtual age: A perspective on the future of fNIRS hyperscanning. Frontiers in Human Neuroscience, 14, 458. Balters, S., Weinstein, T. J., Hawthorne, G., & Reiss, A. L. (2022). Interpersonal trust activity to increase team creativity outcome: An fNIRS hyperscanning approach. Design Thinking Research. Springer. Balters, S., Miller, J. G., Li, R., Hawthorne, G., & Reiss, A. L. (2023). Virtual (zoom) interactions alter conversational behavior and interbrain coherence. Journal of Neuroscience, 43(14), 2568– 2578. Bembich, S., Saksida, A., Mastromarino, S., Travan, L., Di Risio, G., Cont, G., & Demarini, S. (2022). Empathy at birth: Mother’s cortex synchronizes with that of her newborn in pain. European Journal of Neuroscience. Boston Consulting Group. (2018). Why women-owned startups are a better bet. Retrieved from https://www.bcg.com/publications/2018/why-women-owned-startups-are-better-bet. Bruno, J. L., Baker, J. M., Gundran, A., Harbott, L. K., Stuart, Z., Piccirilli, A. M., Hosseini, S. H., Gerdes, J. C., & Reiss, A. L. (2018). Mind over motor mapping: Driver response to changing vehicle dynamics. Human Brain Mapping, 39(10), 3915–3927. Chen, C. C., Greene, P. G., & Crick, A. (1998). Does entrepreneurial self-efficacy distinguish entrepreneurs from managers? Journal of Business Venturing, 13(4), 295–316.

A Neuroscience Approach to Women Entrepreneurs’ Pitch Performance:. . .

223

Chen, X. P., Yao, X., & Kotha, S. (2009). Entrepreneur passion and preparedness in business plan presentations: A persuasion analysis of venture capitalists' funding decisions. Academy of Management Journal, 52(1), 199–214. Cheng, X., Li, X., & Hu, Y. (2015). Synchronous brain activity during cooperative exchange depends on gender of partner: A fNIRS-based hyperscanning study. Human Brain Mapping, 36(6), 2039–2048. Clingingsmith, D., Drover, W., & Shane, S. (2022). Examining the outcomes of entrepreneur pitch training: An exploratory field study. Small Business Economics, 1–28. Cohen, S., Kamarck, T., Mermelstein, R., et al. (1994). Perceived stress scale. Measuring Stress: A Guide for Health and Social Scientists, 235–283. Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58(4), 644. Cruncchbase. (2020). Global VC funding to female founders dropped dramatically this year. Retrieved from https://news.crunchbase.com/news/global-vc-funding-to-female-founders/ Cui, X., Bray, S., & Reiss, A. L. (2010). Functional near infrared spectroscopy (NIRS) signal improvement based on negative correlation between oxygenated and deoxygenated hemoglobin dynamics. NeuroImage, 49(4), 3039–3046. Cui, X., Bryant, D. M., & Reiss, A. L. (2012). NIRS-based hyperscanning reveals increased interpersonal coherence in superior frontal cortex during cooperation. NeuroImage, 59(3), 2430–2437. Drover, W., Busenitz, L., Matusik, S., Townsend, D., Anglin, A., & Dushnitsky, G. (2017). A review and road map of entrepreneurial equity financing research: Venture capital, corporate venture capital, angel investment, crowdfunding, and accelerators. Journal of Management, 43(6), 1820–1853. Duan, L., Dai, R., Xiao, X., Sun, P., Li, Z., & Zhu, C. (2015). Cluster imaging of multi-brain networks (CIMBN): A general framework for hyperscanning and modeling a group of interacting brains. Frontiers in Neuroscience, 9, 267. Feng, X., Sun, B., Chen, C., Li, W., Wang, Y., Zhang, W., et al. (2020). Self–other overlap and interpersonal neural synchronization serially mediate the effect of behavioral synchronization on prosociality. Social Cognitive and Affective Neuroscience, 15(2), 203–214. Ferrari, M., & Quaresima, V. (2012). A brief review on the history of human functional nearinfrared spectroscopy (fNIRS) development and fields of application. NeuroImage, 63(2), 921–935. Fishburn, F. A., Murty, V. P., Hlutkowsky, C. O., MacGillivray, C. E., Bemis, L. M., Murphy, M. E., Huppert, T. J., & Perlman, S. B. (2018). Putting our heads together: Interpersonal neural synchronization as a biological mechanism for shared intentionality. Social Cognitive and Affective Neuroscience, 13(8), 841–849. Foo, M. D., Wong, P. K., & Ong, A. (2005). Do others think you have a viable business idea? Team diversity and judges’ evaluation of ideas in a business plan competition. Journal of Business Venturing, 20, 385–402. Franke, N., Gruber, M., Harhoff, D., & Henkel, J. (2008). Venture capitalists’ evaluations of startup teams: Trade-offs, knock-out criteria, and the impact of VC experience. Entrepreneurship: Theory & Practice, 32, 459–483. Funane, T., Kiguchi, M., Atsumori, H., Sato, H., Kubota, K., & Koizumi, H. (2011). Synchronous activity of two people’s prefrontal cortices during a cooperative task measured by simultaneous near-infrared spectroscopy. Journal of Biomedical Optics, 16(7), 077011. Grinsted, A., Moore, J. C., & Jevrejeva, S. (2004). Application of the cross wavelet transform and wavelet coherence to geophysical time series. Nonlinear Processes in Geophysics, 11(5/6), 561–566. Hohl, L., Bican, P. M., Guderian, C. C., & Riar, F. J. (2021). Gender diversity effects in investment decisions. The Journal of Entrepreneurship, 30(1), 134–152. Holper, L., Scholkmann, F., & Wolf, M. (2012). Between-brain connectivity during imitation measured by fNIRS. NeuroImage, 63(1), 212–222.

224

S. Balters et al.

Hove, M. J., & Risen, J. L. (2009). It’s all in the timing: Interpersonal synchrony increases affiliation. Social Cognition, 27(6), 949–960. Hsu, D. K., Simmons, S. A., & Wieland, A. M. (2017). Designing entrepreneurship experiments: A review, typology, and research agenda. Organizational Research Methods, 20(3), 379–412. Hu, W.-L., Booth, J. W., & Reid, T. (2017). The relationship between design outcomes and mental states during ideation. Journal of Mechanical Design, 139(5), Article 5. Huppert, T., Barker, J., Schmidt, B., Walls, S., & Ghuman, A. (2017). Comparison of group-level, source localized activity for simultaneous functional near-infrared spectroscopy-magnetoencephalography and simultaneous fNIRS-fMRI during parametric median nerve stimulation. Neurophotonics, 4(1), 015001. Ikeda, S., Nozawa, T., Yokoyama, R., Miyazaki, A., Sasaki, Y., Sakaki, K., & Kawashima, R. (2017). Steady beat sound facilitates both coordinated group walking and inter-subject neural synchrony. Frontiers in Human Neuroscience, 11, 147. Kruse, L., Kochenderfer, M. J., Reiss, A. L., & Balters, S. (2021). Dyadic sex composition and task classification using fNIRS hyperscanning data. Proceedings of the IEEE International Conference on Machine Learning and Applications. December13–16, 2021, Pasadena, USA. Lancaster, J. L., Woldorff, M. G., Parsons, L. M., Liotti, M., Freitas, C. S., Rainey, L., Kochunov, P. V., Nickerson, D., Mikiten, S. A., & Fox, P. T. (2000). Automated Talairach atlas labels for functional brain mapping. Human Brain Mapping, 10(3), 120–131. Li, R., Mayseless, N., Balters, S., & Reiss, A. L. (2021). Dynamic inter-brain synchrony in real-life inter-personal cooperation: A functional near-infrared spectroscopy hyperscanning study. NeuroImage, 238, 118263. Liu, N., Cliffer, S., Pradhan, A. H., Lightbody, A., Hall, S. S., & Reiss, A. L. (2016a). Opticalimaging-based neurofeedback to enhance therapeutic intervention in adolescents with autism: Methodology and initial data. Neurophotonics, 4(1), 011003. Liu, N., Mok, C., Witt, E. E., Pradhan, A. H., Chen, J. E., & Reiss, A. L. (2016b). NIRS-based hyperscanning reveals inter-brain neural synchronization during cooperative Jenga game with face-to-face communication. Frontiers in Human Neuroscience, 10, 82. Liu, T., Saito, G., Lin, C., & Saito, H. (2017). Inter-brain network underlying turn-based cooperation and competition: A hyperscanning study using near-infrared spectroscopy. Scientific Reports, 7(1), 1–12. Lu, K., Teng, J., & Hao, N. (2020). Gender of partner affects the interaction pattern during group creative idea generation. Experimental Brain Research, 238(5), 1157–1168. Lu, K., Xue, H., Nozawa, T., & Hao, N. (2019). Cooperation makes a group be more creative. Cerebral Cortex, 29(8), 3457–3470. McCrae, R. R., & Costa, P. T., Jr. (2007). Brief versions of the NEO-PI-3. Journal of Individual Differences, 28(3), 116–128. McSweeney, J. J., McSweeney, K. T., Webb, J. W., & Devers, C. E. (2022). The right touch of pitch assertiveness: Examining entrepreneurs’ gender and project category fit in crowdfunding. Journal of Business Venturing, 37(4), 106223. Miao, H. (2022). Closing the gender gap for women-led businesses could boost global GDP by $2 trillion, Citi says. CNBC. Retrieved from https://www.cnbc.com/2022/03/07/closing-gendergap-could-boost-global-gdp-by-2-trillion-citi-says.html Molavi, B., & Dumont, G. A. (2012). Wavelet-based motion artifact removal for functional nearinfrared spectroscopy. Physiological Measurement, 33(2), 259. Murnieks, C. Y., Haynie, J. M., Wiltbank, R. E., & Harting, T. (2011). ‘I like how you think’: Similarity as an interaction bias in the investor–entrepreneur dyad. Journal of Management Studies, 48(7), 1533–1561. Niu, R., Yu, Y., Li, Y., & Liu, Y. (2019). Use of fNIRS to characterize the neural mechanism of inter-individual rhythmic movement coordination. Frontiers in Physiology, 10, 781. Nozawa, T., Sakaki, K., Ikeda, S., Jeong, H., Yamazaki, S., dos Santos Kawata, K. H., dos Santos Kawata, N. Y., Sasaki, Y., Kulason, K., Hirano, K., et al. (2019). Prior physical synchrony

A Neuroscience Approach to Women Entrepreneurs’ Pitch Performance:. . .

225

enhances rapport and inter-brain synchronization during subsequent educational communication. Scientific Reports, 9(1), 1–13. Osaka, N., Minamoto, T., Yaoi, K., Azuma, M., & Osaka, M. (2014). Neural synchronization during cooperated humming: A hyperscanning study using fNIRS. Procedia-Social and Behavioral Sciences, 126(1), 241–243. Osaka, N., Minamoto, T., Yaoi, K., Azuma, M., Shimada, Y. M., & Osaka, M. (2015). How two brains make one synchronized mind in the inferior frontal cortex: FNIRS-based hyperscanning during cooperative singing. Frontiers in Psychology, 6, 1811. Plichta, M. M., Herrmann, M. J., Baehne, C., Ehlis, A.-C., Richter, M., Pauli, P., & Fallgatter, A. J. (2006). Event-related functional near-infrared spectroscopy (fNIRS): Are the measurements reliable? NeuroImage, 31(1), 116–124. Pollonini, L., Bortfeld, H., & Oghalai, J. S. (2016). PHOEBE: A method for real time mapping of optodes-scalp coupling in functional near-infrared spectroscopy. Biomedical Optics Express, 7(12), 5104–5119. Russell, J. A. (1980). A Circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178. https://doi.org/10.1037/h0077714 Sanchez-Ruiz, P., Wood, M. S., & Long-Ruboyianes, A. (2021). Persuasive or polarizing? The influence of entrepreneurs’ use of ingratiation rhetoric on investor funding decisions. Journal of Business Venturing, 36(4), 106120. Santosa, H., Zhai, X., Fishburn, F., & Huppert, T. (2018). The NIRS brain AnalyzIR toolbox. Algorithms, 11(5), 73. Shane, S., Drover, W., Clingingsmith, D., & Cerf, M. (2020). Founder passion, neural engagement and informal investor interest in startup pitches: An fMRI study. Journal of Business Venturing, 35(4), 105949. Singh, A. K., Okamoto, M., Dan, H., Jurcak, V., & Dan, I. (2005). Spatial registration of multichannel multi-subject fNIRS data to MNI space without MRI. NeuroImage, 27(4), 842–851. Strangman, G., Culver, J. P., Thompson, J. H., & Boas, D. A. (2002). A quantitative comparison of simultaneous BOLD fMRI and NIRS recordings during functional brain activation. NeuroImage, 17(2), 719–731. Valdesolo, P., & DeSteno, D. (2011). Synchrony and the social tuning of compassion. Emotion, 11(2), 262. Walton, G. M., Brady, S. T., & Crum, A. J. (2020). The social-belonging intervention. In Handbook of wise interventions: How social psychology can help people change (pp. 36–62). Wiltermuth, S. S., & Heath, C. (2009). Synchrony and cooperation. Psychological Science, 20(1), 1–5. Wong, C.-S., & Law, K. S. (2002). The effects of leader and follower emotional intelligence on performance and attitude: An exploratory study. In The Leadership Quarterly (pp. 243–274). Routledge. Wyatt, J. S., Delpy, D. T., Cope, M., Wray, S., & Reynolds, E. (1986). Quantification of cerebral oxygenation and haemodynamics in sick newborn infants by near infrared spectrophotometry. The Lancet, 328(8515), 1063–1066. Yang, J., Zhang, H., Ni, J., De Dreu, C. K., & Ma, Y. (2020). Within-group synchronization in the prefrontal cortex associates with intergroup conflict. Nature Neuroscience, 1–7. Zhang, M., Liu, T., Pelowski, M., Jia, H., & Yu, D. (2017a). Social risky decision-making reveals gender differences in the TPJ: A hyperscanning study using functional near-infrared spectroscopy. Brain and Cognition, 119, 54–63. Zhang, M., Liu, T., Pelowski, M., & Yu, D. (2017b). Gender difference in spontaneous deception: A hyperscanning study using functional near-infrared spectroscopy. Scientific Reports, 7(1), 1–13. Zhou, C., Cheng, X., Liu, C., & Li, P. (2022). Interpersonal coordination enhances brain-to-brain synchronization and influences responsibility attribution and reward allocation in social cooperation. NeuroImage, 119028.

Priming Activity to Increase Interpersonal Closeness, Inter-Brain Coherence, and Team Creativity Outcome Stephanie Balters, Grace Hawthorne, and Allan L. Reiss

Abstract Organizational research demonstrates that team interpersonal closeness enhances team performance and creativity. Design thinking practitioners and educators have adopted the concepts of interpersonal closeness and developed priming activities to propel subsequent creative-innovation tasks. In recent years, it has become paramount that these activities are effective in in-person and virtual (Zoom®) interaction settings. In this chapter, we present a design thinking (DT) activity to increase interpersonal closeness in in-person and virtual teams. We derived the DT activity from a Nonviolent Communication exercise frequently used to increase interpersonal closeness between individuals. In an empirical study (N = 72 participants, N = 36 dyads), we assessed whether the DT activity increased interpersonal closeness compared to two control tasks (i.e., a problem-solving and a creative-innovation task). Dyad partners engaged in either an in-person or virtual interaction group throughout the experiment (between-subject design). We also captured inter-brain signatures between dyad partners with portable functional near-infrared spectroscopy neuroimaging during the entire study. Results show that the DT activity increased interpersonal closeness in the in-person and virtual groups compared to the control tasks. We identified a distinct inter-brain signature in the right frontocortical region linked to the DT activity. Notably, this inter-brain signature differed between in-person and virtual groups. This finding suggests that conducting the DT activity in person may be more conducive to this prosocial interbrain coherence pattern than the virtual interaction setting. Finally, preliminary results (N = 12 dyads) suggest that the DT activity increased performance in a subsequent creative-innovation task. Future research needs to confirm this hypothesis.

S. Balters (✉) · A. L. Reiss Center for Interdisciplinary Brain Sciences Research, Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, USA e-mail: [email protected]; [email protected] G. Hawthorne Hasso Plattner Institute of Design (d.school), Stanford, CA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_12

227

228

S. Balters et al.

1 Introduction Design thinking practice has adopted the concept of interpersonal closeness to improve team collaboration and creative-innovation (Druskat & Wolff, 2001; Uebernickel & Thong, 2021). Research shows that design thinking training can increase team collaboration and creative confidence (Jobst et al., 2012; Kelley & Kelley, 2013; Rauth et al., 2010; Royalty et al., 2012, 2014) through practicing empathy and social interaction skills in a design thinking team setting (Noweski et al., 2012; Plank et al., 2021; Traifeh et al., 2020; von Thienen et al., 2017). Practitioners are using many design thinking activities—explicitly or implicitly—to help increase interpersonal closeness and thereby enhance team collaboration and creative-innovation (Kerguenne, 2021; Koch, 2021; Ney & Meinel, 2019). For example, team “check-ins” and “checkouts” before and after each workday to share momentary personal sensitivities, opinions, and feelings with the team can improve team collaboration (Ney & Meinel, 2019). If a team or interpersonal conflict needs to be resolved, design thinking encourages the use of structured formats for communicating and receiving feedback (Ney & Meinel, 2019). Other activities serve as primers for specific collaborative work modes (e.g., creative-innovation sessions) and are typically applied when initiating a new design thinking work phase. Such design thinking (DT) activities usually involve stepping out of one’s comfort zone and often result in some form of playful body movement or verbal interaction (Rothouse, 2020; West et al., 2017). As a result, these DT activities contribute to establishing a work culture of psychological trust (Auernhammer & Roth, 2021; Edmondson, 1999; Liedtka, 2017), in which one is permitted to make mistakes and show vulnerabilities in front of colleagues without fearing rejection (West et al., 2017). In the last year’s edition of Design Thinking Research, we introduced a DT activity designed to enhance interpersonal closeness and team creative-innovation in in-person and virtual (Zoom®) team interactions (Balters et al., 2022). We derived the DT activity from Nonviolent Communication practices, a communication method to increase interpersonal closeness and trust between individuals (Rosenberg & Chopra, 2015). Here, we provide empirical validation of the effectiveness of this DT activity in both in-person and virtual (Zoom®) interaction settings. We invited a total of N = 72 participants who interacted with their dyad partner either in person or virtually during the duration of the study. Dyads collaborated during the DT activity and two control tasks (i.e., a problem-solving and a creative-innovation task). After each of the three tasks, participants rated measures of interpersonal closeness (i.e., connectedness, trust, likability, other-in-self, and similarity). We also captured interbrain signatures between dyad partners using functional near-infrared spectroscopy neuroimaging during the entire study. Our analyses focused on (1) elucidating the behavioral and inter-brain correlates of the DT activity and (2) assessing whether the effectiveness of the interaction differed between in-person and virtual interactions.

Priming Activity to Increase Interpersonal Closeness, Inter-Brain. . .

229

2 DT Activity to Increase Interpersonal Closeness between Dyad Partners During the DT activity, participants engage in a modified version of a Nonviolent Communication exercise used to increase interpersonal closeness and trust between individuals (Rosenberg & Chopra, 2015). Participants are provided with a list of “Needs We All Have” (Fig. 1). Participants are asked to collaborate and identify four needs from the list that are most meaningful to them. To emphasize the importance of each need, they are asked to describe a situation from their life when this need was not met and how it made them feel. The partner is instructed to actively listen, acknowledge the feelings of the one who shared, and describe why the need is also meaningful to them. The design thinking facilitator keeps a time of 8 min.

3 Study Methodology to Assess the Efficacy of Need Sharing Activity The study methodology was approved by the Stanford University Institutional Review Board (IRB #18160) and followed COVID-19 regulations for human experimentation as defined by the Stanford University School of Medicine. Written

Needs we all have acceptance affection appreciation belonging cooperation communication closeness community companionship compassion consideration consistency empathy inclusion

intimacy love mutuality nurturing respect/self-respect safety security stability support to know and to be known to see and be seen to understand and be understood trust warmth

Introduction: For the next eight minutes collaborate with your partner. Together, identify four needs from the list above that are most meaningful to both of you. To emphasize the importance of each need, describe a situation from your own life when that need was not met and how it made you feel. As a partner, listen actively, acknowledge the feelings of the one who shared, and describe why the need is also meaningful to you.

Fig. 1 Instructions for the DT activity. The figure is derived from (Balters et al., 2022)

230

S. Balters et al.

consent was obtained from all participants. The study methodology has been previously presented in (Balters et al., 2023a). Below, we summarize the study methodology and refer the reader to the original paper for detailed information.

3.1

Participants

A total of 72 adults participated in the study (36 females, 36 males, mean age: 27.11 years, SD = 7.57 years). The racial and ethnic composition of the sample was 3% African American/Black, 23% Asian/Pacific Islander, 6% biracial/multiracial, 15% Hispanic/Latinx, 6% Middle Eastern, 22% South Asian, and 25% Caucasian/ White. All participants were right-handed and healthy, with normal or corrected to normal hearing and vision. The study followed a between-subject design, and participants interacted with their dyad partner either in person or virtually throughout the experiment. The previously unacquainted dyad partners were randomly assigned to either interaction condition. Groups were matched based on age, sex, and race/ ethnicity. Both interaction conditions contained six female-female, six female-male, and six male-male dyads. The experimental procedure lasted for approximately 3 hours, and participants were compensated with an Amazon gift card ($25 USD per hour).

3.2

Experimental Procedure

In the in-person group, dyads sat face-to-face at a square table 9 feet away, following COVID-19 guidelines (Fig. 2). To decrease obstruction of faces, participants wore transparent, antifog facemasks (ClearMask™). Dyads of the virtual group sat at desks in two separate rooms and interacted over Zoom® video conferencing. The Zoom® window was maximized, and no self-view window was displayed. We used two identical laptops for video conferencing (Lenovo Yoga 730-15IKB, 15.6″) and

Fig. 2 Between-subject study setup. Thirty-six dyads interacted either in person (a) or virtually (b) during the study

Priming Activity to Increase Interpersonal Closeness, Inter-Brain. . .

231

placed the laptops to approximate the facial proportions of the in-person group. Participants of the virtual group also wore a facemask to prevent bias between the conditions. Before the experiment, participants had 3 min to introduce themselves to one another. During the experiment, participants were alone in the room(s), and we provided instructions through audio prompts. After the experiment, participants filled out post-experimental questionnaires in two separate rooms.

3.3

Experimental Tasks

In addition to the 8-min-long DT activity described in Sect. 2, participants engaged in two other collaborative tasks (i.e., a problem-solving and a creative-innovation task). Both tasks serve as control tasks in data analyses that test the effectiveness of the DT activity. In the problem-solving task, dyads were instructed to collaborate and identify four traffic rules that significantly impact safety on US highways. To emphasize the importance of each rule, participants had to describe how a chosen rule enhances safety on US highways and why the rule was more important than other traffic rules. In the creative-innovation task, dyads were instructed to collaborate and to design a solution to increase water conservation in California households. The solution could take any form (i.e., product, process, campaign, etc.). Dyads were instructed to write down their solution after completion of the task. Dyads collaborated on each task (i.e., DT activity, problem-solving task, and creative-innovation task) for 8 min without interruptions. The task order was randomized across dyads, and we separated tasks by a 2-min calming video of a beach to minimize carryover effects across tasks.

3.4

Neuroimaging with Functional Near-Infrared Spectroscopy (fNIRS)

Functional near-infrared spectroscopy (fNIRS) is a portable neuroimaging technology that has become popular in the field of design (thinking) research (Balters et al., 2022, 2023b). Compared to portable electroencephalography (EEG) neuroimaging, fNIRS has a higher spatial resolution (~1 cm) and higher robustness to motion artifacts (Li et al., 2017). These advantages make fNIRS an ideal tool for assessing cortical brain function in applied design (thinking) contexts (Balters et al., 2022, 2023b). For example, design (thinking) researchers have utilized fNIRS to investigate the neural signatures associated with unstructured idea generation (i.e., brainstorming, Hu et al., 2021) and structured idea generation via design science tools such as Theory of Inventive Problem-solving “TRIZ” (Hu et al., 2018; Shealy et al., 2018) or sketching (Kato et al., 2017, 2018). About a decade ago, researchers extended single-brain assessments to hyperscanning modes in which two or more brains are scanned simultaneously. Researchers have focused on assessing when and

232

S. Balters et al.

Fig. 3 Functional NIRS regions of interest. We measured concentration changes of oxygenated hemoglobin (HbO) in 32 regions of interest across the cortex

how neural processes become synchronized and how this inter-brain coherence (IBC; correlation of cortical activity between brains) relates to behavioral measures (Babiloni & Astolfi, 2014; Balters et al., 2020; Czeszumski et al., 2020). Results from fNIRS hyperscanning studies have shown increased IBC during enhanced levels of dyadic interaction, such as during cooperative games (Baker et al., 2016; Cui et al., 2012; Kruse et al., 2021), in therapy sessions (Zhang et al., 2018), in mother-child bonding (Bembich et al., 2022), and after gift exchanges (Balconi et al., 2019; Balconi & Fronda, 2020). For a more comprehensive introduction to fNIRS hyperscanning, we refer the reader to Balters et al. (2020). We utilized fNIRS hyperscanning in this study to elucidate the inter-brain correlates of the DT activity and to assess whether the effectiveness of the interaction differed between in-person and virtual interactions. Specifically, we recorded the cortical hemodynamic activity of each participant using a continuous wave fNIRS system (NIRSport2 System, NIRX, Germany). We utilized two wavelengths (760 and 850 mm) and a sampling frequency of 10.2 Hz. The system high density contains 64 sources and 64 detectors, which we divided between the 2 participants to generate 100 measurement channels per person. According to the international 10–20 EEG placement system, we placed the channels over the entire cortex. For data processing, we followed rigorous scientific procedures (see Balters et al., 2023a for detailed processing steps) to derive concentration changes of oxygenated hemoglobin (HbO) for a total of 32 regions of interest (ROIs, Fig. 3). We then applied Wavelet Transform Coherence analysis (Cui et al., 2012) to assess averaged IBC values across each 8-minute task. (Note: Inter-brain coherence is a measure of similarity between NIRS signals of dyad partners across a specific time duration.) For each task (i.e., DT activity, problem-solving task, and creative-innovation task), we calculated IBC values for each possible ROI pair between participants. We utilized these IBC values in statistical analyses.

3.5

Interpersonal Closeness Measures

Participants rated the subjective sense of closeness toward their dyad partner (i.e., “interpersonal closeness” [Tarr et al., 2015]) on five seven-point Likert subscales. This included questions about connectedness and trust (Wiltermuth & Heath, 2009), likeability (Hove & Risen, 2009), similarity in personality (Valdesolo & DeSteno,

Priming Activity to Increase Interpersonal Closeness, Inter-Brain. . .

233

2011), and an adapted version of the inclusion of other-in-self scale (i.e., “people’s sense of being interconnected with another”; Aron et al., 1992). We used the five interpersonal closeness measures to test if the DT activity, in contrast to the two control tasks, was effective in increasing interpersonal closeness in the in-person and virtual dyads.

3.6

Post-Experimental Assessments

After the experiment, participants filled out personality trait surveys (i.e., NEO-FFI3 survey [McCrae & Costa Jr., 2007], Adult Attachment Scale Survey [Collins & Read, 1990], and Wong and Law’s Emotional Intelligence Survey [Wong & Law, 2002]) to capture personality traits. They also executed the Alternate Uses Task (AUT) to assess individual levels of divergent thinking and creativity (Guilford, 1967). Lastly, participants rated their prior experience and proficiency with Zoom® video conferencing. We used these measures to assess whether the two groups (in-person, virtual) matched on various individual difference variables or whether certain covariates would need to be considered in statistical analyses.

4 Study Results and Discussion 4.1

Conditions Were Matched on Individual Difference Variables

In the original paper (Balters et al., 2023a), we demonstrated that the two groups (in-person, virtual) matched on various individual difference variables. Specifically, subjects participating in the two groups were matched on age, inter-dyad age differences, personality traits (i.e., NEO-FFI-3 T scores, adult attachment style, emotional intelligence), creative ability (i.e., Alternate Uses Task (AUT) fluency, AUT originality), and familiarity with Zoom® video conferencing (i.e., experience and proficiency). These findings allowed us to execute the primary research analyses without controlling specific covariates.

4.2

Activity Increases Interpersonal Closeness

While we utilized the average score of all five interpersonal closeness measures in the original paper (Balters et al., 2023a), we present a novel measure-specific in this chapter. For each of the five interpersonal closeness measures (i.e., connectedness, trust, likability, other-in-self, and similarity), we ran a two-way analysis of variance

234

S. Balters et al.

Fig. 4 Results of the interpersonal closeness measures. Four of the five interpersonal closeness measures (i.e., connectedness, trust, likability, and other-in-self) showed higher values in the DT activity as compared to the two control tasks (i.e., problem-solving and creative-innovation tasks). We did not find main or interaction effects of group on the measures. These findings suggest that the DT activity was effective in increasing measures of interpersonal closeness in both in-person and virtual teams. Statistically significant differences at FDR-corrected p < 0.05 are indicated by *. Variance is illustrated as the standard error of the means

(ANOVA) with task (problem-solving, creative-innovation, DT activity) as withinsubject factor and group (in-person, virtual) as between-subject factor. These analyses tested for potential main effects of task, main effects of group, and interaction effects of task and group on the five measures. We applied false discovery rate (FDR) correction for multiple testing on the resulting p values (i.e., five testings). The results showed significant main effects of task on connectedness (F [2140] = 9.458, p < 0.001, partial η2 = 0.119), trust (F[2140] = 8.556, p < 0.001, partial η2 = 0.109), likability (F[2140] = 10.590, p < 0.001, partial η2 = 0.131), and other-in-self (F[2140] = 9.458, p = 0.013, partial η2 = 0.064), but not on similarity ( p = 0.068). We conducted subsequent post hoc pairwise comparison analyses with FDR correction for multiple comparisons (i.e., three comparisons per measure). The results for all four measures showed increased values for the DT activity compared to the problem-solving and creative-innovation task (Fig. 4). We summarize the statistics in Table 1. There were no significant differences between the problem-solving and the creative-innovation tasks. The results indicate that the DT activity effectively increased connectedness, trust, likability, and other-in-self measures in contrast to both control tasks. Notably, we did not find the main effects

Priming Activity to Increase Interpersonal Closeness, Inter-Brain. . .

235

Table 1 Overview of the pairwise comparison statistics Interpersonal closeness measures Connectedness DT activity > problem-solving task DT activity > creative-innovation task Trust DT activity > problem-solving task DT activity > creative-innovation task Likability DT activity > problem-solving task DT activity > creative-innovation task Other-in-self DT activity > problem-solving task DT activity > creative-innovation task

Mean difference between task with adjusted p values 0.389 [95% CI, 0.181 to 0.596, p < 0.001] 0.458 [95% CI, 0.214 to 0.702, p < 0.001] 0.208 [95% CI, 0.040 to 0.377, p = 0.024] 0.347 [95% CI, 0.170 to 0.525, p < 0.001] 0.222 [95% CI, 0.009 to 0.355, p = 0.002] 0.403 [95% CI, 0.222 to 0.583, p < 0.001] 0.208 [95% CI, 0.022 to 0.394, p = 0.044] 0.319 [95% CI, 0.072 to 0.567, p = 0.036]

CI confidence interval

of group or interaction effects of task and group on any of the five measures ( p > 0.155). These results suggest that the DT activity effectively increased interpersonal closeness measures independent of the interaction settings. For the four interpersonal closeness measures that showed significant univariate effects, we conducted subsequent correlation analyses with each of the other three measures. These analyses focused on data from the DT activity only. We used FDR correction to adjust for multiple testing (i.e., six correlation analyses). Results showed statistically significant positive correlations for all six analyses (0.40 < r < 0.84, p < 0.001; Fig. 5). These findings validate that connectedness, trust, likability, and other-in-self are all measures of the same interpersonal closeness construct.

4.3

Inter-Brain Coherence

In the original paper (Balters et al., 2023a), we executed statistical analyses to assess whether there were main effects of task (problem-solving task, creative-innovation task, DT activity), main effects of group (in-person, virtual), or interaction effects of task and group on inter-brain coherence values. The results showed a statistically significant interaction effect of task and group for an ROI pair spanning the right dorsal frontopolar area (dFPA) across dyad partners (Fig. 6a). In other words, specifically for the DT activity (and not the other two tasks), IBC significantly differed between in-person and virtual dyads. As demonstrated in Fig. 6b, IBC was higher in the in-person group than in the virtual group (mean difference = 0.084 [95% CI, 0.045 to 0.123, p < 0.001]). Prior research has consistently found increased IBC, particularly in right prefrontal ROI pairs, to be positively associated with prosocial behavior (Cui et al., 2012; Dai et al., 2018; Liu et al., 2016; Lu et al.,

236

S. Balters et al.

Fig. 5 Results of the correlation analyses across interpersonal closeness measures. Results showed statistically significant correlations between all four interpersonal closeness measures that demonstrated significant differences in the primary analyses. The moderate to strong associations are positive. These findings validate that connectedness, trust, likability, and other-in-self are all measures of the same interpersonal closeness construct

Fig. 6 Results of the inter-brain coherence analyses. Results showed a statistically significant interaction effect between task and group for the ROI pair spanning the right dorsal frontopolar area (dFPA) across dyad partners (a). In other words, specifically for the DT activity (and not the other two tasks), IBC significantly differed between in-person and virtual dyads (b). Statistically significant differences at FDR-corrected p < 0.05 are indicated by *. Variance is illustrated as the standard error of the means

2019; Miller et al., 2019; Pan et al., 2017; Xue et al., 2018). Therefore, our findings suggest that conducting the DT activity in person elicits IBC, which may be more conducive to prosocial interaction (Balters et al., 2023a).

Priming Activity to Increase Interpersonal Closeness, Inter-Brain. . .

4.4

237

Preliminary Performance Analysis in Subsequent Creative-Innovation Task

Lastly, we were interested in getting a first glimpse of whether the DT activity effectively increased performance in a subsequent creative-innovation session. We conducted a preliminary analysis of parts of the existing data. Specifically, we utilized data from task order 1 and task order 6 that resulted from this study’s randomized task order methodology (with N = 6 dyads per order across both groups; Fig. 7a). Dyads with task order 1 experienced the DT activity before the creativeinnovation task. In contrast, dyads with task order 6 experienced the problemsolving task before the creative-innovation task. This preliminary analysis used the latter task order as a control condition. We define task order 1 as the “priming condition” and task order 6 as the “control condition.” Two researchers rated the level of creative-innovation for all dyads to generate performance measures of the creative-innovation task. Both raters had expertise in assessing creative-innovation according to the Stanford Design School principles (authors S.B. and G.H.). Specifically, the raters assessed the creative-innovation rating based on four subscales, including fluency (i.e., the total number of elements in a design solution), originality (i.e., the statistical rarity of the response across answers in this study), elaboration (i.e., the level of imagination and exposition of detail), and accountability (i.e., the effectiveness of water conservation) as derived from (Torrance, 1974). We obtained the ratings on a five-point scale ranging from 1 (“very low”) to 5 (“very high”) and averaged the scores from these four subscales to get the final creative-innovation performance score. Inter-rater reliability index was good (ICC = 0.855). We extracted the scores from dyads belonging to the priming condition and the control condition. We removed outliers with more than two standard deviations from the mean, which resulted in data for N = 5 dyads in the priming condition and N = 6 in the control condition. Due to the low statistical power, we refrained from executing formal statistical analyses and inspected descriptive statistics, including effect sizes. As depicted in Fig. 7b, the average

Fig. 7 Descriptive results of the preliminary analysis of performance in subsequent creativeinnovation task. (a) Our study methodology comprised six different task orders across dyads due to the randomization of the three experimental tasks. (b) Descriptive findings indicate that the DT activity (here labeled priming condition) could effectively increase performance in a subsequent creative-innovation task compared to the control condition. Horizontal lines represent the median values, and “x” represents the mean values

238

S. Balters et al.

task performance in the creative-innovation task was higher for the priming condition than the control condition, with a moderate effect size of Cohen’s d = 0.500. These descriptive results could suggest that the DT activity was effective in increasing performance in the subsequent creative-innovation task, compared to a preceding problem-solving task. Subsequent correlation analyses between the performance score and each creative ability score (i.e., AUT fluency and AUT originality from Sect. 4.1) across all N = 11 dyads were not significant ( p > 0.223). This additional preliminary finding suggests that the observed descriptive difference in performance score could be attributed to the DT activity rather than differences in creative ability traits between dyads in the two conditions. Future research needs to formally test the hypothesis that the DT activity, compared to a control task (e.g., problem-solving task), increases performance in a subsequent creative-innovation task.

5 Conclusion This chapter presented a DT activity to increase interpersonal closeness and creativeinnovation in in-person and virtual design thinking teams. We derived the DT activity from a Nonviolent Communication exercise frequently used to increase interpersonal closeness between individuals. We tested the effectiveness of the DT activity through an empirical study that considered both in-person (N = 18 dyads) and virtual interaction settings (N = 18 dyads). The study results provide evidence that the DT activity, in contrast to two other collaborative control tasks (i.e., problem-solving and creative-innovation tasks), effectively increased measures of interpersonal closeness (i.e., connectedness, trust, likability, and other-in-self) in both in-person and virtual interaction settings. Functional NIRS neuroimaging results demonstrated that the DT activity elicited inter-brain coherence in regions associated with prosocial behavior. However, our findings also indicate that conducting the DT activity in person could be more conducive to this prosocial inter-brain coherence pattern than the virtual interaction setting. Finally, preliminary results indicated that the DT activity could effectively increase team performance in a subsequent creative-innovation task. Future research needs to validate these preliminary findings. We hope this chapter provides practitioners and educators alike with a novel DT activity to increase interpersonal closeness in in-person and virtual design thinking teams.

References Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of Personality and Social Psychology, 63(4), 596.

Priming Activity to Increase Interpersonal Closeness, Inter-Brain. . .

239

Auernhammer, J., & Roth, B. (2021). The origin and evolution of Stanford University’s design thinking: From product design to design thinking in innovation management. Journal of Product Innovation Management, 38(6), 623–644. Babiloni, F., & Astolfi, L. (2014). Social neuroscience and hyperscanning techniques: Past, present and future. Neuroscience & Biobehavioral Reviews, 44, 76–93. Baker, J. M., Liu, N., Cui, X., Vrticka, P., Saggar, M., Hosseini, S. H., & Reiss, A. L. (2016). Sex differences in neural and behavioral signatures of cooperation revealed by fNIRS hyperscanning. Scientific Reports, 6(1), Article 1. Balconi, M., & Fronda, G. (2020). The “gift effect” on functional brain connectivity. Inter-brain synchronization when prosocial behavior is in action. Scientific Reports, 10(1), 1–10. Balconi, M., Fronda, G., & Vanutelli, M. E. (2019). Donate or receive? Social hyperscanning application with fNIRS. Current Psychology, 38(4), 991–1002. Balters, S., Baker, J. M., Hawthorne, G., & Reiss, A. L. (2020). Capturing human interaction in the virtual age: A perspective on the future of fNIRS hyperscanning. Frontiers in Human Neuroscience, 14, 458. Balters, S., Miller, J. G., Li, R., Hawthorne, G., & Reiss, A. L. (2023a). Virtual (zoom) interactions alter conversational behavior and interbrain coherence. Journal of Neuroscience, 43(14), 2568– 2578. Balters, S., Weinstein, T., Mayseless, N., Auernhammer, J., Hawthorne, G., Steinert, M., Meinel, C., Leifer, L. J., & Reiss, A. L. (2023b). Design science and neuroscience: A systematic review of the emergent field of design neurocognition. Design Studies, 84, 101148. Balters, S., Weinstein, T. J., Hawthorne, G., & Reiss, A. L. (2022). Interpersonal trust activity to increase team creativity outcome: An fNIRS hyperscanning approach. Design Thinking Research. Springer. Bembich, S., Saksida, A., Mastromarino, S., Travan, L., Di Risio, G., Cont, G., & Demarini, S. (2022). Empathy at birth: Mother’s cortex synchronizes with that of her newborn in pain. European Journal of Neuroscience, 55(6), 1519–1531. Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58(4), 644. Cui, X., Bryant, D. M., & Reiss, A. L. (2012). NIRS-based hyperscanning reveals increased interpersonal coherence in superior frontal cortex during cooperation. NeuroImage, 59(3), Article 3. Czeszumski, A., Eustergerling, S., Lang, A., Menrath, D., Gerstenberger, M., Schuberth, S., Schreiber, F., Rendon, Z. Z., & König, P. (2020). Hyperscanning: A valid method to study neural inter-brain underpinnings of social interaction. Frontiers in Human Neuroscience, 14, 39. Dai, B., Chen, C., Long, Y., Zheng, L., Zhao, H., Bai, X., et al. (2018). Neural mechanisms for selectively tuning in to the target speaker in a naturalistic noisy situation. Nature Communications, 9(1), 2405. Druskat, V. U., & Wolff, S. B. (2001). Building the emotional intelligence of groups. Harvard Business Review, 79(3), 80–91. Edmondson, A. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350–383. Guilford, J. P. (1967). The nature of human intelligence. McGraw-Hill. Hove, M. J., & Risen, J. L. (2009). It’s all in the timing: Interpersonal synchrony increases affiliation. Social Cognition, 27(6), 949–960. Hu, M., Shealy, T., & Gero, J. S. (2018). Neuro-cognitive differences among engineering students when using un-structured, partially structured, and structured design concept generation techniques. 125th ASEE Annual Conference and Exposition. Salt Lake City, UT: American Society for Engineering Education. Hu, M., Shealy, T., & Milovanovic, J. (2021). Cognitive differences among first-year and senior engineering students when generating design solutions with and without additional dimensions of sustainability. Design Science, 7.

240

S. Balters et al.

Jobst, B., Köppen, E., Lindberg, T., Moritz, J., Rhinow, H., & Meinel, C. (2012). The faith-factor in design thinking: Creative confidence through education at the design thinking schools Potsdam and Stanford? In Design thinking research (pp. 35–46). Springer. Kato, T., Okada, H., & Izu, Y. (2017). Brain activities of idea generation using sketches. In The 4th International Conference on Design Engineering and Science, Aachen, Germany. Kato, T., Okada, H., Izu, Y., et al. (2018). Measurement of brain activities of idEA generation (Sketch). DS 92: Proceedings of the DESIGN 2018 15th International Design Conference, 2027–2034. Kelley, T., & Kelley, D. (2013). Creative confidence: Unleashing the creative potential within us all. Currency. Kerguenne, A. (2021). Strategisches Design Thinking als neuer Weg in der Führung von digitalen Transformations-Prozessen. In C. Meinel & T. Krohn (Eds.), Design Thinking in der Bildung— Innovation kann man lernen (pp. 285–302). Wiley-VCH. Koch, J. (2021). Human Centeredness in Professional Education—Über den Nutzen und die Anwendung eines menschenzentrierten Ansatzes im Bereich Professional Education. In C. Meinel & T. Krohn (Eds.), Design Thinking in der Bildung—Innovation kann man lernen (pp. 303–314). Wiley-VCH. Kruse, L., Kochenderfer, M. J., Reiss, A. L., & Balters, S. (2021). Dyadic sex composition and task classification using fNIRS hyperscanning data. Proceedings of the IEEE International Conference on Machine Learning and Applications. December13–16, 2021, Pasadena, USA. Li, R., Potter, T., Huang, W., & Zhang, Y. (2017). Enhancing performance of a hybrid EEG-fNIRS system using channel selection and early temporal features. Frontiers in Human Neuroscience, 11, 462. Liedtka, J. (2017). Evaluating the impact of design thinking in action. Academy of Management Proceedings, 2017(1), 10264. Liu, N., Mok, C., Witt, E. E., Pradhan, A. H., Chen, J. E., & Reiss, A. L. (2016). NIRS-based hyperscanning reveals inter-brain neural synchronization during cooperative Jenga game with face-to-face communication. Frontiers in Human Neuroscience, 10, 82. Lu, K., Xue, H., Nozawa, T., & Hao, N. (2019). Cooperation makes a group be more creative. Cerebral Cortex, 29(8), 3457–3470. McCrae, R. R., & Costa, P. T., Jr. (2007). Brief versions of the NEO-PI-3. Journal of Individual Differences, 28(3), 116–128. Miller, J. G., Vrtička, P., Cui, X., Shrestha, S., Hosseini, S. H., Baker, J. M., & Reiss, A. L. (2019). Inter-brain synchrony in mother-child dyads during cooperation: An fNIRS hyperscanning study. Neuropsychologia, 124, 117–124. Ney, S., & Meinel, C. (2019). Putting design thinking to work: How large organizations can embrace messy institutions to tackle wicked problems. Springer International Publishing. Noweski, C., Scheer, A., Büttner, N., von Thienen, J., Erdmann, J., & Meinel, C. (2012). Towards a paradigm shift in education practice: Developing twenty-first century skills with design thinking. In Design thinking research (pp. 71–94). Springer. Pan, Y., Cheng, X., Zhang, Z., Li, X., & Hu, Y. (2017). Cooperation in lovers: An fNIRS-based hyperscanning study. Human Brain Mapping, 38(2), 831–841. Plank, I. S., von Thienen, J. P. A., & Meinel, C. (2021). The neuroscience of empathy: Researchoverview and implications for human-centred design. Design Thinking Research, 89–124. Rauth, I., Köppen, E., Jobst, B., Meinel, C., & et al. (2010). Design thinking: An educational model towards creative confidence. DS 66–2: Proceedings of the 1st International Conference on Design Creativity (ICDC 2010). Rosenberg, M. B., & Chopra, D. (2015). Nonviolent communication: A language of life: Lifechanging tools for healthy relationships. PuddleDancer Press. Rothouse, M. J. (2020). A mindful approach to team creativity and collaboration in organizations: Creating a culture of innovation. Springer. Royalty, A., Oishi, L., & Roth, B. (2012). “I use it every day”: Pathways to adaptive innovation after graduate study in design thinking. In Design thinking research (pp. 95–105). Springer.

Priming Activity to Increase Interpersonal Closeness, Inter-Brain. . .

241

Royalty, A., Oishi, L. N., & Roth, B. (2014). Acting with creative confidence: Developing a creative agency assessment tool. In Design thinking research (pp. 79–96). Springer. Shealy, T., Hu, M., & Gero, J. (2018). Patterns of cortical activation when using concept generation techniques of brainstorming, morphological analysis, and TRIZ. ASME 2018 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. Tarr, B., Launay, J., Cohen, E., & Dunbar, R. (2015). Synchrony and exertion during dance independently raise pain threshold and encourage social bonding. Biology Letters, 11(10), Article 10. Torrance, E. P. (1974). The Torrance tests of creative thinking-TTCT manual and scoring guide: Verbal test a, figural test. Ginn. Traifeh, H., Nicolai, C., Refaie, R., Meinel, C., & et al. (2020). Engaging digital engineering students in design thinking. DS 101: Proceedings of NordDesign 2020, Lyngby, Denmark, 12th–14th August 2020, 1–14. Uebernickel, F., & Thong, C. (2021). Die Kontextualisierung des Design Thinking mit multiplen Intelligenzen. In C. Meinel & T. Krohn (Eds.), Design Thinking in der Bildung—Innovation kann man lernen (pp. 387–428). Wiley-VCH. Valdesolo, P., & DeSteno, D. (2011). Synchrony and the social tuning of compassion. Emotion, 11(2), 262. von Thienen, J., Royalty, A., & Meinel, C. (2017). Design thinking in higher education: How students become dedicated creative problem solvers. In Handbook of research on creative problem-solving skill development in higher education (pp. 306–328). IGI Global. West, S., Hoff, E., & Carlsson, I. (2017). Enhancing team creativity with playful improvisation theater: A controlled intervention field study. International Journal of Play, 6(3), 283–293. Wiltermuth, S. S., & Heath, C. (2009). Synchrony and cooperation. Psychological Science, 20(1), 1–5. Wong, C.-S., & Law, K. S. (2002). The effects of leader and follower emotional intelligence on performance and attitude: An exploratory study. In The Leadership Quarterly (pp. 243–274). Routledge. Xue, H., Lu, K., & Hao, N. (2018). Cooperation makes two less-creative individuals turn into a highly-creative pair. Neuroimage, 172, 527–537. Zhang, Y., Meng, T., Hou, Y., Pan, Y., & Hu, Y. (2018). Interpersonal brain synchronization associated with working alliance during psychological counseling. Psychiatry Research: Neuroimaging, 282, 103–109.

Design the Future with Emotion: Crucial Cultural Perspectives Chunchen Xu, Xiao Ge, Nanami Furue, Daigo Misaki, Hazel Markus, and Jeanne Tsai

Abstract No one universal affective route leads to creative ideas. Rather, the designers’ affective experience is influenced by the cultural contexts they are in. However, scant research has examined how culture shapes designers’ emotion in creative problem-solving activities. We present two survey studies that explore the interplay between affect, culture, and idea generation. The findings suggest that people tend to associate low-arousal, positive emotion with idea generation in Japanese contexts, compared with high-arousal, positive emotion in American contexts. We also found that Japanese participants expressed more socially engaging emotions, had higher levels of emotional fluctuation, and reported lower levels of emotional expressiveness than their American counterparts. This research contributes to the emerging field of emotion research in design by examining the cultural shaping of affect in idea generation. We call for more cultural research to enable designers to provide insights into the profound roles of affective experience and expression in creative processes and how it may vary across cultures. In doing so, we hope to offer new vistas for enhancing creative performance and enabling crosscultural collaboration in creative work.

C. Xu (✉) Stanford Psychology Department, Stanford University, Stanford, CA, USA e-mail: [email protected] X. Ge Center for Design Research, Stanford University, Stanford, CA, USA N. Furue School of Management, Tokyo University of Science, Chiyoda-ku, Tokyo, Japan D. Misaki School of Engineering, Department of Mechanical Systems Engineering, Kogakuin University, Shinjukuku, Tokyo, Japan H. Markus · J. Tsai Stanford Psychology Department, Stanford, CA, USA © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_13

243

244

C. Xu et al.

1 Introduction The way that designers think, feel, and act is constantly influenced by their cultural contexts. For example, while the Chinese saying “急中生智” depicts a calm thinker who comes up with ingenious solutions amid crisis, such as in the famous Chinese folktale about Sima Guang and the Water Tank, the American mantra “unleash your creative potential” advocates excited self-expression as a way to elicit a generative state of mind. Nonetheless, most current theories and practices aimed at enhancing creativity in the USA have been chiefly based on cultural norms, values, and beliefs that are prevalent in European American, middle-class contexts. As a result, these theories and practices are less likely to resonate with or empower people from other cultural backgrounds. As our globalized society continues to evolve, it is of vital importance to uncover cultural groundings of theories and practices in regards to creativity. One crucial step toward this goal is to understand how people in different cultures feel when they engage in creative problem-solving. We use the term “designer” to refer to anyone who engages in creative processes. As a rich, dynamic, yet ubiquitous aspect of human experience, emotion can powerfully drive or derail a generation of novel ideas. A competent designer needs to understand their own and others’ emotions, as they embark on a journey to bring new ideas and products to the world. However, we currently know little about how emotion impacts creative performance across the globe. The lack of rigorous research on emotion impedes developments of sound practices to guide designers’ learning and growth and to facilitate crosscultural collaboration. We seek to advance the emerging research field centered on affect in design and contribute to design thinking research by illuminating cultural variations in designers’ affective processes in creative problem-solving. In the following sections, we first review how emotion is currently understood in design research and practices. We then introduce our theoretical perspective on the cultural shaping of emotion. After that, we report two empirical studies, which suggest that culturally normative emotions are linked to self-reported creative outcomes such as the novelty of ideas. The studies also explore cultural variations in a few other emotional tendencies in creative contexts. Finally, we discuss implications of this work and describe future directions.

2 Theoretical Background Scientific theories of emotion have evolved greatly over the past century. Throughout this article, we use the term “emotion” and “affect” to refer to people’s subjective feeling states. Classical views of emotion assume categories (e.g., anger, joy) that are distinguishable as fingerprints at surface levels (e.g., facial behavior) and/or at neurological levels (e.g., patterns of autonomic nervous system). These basic

Design the Future with Emotion: Crucial Cultural Perspectives

245

views of emotion (also known as classical, essentialist views), though intuitive for people from certain cultures, are repeatedly found problematic in light of recent research findings (Barrett, 2017). Alternative views of emotion have flourished. Appraisal views, for instance, generally expect variability in emotional responses and are agnostic to mechanistic causes of emotion. One of the most adopted views is core affect. In the core affect model, two subjective feelings are used to characterize all sorts of emotion concepts across cultures; they are considered orthogonal to each other—valence (pleasuredispleasure) and arousal (sleepy-activated). The core affect view provides a potentially more culturally responsive language to talk about emotion. Russell and Barrett (Russell, 2003; Russell & Barrett, 1999) who developed the core affect concept are ambiguous about the causes or the biological mechanisms of basic feelings. Barrett, in her and her colleagues’ recent work on the biological and neurological basis of emotion, explicitly took a complexity-embracing approach and argued that emotion is constructed in situ. This constructed view of emotion proposes that emotion emerges in a brain as it “continually makes meaning of sense data from its body and the world by categorizing those data with situation-specific concepts, thereby constructing experience and guiding action” (Barrett, 2012, 2013). In this paper, we adopt the core affect view to explore self-reported affective states in idea generation. In addition, we use the affect valuation theory to examine people’s ideal and actual affective states based on a broad set of emotion words that are considered common across cultures (Tsai et al., 2006a, b).

2.1

Emotion and Design Research

Emotion is central to discovery and invention, yet its role is often invisible. As a methodology to make explicit implicit principles of creative acts, design thinking has put an emphasis on “thinking,” not “feeling” (Camacho, 2016). This is partly due to a long-standing stigma about emotion at work, especially in male-dominant engineering professions (Picard, 1997; Adams et al., 2011). Worse than being a subordinate to design cognition, emotion is sometimes viewed with disdain (Whitfield, 2007). For decades, research on the design process has focused on deriving the rational and analytical basis. For instance, efforts to improve engineering design largely cast skepticism on intuitive design practices that rely on feelings (Ranscombe et al., 2017). By contrast, an affective basis, and its roles in design, is less acknowledged and studied (Ge et al., 2021). In emotion research within design science, researchers have primarily looked at emotion as induced through designed products (e.g., user emotion), popularized through Don Norman’s work (Norman, 2004), or as a way to understand students’ educational experiences (e.g., academic experience) (Lönngren et al., 2020). Nevertheless, there has been a burgeoning interest in studying emotion in design due to a confluence of several forces. These factors include the recent technological growth of computational emotion sensing tools and models, the rise of

246

C. Xu et al.

Table 1 Emerging research about affect and emotion in design process and behavior Theoretical perspective Basic views

Focus Mapping physio-/psychological status (e.g., facial expression) to “internal” emotional states based on fixed categories

Neuroscience views

Studying affect and emotion based on its relationship with cognition

Core affect views Constructed views

Investigating arousal and valence of feelings Studying physio-/psychological activities as constructed by situations

Mixed views

Taking mixed perspectives based on theoretical considerations or, in the cases of some papers, as a result of empirical choices rather than adequate theoretical conceptions

Examples in design research Studies on designers’ comfort using engineering equipment (Bezawada et al., 2017), emotional experience in engineering design classrooms (Villanueva et al., 2018) during CAD design (Zhou et al., 2021) “[Emotion is the] highest form of thought” (Olson, 2001). “Affective reasoning is the basis of rationality” (Dong et al., 2009). “Design physiology” as part of the cognitive process of design (Gero & Milovanovic, 2020) Group emotional valence across design stages (Ewald et al., 2019) Situated emotion of experienced designers in codesign process (Ge et al., 2021) Studies on team affective behavior to predict team performance (Jung & Leifer, 2011), software engineers’ emotion in remote collaboration (Vrzakova et al., 2020)

human-centered design (e.g., empathy development for designers), as well as new research evidence in various domains showing how emotion is intertwined with cognition, creativity, learning, and performance (e.g., Psychology: Csikszentmihalyi, 2013; Davis, 2009; Gino & Ariely, 2012, Management Science: Barsade, 2002; Learning Sciences: Pekrun et al., 2014). Researchers have studied designers’ emotion from various perspectives—as a function of context, expertise, and design phase or as internal experiences, a joint social process, or a form of thought (see Ge et al., 2021 for a review of studies). A variety of approaches have also been taken across multiple research contexts to examine different kinds of design activities. We summarize these different approaches in Table 1.

2.2

Culture and Emotion in Design

Despite a growing research interest in emotion in the creative process, little research has investigated the crucial role of culture. Current creative practices in the USA have prioritized emotion-related cultural values that are prevalent in European American contexts, while cultural values of many other groups have not received as much attention in theory and practice. As a result, certain design practices

Design the Future with Emotion: Crucial Cultural Perspectives

247

surrounding emotion developed in European American contexts may not be applicable to other cultural contexts. For instance, in their qualitative study about creative design thinking practices at IDEO, Sutton and Hargadon (1996) wrote, “Clients, like others at IDEO brainstorms, are taught to praise ideas, build on suggested ideas, be playful, wait their turn before talking, and not be critical. The justification for such guidelines is that they facilitate the flow of ideas. A side-effect is that participants are encouraged (nearly required) to express only positive feelings. If designers or clients are negative, or just look grumpy, they are gently reminded to be more upbeat.” Here, the ideation practices at IDEO prioritized the experience and expression of high-arousal, positive emotions, which tend to be the normative, desirable affective states in European American contexts. Nonetheless, people in other cultural contexts such as East Asian societies tend to value low-arousal, positive emotions or are more likely to acknowledge the coexistence of both negative and positive emotions. Hence, it is important to understand cultural variations in affective norms and values as they apply to the creative process. Emotions are “socially shared, collective scripts” (Kitayama & Masuda, 1995). People’s experience and expression of their emotions are expected to vary systematically as a function of construals of the self (i.e., cultural schemas that people apply to defining who they are) (Markus & Kitayama, 1991). In societies where the self is typically thought of as being independent from other people and from contextual factors (i.e., independent self-construal), emotion is usually thought of as “internal” and defined as generated from individuals (Barrett, 2017; Mesquita & Markus, 2004). In cultural contexts where people perceive the self as overlapping with others (i.e., interdependent self-construal), emotion tends to be associated with other people and situational factors (Masuda et al., 2008). Tsai and her colleagues also studied how emotions that people generally value and ideally want to experience (i.e., ideal affect) may vary across cultures. They suggest that ideal affect allows people to effectively socialize with others and maintain a sense of self that is generally concordant with their respective contexts (Tsai et al., 2006a, b). In particular, their work showed that high-arousal, positive affect (e.g., excitement) is generally valued in American contexts, whereas low-arousal, positive affect (e.g., calmness) is valued in East Asian contexts.

3 Research Question and Hypothesis Drawing on prior work, we ask the following research questions: (1) What affect do people in different cultural contexts actually and ideally want to experience during idea generation processes? (2) How do these different affective states relate to creative outcomes? As a starting point, we focused on comparing American and Japanese cultural contexts. We hypothesized that people overall tend to align their emotion in the creative process with “ideal affect”—feeling states that are valued by their cultures. Mesquita and Boiger (2014) argued that to the extent that certain emotions produce better

248

C. Xu et al.

outcomes within a sociocultural context, these emotions tend to be experienced more frequently. It is plausible that aligning affective experience with cultural values is associated with better performance in idea generation. For instance, whereas designers in East Asia prefer to engage in silent and reflective “brainwriting” (Ge & Maisch, 2016), designers in the USA enjoy riding on high energy to hunt for novel ideas. This difference in desirable emotion is also readily reflected by different ambiences across cultures. While the Nintendo headquarters in Japan are characterized by calmness and simplicity, Google’s workplace is full of colorful facilities for exciting, free exchange of ideas. We propose that when designers’ actual emotions are consistent with what is valued in their cultures, they are more likely to excel in creative problem-solving. Hypothesis 1a People in American contexts are more likely to experience and ideally want high-arousal, positive emotions (HAP) in idea generation processes than people in Japanese contexts. Hypothesis 1b People in American contexts are less likely to experience or ideally want low-arousal, positive emotions (LAP) in idea generation processes than people in Japanese contexts. Hypothesis 2a Experiencing high-arousal, positive emotions (HAP) in idea generation processes is more likely to predict novelty of ideas among people in American contexts than people in Japanese contexts. Hypothesis 2b Experiencing low-arousal, positive emotions (LAP) in idea generation processes is more likely to predict novelty of ideas among people in Japanese contexts than people in American contexts. In addition to exploring how culture may shape preferred levels of arousal and valence during ideation, we also investigated other dimensions of affect in creative processes, including the extent to which people experience socially engaging (versus disengaging) emotions. While socially engaging emotions (e.g., guilt and friendly feelings) reflect a desire to build, maintain, and repair one’s connection with others and with their surrounding social and physical environments broadly construed, socially disengaging emotions (e.g., anger, pride) are associated with the opposite desire (Kitayama et al., 2006). Cultural contexts consist of an unevenly distributed set of symbolic resources shared by a group of people—for example, narratives, images, and schemas related to seeking harmony can be more easily found in East Asia than in the USA. Consequently, culture holds the potential to foster different appraisals of lived experiences and produce different emotions. Hypothesis 3 During creative problem-solving processes, people in Japanese contexts are more likely to experience socially engaging emotions and less likely to experience socially disengaging emotions than those in American contexts. For exploratory purposes, we also examined a few other emotional tendencies at different stages of creative problem-solving, including cultural variations in emotion expressiveness (how easy or difficult it is for people to express emotion), emotion

Design the Future with Emotion: Crucial Cultural Perspectives

249

fluctuation (how much people’s emotions fluctuate over the course of time), emotion attentiveness (the degree to which people pay attention to their emotions), and emotion ambivalence (a tendency to experience a mix of negative and positive emotions). In the following sections, we presented two empirical survey studies testing the hypotheses. We first conducted a pilot study to simply examine potential differences in actual and ideal affect among American participants and found evidence in support of our theorizing. We then recruited a sample of American and Japanese participants and compared their affective experience and tendencies during creative problem-solving.

4 Pilot Study 4.1 4.1.1

Method Participants

We analyzed responses from 106 American adult participants (Mage (SD) = 32.56 (11.36); 45 men, 59 women, 2 other; 69 White, 11 Black, 6 Latinx, 19 Asian, 2 other). Participants were recruited from Prolific.

4.1.2

Procedure

Participants were asked to briefly write down a recent time that they came up with ideas to solve a particular problem or to create a product. They were instructed to focus on their feelings and emotions when describing their experience and were required to spend at least 1 min on this task with 30–200 words. After that, participants reported their ideal and actual affect during idea generation. The order of the questions about ideal and actual affect was randomized. Participants then evaluated their own ideas in terms how useful and novel the ideas were. In the end, they completed demographic questions including age, gender, race and ethnicity, and annual household income. They also completed a short reading comprehension question as an attention check.

4.1.3

Measures

Ideal and Actual Affect We adopted six items from Affect Valuation Index (AVI) (Tsai et al., 2006a, b). We focused on items examining ideal and actual high-arousal positive states (HAP; elated, excited, enthusiastic) and low-arousal positive states (LAP; calm, peaceful, serene). Participants answered the questions on a five-pt. scale (1 = “Not at all” to 5 = “Extremely”).

250

C. Xu et al.

Self-Appraisal of Ideas We used two items to measure self-reported qualities of ideas (Hennessey & Amabile, 2010): (1) How novel was your idea? (2) How useful was your idea? Participants answered the questions on a five-pt. scale (1 = “Not at all” to 5 = “Extremely”).

4.2

Results

We summarized descriptive statistics and correlations among variables in Table 2.

4.2.1

Differences in Actual and Ideal Affect

We conducted paired t-tests and found significant differences between ideal and actual HAP and LAP in participants’ creative problem-solving processes. As shown in Fig. 1, American participants generally wanted to have both more HAP and LAP than they actually had (HAP: t(105) = 7.68, p < 0.001; LAP: t(105) = 9.70, p < 0.001) during the process of creative problem-solving. In addition to experiencing more HAP than LAP (t(105) = 3.23, p = 0.002), American participants also desired more HAP than LAP (t(105) = 2.00, p = 0.05).

4.2.2

Affect and Self-Reported Novelty and Usefulness of Ideas

We constructed two regression models to predict self-reported novelty and usefulness of ideas with affect (actual and ideal HAP and LAP) while including common demographic variables as covariates including gender (“male” was coded as “0” and “female” and “other” as “1”), age, and annual household income (before tax; measured on a 1–11 scale with each scale representing a range based on a reasonable income distribution in the USA). We found that participants’ actual HAP, but not others (i.e., ideal HAP, ideal and actual LAP), predicted self-appraisals of idea novelty and usefulness. Table 3 presents these results.

4.3

Discussion

In pilot study, we found that American participants expressed high-arousal, positive emotions (i.e., HAP) that were generally consistent with what is considered valued in their cultural context. In addition, experiencing culturally normative emotions (i.e., HAP) predicted creative outcomes such as self-reported novelty and usefulness of ideas.

1 – 0.42*** 0.12 0.52*** 0.07 0.40*** -0.01 -0.35*** 0.17 – 0.24* 0.41*** 0.09 0.33*** 0.06 -0.15 0.28**

2

Note: Gender was coded as male = 0, female and other = 1 *p < 0.05; **p < 0.01; ***p < 0.001

1. Self-reported novelty of ideas 2. Self-reported usefulness of ideas 3. Ideal HAP 4. Actual HAP 5. Ideal LAP 6. Actual LAP 7. Age 8. Gender 9. Annual household income – 0.39*** 0.41*** 0.18 -0.12 -0.17 0.04

3

– 0.14 0.57*** 0.09 -0.29** 0.18

4

Table 2 Descriptive statistics and correlations among relevant variables in pilot study

– 0.28** -0.05 0.03 0.05

5

– 0.08 -0.12 0.28**

6

– 0.07 -0.01

7

– -0.17

8

M 2.58 3.59 3.58 2.60 3.36 2.26 32.56 0.58 6.37

SD 1.19 1.07 1.07 1.28 0.97 0.99 11.36 0.50 3.52

Design the Future with Emotion: Crucial Cultural Perspectives 251

252

C. Xu et al.

Experience of Affect (1 = Not at all and 5 = Extremely)

Affect in Idea Generation in Pilot Study 5

4 Types of Affect HAP

3

LAP 2

1 Actual Affect

Ideal Affect

Fig. 1 Ideal and actual affect during idea generation among American participants in pilot study. Error bars represented standard errors Table 3 Regression results predicting novelty and usefulness of ideas in pilot study

Ideal HAP Actual HAP Ideal LAP Actual LAP Gender Age Anual Household Income Constant Observations R2 Adjusted R2

Dependent variable Self-reported novelty of ideas (1) -0.127 (-0.343, 0.089) 0.371*** (0.165, 0.576) 0.020 (-0.208, 0.248) 0.199 (-0.055, 0.454) -0.554** (-0.966, -0.142) -0.005 (-0.023, 0.012) 0.007 (-0.051, 0.064) 2.005*** (0.890, 3.120) 106 0.344 0.297

Self-reported usefulness of ideas (2) 0.127 (-0.082, 0.336) 0.212* (0.013, 0.411) -0.038 (-0.258, 0.183) 0.120 (-0.126, 0.367) -0.016 (-0.415, 0.383) 0.004 (-0.012, 0.021) 0.062* (0.006, 0.117) 1.913*** (0.833, 2.993) 106 0.229 0.174

Note: Gender was coded as male = 0, female and other = 1. Unstandardized coefficients are presented; 95% confidence intervals are in parentheses *p < 0.05; **p < 0.01; ***p < 0.001

5 Main Study 5.1 5.1.1

Methods Participants

We analyzed responses from 127 American participants (Mage (SD) = 39.2 (10.5); 59 women, 68 men; 110 White, 13 Black, 6 Latinx, 4 Asian, 1 other) and

Design the Future with Emotion: Crucial Cultural Perspectives

253

155 Japanese participants (Mage (SD) = 41.7 (9.38); 59 women, 96 men). Participants were recruited from MTurk (US) and Lancers (Japan), respectively.

5.1.2

Procedure

Participants were first asked to write down a recent time that they came up with ideas to solve a problem or improve a situation. After that, participants were instructed to specifically describe and elaborate on their feelings as they came up with ideas, including what the feelings were like and how the feelings changed during the process. To capture a broader range of people’s affective experiences, we instructed participants to describe their subjective feeling states even when they did not readily have words for these feelings. Participants were required to spend at least 1 min on this task with 30–200 words. Next, participants reported their feelings during idea generation in a different way. They were asked to list up to five feelings that they experienced, if any, during the process of problem-solving that they provided. For each listed feeling, they were asked to rate its intensity as well as its timing relative to the whole process (i.e., early, middle, and/or end). In addition, the participants reflected on how difficult it was to describe their feelings and how much fluctuation, awareness, and ambivalence of feelings they experienced. Next, participants reported their ideal and actual affect during their processes of idea generation. The order of the questions was randomized. Following that, participants evaluated their own problem-solving outcomes. Finally, they completed demographic questions including age, gender, race and ethnicity, and annual household income. They also completed a short reading comprehension question as an attention check. American participants completed the survey in English, and Japanese participants completed the same survey in Japanese.

5.1.3

Measures

Ideal and Actual Affect We used the same measure as in pilot study, focusing on examining ideal and actual high-arousal, positive states (HAP) and low-arousal, positive states (LAP). We adopted the translation from prior work (De Almeida & Uchida, 2021) for the Japanese version. Self-Described Emotion during Idea Generation Participants listed up to five different feeling states in their own words (self-listed feelings). Participants also described their emotional experience in written forms ( free-form responses of feelings), which we subsequently coded for their levels of social engagement. Emotional Tendencies We used a few items to identify participants’ emotional tendencies including emotion expressiveness (“How easy or difficult was it for you to describe your feelings?”; 1 = “Extremely easy,” 5 = “Extremely difficult”) (reverse-recoded for subsequent analyses), emotion fluctuation (“To what extent did your feelings fluctuate as you came up with the idea?”; 1 = “None at all,”

254

C. Xu et al.

5 = “A great deal”), emotion attentiveness (To what extent did you pay attention to your feelings while you were coming up with the idea?”; 1 = “Never,” 5 = “Always”), and emotion ambivalence (“To what extent did you simultaneously experience different feelings in the process of coming up with the idea you described?”; 1 = “Never,” 5 = “Always”). Self-Appraisal of Ideas We used the same item to measure novelty of ideas as in pilot study. We also asked participants to answer the question “How good was the idea?” on a five-pt. scale (1 = “Not at all,” 5 = “Extremely”). We added this item as a more general way to gauge the quality of ideas as novelty may be valued more in American contexts than in Japanese contexts (Ge et al., 2022). For exploratory purposes, we also measured the extent to which participants believed that their ideas solved the problems they described on a five-pt. scale (1 = “Not at all,” 5 = “Extremely”).

5.2

Results

We summarized descriptive statistics and correlations among variables in Table 4.

5.2.1

Actual and Ideal Affect in Idea Generation

We conducted one-way ANOVA tests to examine group differences in actual and ideal affect. Here, we used ipsatized score to control for potential response style differences in all subsequent analyses (Tsai et al., 2006a, b). In terms of ideal affect, American participants ideally wanted to experience more HAP and more LAP than Japanese participants, Fs(1, 280) = 30.2 and 5.25, p < 0.001 and p = 0.02, respectively. In terms of actual affect, American participants reported higher actual HAP than Japanese participants, F(1, 280) = 3.23, p = 0.07. Japanese participants reported actually experiencing more LAP than American participants, F(1, 280) = 5.13, p = 0.02 (Fig. 2). To examine whether participants’ ideal and actual HAP and LAP would predict self-appraisals of idea outcomes, in particular, how novel the idea is and how good the idea is, we initially constructed linear regression models controlling for the same demographic variables as in pilot study. Due to lack of model fit, backward stepwise regression was performed separately for American and Japanese participants to include other relevant variables. Results for both groups showed that, in addition to some of the ideal and actual affect variables, the problem-solved variable (i.e., how much the problem was solved) emerged as a significant predictor. Therefore, we constructed linear regression models based on demographic variables and the problem-solved variable. As shown in Table 5, we found that for American participants, actual HAP, but not other factors (i.e., ideal HAP, ideal and actual LAP), predicted how good the idea

Idea novelty How good ideas are Ideal HAP Actual HAP Ideal LAP Actual LAP Age Gender Annual household income

1 0.39*** 0.15* 0.23*** -0.13* -0.09 0.02 -0.08 0.12* 0.18** 0.22*** 0.07 -0.05 0.05 0.04 0.16**

2

Note: Gender was coded as male = 0, female and other = 1 *p < 0.05; **p < 0.01; ***p < 0.001

1. 2. 3. 4. 5. 6. 7. 8. 9. 0.53*** -0.17** -0.21*** 0.07 -0.08 0.29***

3

-0.14* -0.11 0.15** 0.04 0.15*

4

Table 4 Descriptive statistics and correlations among relevant variables in main study

0.39*** 0.05 0.13* -0.02

5

0.06 -0.07 -0.15**

6

-0.04 0.04

7

0.19**

8

M 2.60 3.81 0.46 0.44 0.62 0.30 40.54 0.42 5.49

SD 1.12 0.91 0.58 0.66 0.59 0.62 9.95 0.49 3.02

Design the Future with Emotion: Crucial Cultural Perspectives 255

256

C. Xu et al.

a

Ideal Affect in Idea Generation in Main Study

Ipsatized score of affect

1.00

0.75 Group American

0.50

Japanese 0.25

0.00 HAP

LAP Types of affect

b

Actual Affect in Idea Generation in Main Study

Ipsatized score of affect

1.00

0.75 Group American

0.50

Japanese 0.25

0.00 HAP

LAP Types of affect

Fig. 2 (a, b) Ideal and actual affect in idea generation across American and Japanese participants in main study. Error bars represented standard errors

is perceived to be (β = 0.29, t(118) = 3.12, p = 0.008) and how novel the idea is perceived to be (β = 0.30, t(118) = 2.94, p = 0.004). Due to lack of model fit in Model 1-American, we included other ideal and actual affect variables—P (positive), N (negative), HA (high arousal), and LA (low arousal)—as additional predictors, which improved the model (adjusted R2 from 0.07 to 0.12). We found that American participants’ actual HAP (β = 0.35, t(112) = 2.87, p = 0.005), but not other affect variables, predicted self-appraisals of ideal novelty. For Japanese participants, affect related variables did not predict idea novelty; actual LAP, but not other affect variables, marginally predicted self-appraisals of how good the idea is (β = 0.12, t(146) =1.72, p = 0.09).

Design the Future with Emotion: Crucial Cultural Perspectives

257

Table 5 Regression results predicting idea novelty and how good ideas are in main study

Ideal HAP Ideal LAP Actual HAP Actual LAP Problem-solved Gender Age Annual household income Constant Observations R2 Adjusted R2

Dependent variable How novel ideas are (1) Japanese American -0.117 (-0.445, -0.403 (-0.865, 0.211) 0.059) -0.202 (-0.465, -0.466 (-0.906, -0.025) 0.062) 0.216 (-0.065, 0.520 (0.174, 0.866) 0.497) -0.070 (-0.348, 0.116 (-0.227, 0.209) 0.460) -0.004 (-0.280, 0.340 (0.185, 0.494) 0.272) -0.385 (-0.697, -0.078 (-0.509, -0.072) 0.354) 0.008 (-0.008, -0.003 (-0.022, 0.024) 0.017) -0.026 (-0.081, 0.043 (-0.029, 0.029) 0.115) 3.120 (1.690, 1.170 (0.311, 2.030) 4.570) 155 127 0.195 0.127 0.151 0.068

How good ideas are (2) Japanese American 0.054 (-0.189, -0.239 (0.297) 0.525, 0.047) -0.075 (0.198 (-0.075, 0.269, 0.120) 0.470) 0.036 (-0.173, 0.341 (0.127, 0.244) 0.555) 0.180 (-0.026, -0.171 (0.386) 0.383, 0.042) 0.608 (0.494, 0.399 (0.229, 0.570) 0.723) -0.065 (0.143 (-0.124, 0.297, 0.166) 0.410) 0.004 (-0.008, 0.001 (-0.011, 0.015) 0.013) 0.006 (-0.034, -0.003 (0.047) 0.047, 0.042) 1.200 (0.561, 2.220 1.830) (1.330, 3.110) 155 127 0.443 0.267 0.413 0.217

Note: Gender was coded as male = 0, female and other = 1. Unstandardized coefficients are presented; 95% confidence intervals are in parentheses *p < 0.05; **p < 0.01; ***p < 0.001

5.2.2

Self-Listed Feeling States during Idea Generation

We conducted sentiment analysis using feelings that participants described in their own words (i.e., self-listed feelings). Valence and arousal scores of self-listed feelings ranging from 0 (lowest) to 1 (highest) were derived based on National Research Council Canada (NRC) sentiment and emotion lexicons (Version 0.92, Mohammad & Turney, 2013). The Japanese responses were first translated using AI translator Deeply.com and then double-checked by one of our Japanese collaborators. We conducted one-way ANOVA tests and found that Japanese participants generally expressed emotions with lower valence (M = 0.35, SD = 0.17) than the American participants (M = 0.45, SD = 0.22), F(1,280) = 17.07, p < 0.001. Similarly, Japanese participants expressed emotions with lower arousal (M = 0.36, SD = 0.13) than American participants (M = 0.42, SD = 0.12), F(1,280) = 19.24, p < 0.001. Figure 3 shows the results.

258

C. Xu et al. Arousal and valence based on sentiment analysis of self-listed feelings

Level of affect (0-1)

1.00

0.75 Group American Japanese

0.50

0.25

0.00 Arousal

Valence Types of affect

Fig. 3 Cultural differences in levels of arousal and valence as indicated by self-listed feelings

5.2.3

Change in Affect across Different Stages of Idea Generation

Using the wordcloud package in R, we conducted a preliminary text data analysis to examine participants’ emotions at different stages of ideation. We removed common words (e.g., the, and, for, in, to, etc.) and starter words (e.g., “I feel,” “I had the feeling of”) and extracted word stems for frequency analysis. American participants’ average word frequency in describing feelings is higher than Japanese (t(281) = 6, p < 0.001). This result suggests that American participants are more likely to have a shared model about emotion than Japanese participants. In the American context, through socialization, people may have incorporated a set of common emotion concepts (e.g., frustration, happiness, anger) to explain their experiences and behaviors related to creative problem-solving. By comparison, Japanese may have a more diverse understanding about what emotions they associate with idea generation, and their emotional expressions tend to be less institutionalized with weaker cultural scripts to base their expression on. Figure 4 visualized frequency of word usage at early, middle, and ending stages of idea generation for the two samples, respectively. Across the board, words (including their various forms) such as frustration, happiness, proudness, worry, and relief are among the most common (i.e., highly frequent) descriptions of feelings. In the early stage, “worried” was frequently mentioned among Japanese participants, whereas “frustration” was more frequently expressed by American participants. While worry and frustration could both be categorized as low-arousal, negative emotions, they differ in terms of social engagement (i.e., whether emotions tend to connect or disconnect people with others and their surrounding environments). We interpreted this pattern as indicating that Japanese participants experienced more socially engaged emotions that connected them to other people and their

Design the Future with Emotion: Crucial Cultural Perspectives

259

Fig. 4 Word clouds of listed feelings across different stages of idea generation for Japanese and American samples. Larger fonts indicate higher frequency

contexts. At the ending stage, Americans also reported a greater feeling of pride than Japanese. This pattern is consistent with prior research suggesting that Americans tend to express emotions that attribute more agency to the independent self (e.g., pride) than Japanese (Imada & Ellsworth, 2011).

5.2.4

Cultural Variations in Socially Engaging Emotion during Ideation

Two of the authors coded participants’ free-form responses of feelings based on their expressed levels of social engagement (Markus & Kitayama, 1991). Responses expressing one’s inner feelings independent from others were coded as “1,” and responses expressing feelings related to social engagement in reference to others were coded as “0.” We found 41 socially engaging feelings out of 155 responses (25.8%) in the Japanese sample, compared with 13 out of 127 responses (10.2%) in the American sample. A chi-squared test of independence was performed to examine the relation between cultural group and levels of social engagement versus disengagement regarding emotion. Overall, Japanese participants were more likely to construct their emotions in reference to other people than were American participants, χ2 (1, N = 282) = 10, p = 0.001. Examples of interdependent feelings from the Japanese sample included the following: “I wonder how my friends are feeling” and “. . .I was so angry. . .[but] I managed to adjust it because of the other people involved in the project.” Examples of interdependent feelings from the American sample include the following: “I felt really happy with myself when I came up with this idea. I thought others would be proud of me and that I could help others at work with my idea,” and “I felt pretty

260

C. Xu et al.

great contributing to the whole team and providing them with ideas that made the supervisor very happy.”

5.2.5

Other Dimensions of Affective Tendencies during Idea Generation

To examine differences in emotional tendencies, we conducted a one-way multivariate analysis of variance (MANOVA) combining all four DVs (emotion expressiveness, emotion fluctuation, emotion attentiveness, emotion ambivalence) as dependent variables and found a significant multivariate effect of participants’ cultural backgrounds, Pillai’s trace = 0.15, F(4, 276) = 12, p < 0.001, partial η2 = 0.15. Taken separately, this was only significant for emotion expressiveness, F(1, 279) = 45.7, p < 0.001, and emotion fluctuation, F(1, 279) = 4.28, p = 0.039. This means that Japanese participants found it more difficult to express emotion and experienced a higher level of fluctuation in their emotional states than did their American counterparts. We examined how participants’ emotional tendencies may predict outcomes including how novel and how good they perceived their ideas to be. First, we conducted an exploratory correlation analysis as shown in Table 6. The correlation analyses show that emotional tendency variables differentially predicted idea generation outcomes for these two samples. For instance, emotion fluctuation was positively associated with qualities of ideas and learning among Japanese participants, whereas it only predicted learning for Americans. Furthermore, we again constructed regression models to predict the three outcome variables (e.g., novelty of ideas, how good the ideas were, and levels of learning) using these four emotion tendency variables while controlling for common demographic characteristics including age, gender, and annual household income and how much their problem was solved. Overall, we found that only emotion fluctuation and emotion attentiveness predicted Japanese participants’ self-appraisal of idea novelty ( ps = 0.08 and 0.03, respectively). By comparison, none of these emotional tendencies predicted outcome variables for American participants.

5.3

Discussion

In this study, we found evidence supporting H2a that experiencing HAP during idea generation is more likely to predict novelty of ideas for people in American contexts than those in Japanese contexts, despite no significant difference in actual HAP was found across the two groups. This finding suggests that Americans may use their own emotion to judge ideas whereas Japanese may take into considerations multiple factors (e.g., others’ opinions) and rely less on their own emotions to evaluate ideas. We did not find evidence supporting H2b. Yet we found that for Japanese, experiencing LAP during idea generation tended to predict how good the idea was. It is important to note that these two cultures may place relatively different

Note: *p < 0.05; **p < 0.01; ***p < 0.001

How novel was my idea? How good was my idea? How much did I learn during this process?

Emotion fluctuation American Japanese 0.02 0.27*** 0.00 0.17* 0.19* 0.31*** Emotion attentiveness American Japanese 0.13 0.24** 0.20* 0.14 0.14 0.31***

Table 6 Correlations between emotional tendencies and various dependent variables in main study Emotion ambivalence American Japanese 0.05 0.25** 0.02 0.19* 0.15 0.32***

Emotion expressiveness American Japanese -0.02 -0.10 0.23** 0.07 0.05 0.00

Design the Future with Emotion: Crucial Cultural Perspectives 261

262

C. Xu et al.

emphases on how important it is for an idea to be novel. Hence, a question such as “how good is this idea?” may prove to be a general way to gauge the quality of ideas as valued by people in different cultural contexts. Results of exploratory text and sentiment analyses were generally consistent with respective cultural models of emotion in these two contexts. We found people in Japanese contexts were more likely to experience socially engaging emotions and less likely to experience socially disengaging emotions than those in American contexts. In addition, a series of emotional tendencies (e.g., ambivalence, fluctuations) showed different associations with the self-appraised creative outcomes for these two samples.

6 General Discussion Overall, emotion-shaping through culture has been under-recognized in creative research and practices. Popular practices tend to prescribe a particular emotional route to ideation in the USA—high-arousal, positive emotions (e.g., excitement). However, we argue that people’s experiences and expressions of emotion in creative problem-solving are influenced by historically derived cultural norms and values. While an excitement-oriented affective route is consistent with cultural norms regarding emotion in the European American contexts, people in many interdependent contexts ideally want to experience low-arousal, positive emotions, and they normatively acknowledge mix-valuenced emotions. Furthermore, people in these contexts often construct their emotion in reference to other people or to their socio-physical surroundings. Hence, they are more likely to experience socially engaging emotions that motivate building and maintaining social connection. As shown by our empirical data, compared with American participants, Japanese participants reported more frequent experiences of low-arousal, positive emotion (e.g., calmness) as well as more socially engaging emotions in an idea generation task. Our findings serve to remind designers of the critical role emotion plays in provoking a generative mind state. We also encourage future researchers to further explore how culture underpins expressions of emotion and why and in what ways emotion can be leveraged to make creative practices culturally responsive.

6.1

Limitations

We acknowledge some important limitations of the current work. First, our study designs were correlational in nature and we used self-reported creative outcomes, which are the participants’ own subjective evaluations of their ideas. However, given that emotion casts an influence on people’s cognitive processes, it is plausible that experiencing positive emotion can lead people to appraise their ideas more positively (these associations, however, are likely to be culturally variable as well). In the

Design the Future with Emotion: Crucial Cultural Perspectives

263

meanwhile, having positive idea generation outcomes may produce positive emotion. Although participants were asked to report their emotion during idea generation, they may not have perfectly separated emotion at different stages. Hence, it would be important to design studies to determine the causal mechanism and to use other ratings for qualities of ideas. Second, in this study, we did not address the within-culture diversity in American society. The foregoing theorizing regarding high-arousal, positive emotion as the ideal affect mainly applies to middle-class, European American contexts. Different sociocultural groups in the USA likely have different norms and values regarding emotional experience and expression. It would be critical to replicate our findings with more diverse samples to include clear comparisons between European American participants (as opposed to Americans in general) and participants in different East Asian societies (e.g., Korean, Chinese). Third, there is immense heterogeneity in how creativity manifests in different domains. For example, artistic creativity may involve different affective experience from pragmatic problem-solving creativity. Our study design used a recall paradigm to ask participants to describe any recent creative problem-solving experience. Thus, it is possible that our American and Japanese samples may have provided ideas for solving different problems, which may in turn account for different affect being reported. Fourth, we used a one-time, self-reported measure of emotion in our studies. However, given that emotion is fleeting and dynamic, it would be important to collect data at various points using designs such as repeated measure studies with experience sampling methods and adopt multiple diverse measures of emotion (e.g., physiological measures).

6.2

Implications

Emotion profoundly influences people’s cognitive processes, including idea generation, judgment, and decision-making (Lerner et al., 2015). Hence, attending to how emotion affects creative outcomes pinpoints a promising new route to encourage people to come up with better ideas to solve problems in their respective contexts. Research on affect in design can shed light on how to foster an environment conducive to idea generation in a culturally resonant way. For instance, should a particular social and physical environment for idea generation be designed to evoke high-arousal, positive (versus low-arousal, positive) emotion? Should educators highlight and encourage emotion ambivalence or fluctuation? How does expressing one’s or acknowledging others’ emotions affect the quality of ideas generated? Insights into managing the affective dimension of creative problem-solving could bolster creative performance. Highlighting emotion as a cultural product can also guide people to incorporate diverse cultural values in designing products with a direct link to emotion. At a

264

C. Xu et al.

fundamental level, the very concept of emotion is culturally variable. But such variability in emotional experience has not been well reflected in design practices. For instance, there has been a lot of enthusiasm about building artificially intelligent models that detect and perform emotion. Current conceptual models of emotion that guide the algorithmic developments are likely to be based on theories of emotion connected with cultural assumptions in European American contexts (White & Katsuno, 2022). Popular theories about emotion (e.g., basic emotion) adopted in affective AI products tend to encode a universalistic fallacy that people around the globe experience a set of similar, “basic” feelings. Nonetheless, such AI products may be less effective or biased when used in diverse cultural contexts. A cultural consideration of affective AI products is important to ensure equitable design that benefits a broader range of the population across the globe. Finally, emotion is an intuitive way for people to empathize with and relate to others. It is thus a key factor to fostering effective cross-cultural collaboration. For example, in multicultural teams, people from different cultural backgrounds are likely to experience and exhibit different emotional patterns when interacting with each other. A lack of cultural perspective and knowledge about emotion can cause confusion and conflict to the detriment of team cohesion and performance. On the other hand, adequately understanding of cultural underpinnings of affect would encourage team members to better negotiate communication norms. Such a scenario would encourage people from minoritized backgrounds to voice their viewpoints.

6.3

Future Direction

It would be interesting to examine whether experiencing culturally counternormative emotions may yield better creative performance under certain circumstances as well. This is because emotions that do not conform to cultural norms may be less frequently experienced and hence may elicit more novelty in thoughts. Future research could examine this possibility by treating creative processes as a continuous experience and taking a longer horizon in study designs. Another fruitful direction is to examine cultural norms regarding the expression of emotion. Culture can be viewed as a toolkit and instruments (e.g., languages) for people to encode and express their emotion. However, languages are powerful tools for activating existing cultural values and may constrain the experience and expression of emotion. Many fleeting emotions may not be readily recorded and thereby may go unnoticed, as participants do not have the tools to capture their own subjective experience or to share it with others. These subjective experiences, however, may present a window to capturing novelty. For instance, Isbister et al. (2006) explored a body-based, nonverbal means for evaluating a system’s affective impact on users. The authors created a set of sensual evaluation instrument objects as a way to measure affect. Their approach presents a way to record affect by bypassing languages as a constraint.

Design the Future with Emotion: Crucial Cultural Perspectives

265

Future research can investigate if providing participants with the means to record their emotion may lead participants to generate more ideas that are novel. For example, can integrating visual means (e.g., drawing and sketching) or movements (e.g., dancing) based on people’s emotion boost performance in creative problem-solving? Likewise, it is worthwhile to explore if creating novel symbols for recording emotion can serve as vehicles to enhance awareness of one’s subjective experience and help facilitate the expression of novel ideas.

7 Conclusion Emotion is a ubiquitous part of the creative process in different societies. The ways that people experience and express emotion are profoundly shaped by the particular cultural contexts they are in. We seek to advance nascent field of scientific research on the cultural shaping of emotion in creative activities. Our exploratory work has shown that Japanese and American participants report distinct emotional patterns, which then differentially predict self-appraised creative outcomes. We call for future work to build upon our findings and further investigate how understanding emotion can pinpoint new ways to encourage creative problem-solving in different cultures.

References Adams, R., Evangelou, D., English, L., De Figueiredo, A. D., Mousoulides, N., Pawley, A. L., Schiefellite, C., Stevens, R., Svinicki, M., Trenor, J. M., & Wilson, D. M. (2011). Multiple perspectives on engaging future engineers. Journal of Engineering Education, 100(1), 48–88. Barrett, L. F. (2012). Emotions are real. Emotion, 12(3), 413–429. Barrett, L. F. (2013). Psychological construction: The Darwinian approach to the science of emotion. Emotion Review, 5(4), 379–389. Barrett, L. F. (2017). How emotions are made: The secret life of the brain. Palgrave Macmillan. Barsade, S. G. (2002). The ripple effect: Emotional contagion and its influence on group behavior. Administrative Science Quarterly, 47, 644–675. Bezawada, S., Hu, Q., Gray, A., Brick, T., & Tucker, C. (2017). Automatic facial feature extraction for predicting designers' comfort with engineering equipment during prototype creation. Journal of Mechanical Design, 139, 2. Camacho, M. (2016). David Kelley: From design to design thinking at Stanford and IDEO. She Ji: The Journal of Design, Economics, and Innovation, 2, 1. https://doi.org/10.1016/j.sheji.2016. 01.009 Csikszentmihalyi, M. (2013). Flow: The psychology of happiness. Random House. Davis, M. A. (2009). Understanding the relationship between mood and creativity: A meta-analysis. Organizational Behavior and Human Decision Processes, 108, 25–38. De Almeida, I., & Uchida, Y. (2021). Who can buffer marginalization risk? Affect experience, affect valuation, and social marginalization in Japan and Brazil. Frontiers in Psychology, 12, 501165. Dong, A., Kleinsmann, M., & Valkenburg, R. (2009). Affect-in-cognition through the language of appraisals. Design Studies, 30(2), 138e153.

266

C. Xu et al.

Ewald, B., Menning, A., Nicolai, C., & Weinberg, U. (2019). Emotions along the design thinking process. In Design thinking research (pp. 41e60). Cham: Springer. Ge, X., Leifer, L., & Shui, L. (2021). Situated emotion and its constructive role in collaborative design: A mixed-method study of experienced designers. Design Studies, 75, 101020. Ge, X., & Maisch, B. (2016). Industrial design thinking at Siemens Corporate Technology, China. In Design thinking for innovation (pp. 165–181). Springer International Publishing. Ge, X., Xu, C., Furue, N., Misaki, D., Lee, C., & Markus, H. R. (2022). The cultural construction of creative problem-solving: A critical reflection on creative design thinking, teaching, and learning. In Design thinking research: Achieving real innovation (pp. 291–323). Cham: Springer International Publishing. Gero, J. S., & Milovanovic, J. (2020). A framework for studying design thinking through measuring designers’ minds, bodies and brains. Design Science, 6. Gino, F., & Ariely, D. (2012). The dark side of creativity: Original thinkers can be more dishonest. Journal of Personality and Social Psychology, 102, 445. Hennessey, B. A., & Amabile, T. M. (2010). Creativity. Annual Review of Psychology, 61, 569–598. Imada, T., & Ellsworth, P. C. (2011). Proud Americans and lucky Japanese: Cultural differences in appraisal and corresponding emotion. Emotion, 11(2), 329–345. Isbister, K., Höök, K., Sharp, M., & Laaksolahti, J. (2006). The sensual evaluation instrument: developing an affective evaluation tool. In Proceedings of the SIGCHI conference on Human Factors in computing systems (pp. 1163–1172). Jung, M. F., & Leifer, L. J. (2011). A method to study affective dynamics and performance in engineering design teams. In DS 68–7: Proceedings of the 18th International Conference on Engineering Design (ICED 11), Impacting Society through Engineering Design, Vol. 7: Human Behaviour in Design, Lyngby/Copenhagen, Denmark, 15.-19.08. 2011. Kitayama, S., & Masuda, T. (1995). Reappraising cognitive appraisal from a cultural perspective. Psychological Inquiry, 6(3), 217–223. Kitayama, S., Mesquita, B., & Karasawa, M. (2006). Cultural affordances and emotional experience: Socially engaging and disengaging emotions in Japan and the United States. Journal of Personality and Social Psychology, 91(5), 890–903. https://doi.org/10.1037/0022-3514.91. 5.890 Lerner, J. S., Li, Y., Valdesolo, P., & Kassam, K. S. (2015). Emotion and decision making. Annual Review of Psychology, 66, 799–823. Lönngren, J., Adawi, T., Berge, M., Huff, J., Murzi, H., Direito, I., . . . Sultan, U. (2020). Emotions in engineering education: Towards a research agenda. In 2020 IEEE Frontiers in Education Conference (FIE) (pp. 1–5). IEEE. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98(2), 224. Masuda, T., Ellsworth, P. C., Mesquita, B., Leu, J., Tanida, S., & Van de Veerdonk, E. (2008). Placing the face in context: Cultural differences in the perception of facial emotion. (Vol. 94) (No. 3). American Psychological Association. doi: https://doi.org/10.1037/0022-3514.94.3.365. Mesquita, B., & Boiger, M. (2014). Emotions in context: A sociodynamic model of emotions. Emotion Review, 6(4), 298–302. Mesquita, B., & Markus, H. R. (2004). Culture and emotion: Models of agency as sources of cultural variation in emotion. In Feelings and emotions: The Amsterdam symposium (pp. 341–358). Mohammad, S. M., & Turney, P. D. (2013). Nrc emotion lexicon. National Research Council, Canada, 2, 234. Norman, D. A. (2004). Emotional design: Why we love (or hate) everyday things. Civitas Books. Olson, G. B. (2001). Brains of steel: Mind melding with materials. International Journal of Engineering Education, 17(4/5), 468–471. Pekrun, R., Hall, N. C., Goetz, T., & Perry, R. P. (2014). Boredom and academic achievement: Testing a model of reciprocal causation. Journal of Educational Psychology, 106, 696.

Design the Future with Emotion: Crucial Cultural Perspectives

267

Picard, R. W. (1997). Affective computing. MIT press. Ranscombe, C., Kinsella, P., & Blijlevens, J. (2017). Data-driven styling: Augmenting intuition in the product design process using holistic styling analysis. Journal of Mechanical Design, 139, 11. Russell, J. A. (2003). Core affect and the psychological construction of emotion. Psychological Review, 110 , 145–172. (: American Psychological Association). Russell, J. A., & Barrett, L. F. (1999). Core affect, prototypical emotional episodes, and other things called emotion: Dissecting the elephant. Journal of Personality and Social Psychology, 76, 805. Sutton, R. I., & Hargadon, A. (1996). Brainstorming groups in context: Effectiveness in a product design firm. Administrative Science Quarterly, 685–718. Tsai, J. L., Knutson, B., & Fung, H. H. (2006a). Cultural variation in affect valuation. Journal of Personality and Social Psychology, 90, 288–307. Tsai, J. L., Knutson, B., & Fung, H. H. (2006b). Cultural variation in affect valuation. Journal of Personality and Social Psychology, 90(2), 288. Villanueva, I., Campbell, B. D., Raikes, A. C., Jones, S. H., & Putney, L. G. (2018). A multimodal exploration of engineering students’ emotions and electrodermal activity in design activities. Journal of Engineering Education, 107(3), 414e441. Vrzakova, H., Begel, A., Mehtätalo, L., & Bednarik, R. (2020). Affect recognition in code review: An in-situ biometric study of reviewer’s affect. Journal of Systems and Software, 159, 110434. White, D., & Katsuno, H. (2022). Artificial emotional intelligence beyond east and west. Internet Policy Review, 11(1), 1–17. Whitfield, T. A. (2007). Feelings in design–A neuroevolutionary perspective on process and knowledge. The Design Journal, 10(3), 3–15. Zhou, J. J., Phadnis, V., & Olechowski, A. (2021). Analysis of designer emotions in collaborative and traditional computer-aided design. Journal of Mechanical Design, 143, 2.

Part 4

Design Thinking Best Practices and Strategy

Opportunities and Limitations of Design Thinking as Strategic Approach for Navigating Digital Transformation in Organizations Annie Kerguenne, Mara Meisel, and Christoph Meinel

Abstract Digital transformation is fueled by the growing offer of digital technology, promising new business models, and growing revenues. The inevitable systemic change needed for digital transformation accompanied by the dynamic conditions for such is what contributes to the growing complexity of transformation projects in organizations, bringing new opportunities and hurdles for the people affected. This complexity results in most digital transformation activities falling short of their goal and leaders not being able to create the desired impact. There is evidence that design thinking has the potential to significantly contribute to the field of digital transformation. This research project aims at getting a better understanding of how design thinking as strategic approach can contribute to the success of digital transformation activities and, if so, to what extent. With a mixed-method approach, including a literature review and qualitative research, we conducted three case studies. Our learnings show that digital transformation is very individual to the respective organization regarding scope, motivations, goals, and starting points. Therefore, there is no general or linear approach to digital transformation processes. Our findings show that it can start anywhere and includes forward as well as backward steps. The individual transformation needs of organizations are also based on the high complexity of transformation processes and activities. Orchestrating complex transformation projects requires a systematic, clearly structured, and integrated process. This process needs both diverging and converging activities in the areas of analysis, vision, learning, and diffusion. These activities enable ambidextrous navigation through all transformation areas. To overcome transformation hurdles, the constant integration of human, technology, and system perspective is crucial to mirror the interrelation of all three systems. We found that the organizational culture plays a significant role in successful transformation projects. On a

A. Kerguenne (✉) · C. Meinel Hasso-Plattner-Institut, Prof.-Dr.-Helmert-Straße 2-3, Potsdam, Germany e-mail: [email protected]; offi[email protected] M. Meisel Hasso-Plattner-Institut, Prof.-Dr.-Helmert-Straße 2-3, Potsdam, Germany e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_14

271

272

A. Kerguenne et al.

strategic level, the cultural digital transformation maturity can serve as decision reference to define starting point, metrics, and individual identification. On a systemic level, we learned that methods are less likely to be transferred into the organizational value creation practice than concepts/principles (as individual lessons learned). Our cases show the vulnerability and fragility of transformation processes: On an operational level, transformation gets blocked by daily business or urgent matters leading to teams hitting the breaks in transformation activities. To make our research actionable, we developed an integrated and clearly structured digital transformation process and transferred the findings of this project for creating an adaptive digital transformation strategy kit, that aims at closing the research-practice gap and offers practitioners a scientifically substantiated strategic decision support on how to navigate their own digital transformation endeavors.

1 Introduction Today, the opportunities for creating new business value are within easy reach for most organizations. At the same time, these opportunities have proven rather elusive. Digital technology offers speed and scalability that drives successful innovation and adaptivity. This continues to create major disruptions in how businesses work in terms of new offerings, customer and partner relations, infrastructure, and organizational structure. Accordingly, today’s leaders are charged with navigating the dynamic complexity that emerges from the nexus of digitalization, organizational change, as well as turbulence in wider business environments. So, more than ever, leadership requires a holistic multi-perspective approach rather than a singleperspective one. There is evidence that design thinking can support the navigation in dynamic complexity. Since the Harvard Business Review identified design thinking as an “approach to devise strategy and to manage change” (Kolko, 2015), the thought school has made considerable inroads into the service portfolios of the top global consultancies (Pello, 2018). Advances have also been made into the list of digital technologies, tools, and methods currently used by organizations with successful transformations (de la Boutetière et al., 2018). Yet, there is evidence that the challenge of “How to bring it home?” remains unsolved. The dilemma Larry Leifer perceived regarding design thinking in 2012 implies a tension between disruptive solutions and the organizational challenges of integrating these new, disruptive solutions at both organizational (i.e., structure and procedural) and individual levels (Design Thinking DTINGRE, 2012). We call the result (or, indeed, the lack of results) of this tension as “disruption paralysis.” Disruption paralysis can be measured: “70% of Digital Transformations fall short of their objectives. . .[because] delivering such fundamental change at scale in large, complex organizations is challenging, especially with short-term pressures”(Forth et al., 2020). This translates into approximately 900 billion annual budget resources wasted in 2020 in the USA (Tabrizi et al., 2019). We therefore hypothesize that the main challenge in value creation in the digital era is balancing the capabilities of generating high-value

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

273

disruptive innovations with the ability to “bring them home” by embedding them in an existing organizational structure. Our goal is to explore the opportunities and limitations of design thinking for navigating digital transformation and examine how these could be grounded by what we know about how design thinking works and why. Linking Research to Practical Transformation Activities Matters The initial motivation for the HPDTRP is clearly stated by Hasso Plattner: “We must strive to improve our understanding of innovation processes in order to bring better solutions to society more quickly.” The better we understand the things we do, the less (time, money, energy) waste we produce in achieving good results. Creating successful innovations is already a difficult task embedded in multiple layers of uncertainty (Jalonen, 2011). If we consider digital transformation as a system of nested innovations, this challenge grows in complexity and volume. “Digital is not just a thing that you can buy and plug into the organization. It is multi-faceted and diffuse and doesn’t just involve technology. Digital Transformation is an ongoing process of changing the way you do business. It requires mixing people, machines, and business processes, with all the messiness that it entails. It also requires continuous monitoring and intervention, from the top, to ensure that both digital leaders and non-digital leaders are making good decisions about their transformation efforts” (Davenport & Westerman, 2018). Digital transformation recalls the David against Goliath story with its corresponding message. The bigger and more uncontrollable the challenge, the more precise one should be able to act on the leverage points for the desired impact. We assume that research could have an important impact on equipping actors in digital transformation with well-founded instruments and strategies to identify the areas of action for their specific digital transformation journey and to decide for actions that work. Research insights translated into concrete action guidelines for digital transformation become like David’s knowledge about Goliath’s most sensitive impact point. In this sense, our project intends to continue closing the research-practice gap. Furthermore, researchers say that “the development and application of new DT methods, tools, and frameworks often lack the foundation of rigorous research, and research insights seldom get implemented to inform practice”(J. A. Edelman et al., 2021). Looking back at the history of design thinking, this research-practice gap seems less like a conscience decision of keeping these two areas of knowledge and action apart, but more like a logical result of the methodological diffusion process. Julia von Thienen states in her work about the design thinking foundations that “Design Thinking (. . .) was primarily an export of practices. The available theories were maintained only inhouse (. . .) yet these theories are invaluable in helping Design Thinking practices become fully understandable and applicable with greatest mindfulness and intentionality” (J. P. A. von Thienen et al., 2021, p. 10).

274

A. Kerguenne et al.

2 Research Question and Research Design The proposed project aims to answer one overarching research question: What are the lessons learned for the potential and limitations of design thinking methods and concepts as a strategy for (planning, developing, navigating, integrating) successful digital transformation activities and programs in organizations?

An effective answer to this question will lead to less waste of human, technological, and financial resources and support practitioners as well as coaches in digital transformation processes. Concrete we want to contribute to solving this challenge: How might we help leaders and their teams of teams overcome the “disruption paralysis” and support them in planning, developing, navigating, and integrating successful digital transformation activities and programs in their organizations?

2.1

Research Design: Method Mix for a Practice-Oriented Action Research

Our concept is informed by an initial literature review and empirical data (interviews). Those built the basis for determining suitable conceptual frameworks and a practice-oriented starting point for defining the main terms and concepts of digital transformation, design thinking (as strategic approach), and transformation processes. To build a first concept for a practice-oriented transformation framework, we explored the challenges of digital transformation from a practitioner’s perspective in an expert workshop with innovation and transformation specialists (strategic design, engineering, IT and media, organizational development, AI research) and by analyzing 180 practitioner challenges brought into design thinking executive education programs from 2014 to 2022. By screening selected business magazines and consultancy publications, we identified the main digital transformation hurdles and failure reasons. In a second step, we reviewed existing design thinking methods and, based on their application potential, assigned them to the different phases of the transformation framework draft. For that purpose, we connected the challenges and hurdles with relevant lessons learned from research results of the HPDTRP reports of 2012–2021. With a focus on HPDTRP research, we aim not only to make design thinking research accessible but also to create a meaningful structure for this knowledge pool after 13 years of the program running. This literature based on the impact of design thinking in different contexts was extended to research project findings from the wider HPI design thinking researcher community as well as standard references in the respective areas of transformation. To close gaps and to create tangible prototypes for comprehensive transformative actions in all areas, we extended the method pool. We adapted actions from related fields, such as management, organizational sociology, future studies, and sustainability. The research findings were transferred into the design of a complete digital transformation

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

275

strategy kit with phase explanations cards, action step-by-step cards, and inspirational impulse cards. In parallel, we started our case study research (Yin, 2017) gathering and analyzing the data (interviews, research workshops, and company publications) in the three organizations. In our concept testing phase, we focused on the design of tailor-made transformation action sets for distinct transformation tasks of our research case partners. We gained a deeper understanding of our initial research findings and hypothesis in real-world application. As a final step, we submitted the completed digital transformation strategy framework and kit to the digital transformation experts (whose input helped set the starting point of the research phase) to get their evaluation regarding the remaining gaps, further development potential, and usability of key activity modules. By embedding these findings in structured sets of transformational actions, they can be used in organizational development curricula, either taught in education institutions or run by respective stakeholder (e.g., leading human resources or IT departments) inside of organizations (Fig. 1).

3 Design Thinking for Digital Transformation Today 3.1

Our Perspective on Design Thinking

During an interview at the d.confestival in 2012, Hasso Plattner demonstrates his pragmatic yet deep understanding of the design thinking capacity in his words, “The method is based on common sense” (Potsdam Eins, 2012). This describes the potential of design thinking that made it one of the most popular methods to create innovations and finally led to the holistic application as an approach “to devise strategy and manage change” (Kolko, 2015). Applying the core principles of design, such as user-need focus, multi-perspectivity, or fine-tuning concepts by controlled experiments for business solutions, inspired Hasso Plattner, who became for design thinking what the de Medici had for the Renaissance: namely, an enabler, multiplier, and legitimation agent. From an entrepreneurial perspective, what could make more sense than capitalizing on IT concepts by using them for different user needs. Maybe this was one of the motivations for Plattner to fund the pioneering education institutions at Stanford University in 2005 and, later, 2008 in Potsdam at the HPI. From this practical and application-oriented point of view, design thinking encompasses not just the methods but also the specific attitude toward solution finding (the mindset) and the different concepts or strategic principles that practitioners derive from the lessons of distinct applications. The complete design thinking repertoire is an interplay of methods, mindset, and principles. In applying the methods, users grow a design thinker mindset, which sets the frame for developing individual principles for application in further contexts (Fig. 2). In fact, the historical evolution of design thinking from a method to a mindset and strategic principles appears to be the logical result of an ongoing learning process

Fig. 1 Method mix for a practice-oriented action research

276 A. Kerguenne et al.

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

277

Fig. 2 Experience-based learning according to D. Kolb and Fry (1974)

based on experiences. Kolb’s experience-based learning cycle represents the theoretical concept supporting this assumption (D. A. Kolb, 1984; D. Kolb & Fry, 1974). The experience: Transformation pioneers experience the value of a method, e.g., user interviews, by asking open question, inquiring about good and bad experiences, exploring the underlying reasons for positive or negative emotions, and, perhaps, making a discovery, for instance, finding out that a young person does not enjoy doing schoolwork when it is parent-monitored because this gives them the feeling of defeat at not being taken seriously. This surprised the researchers who had assumed students would more likely view such monitoring to the establish orientation and structure. The reflection: When compared to the usual batteries of closed questions used in classic market research, the empathic technique of placing oneself in the user’s shoes through open questions and follow-ups allows a deeper dive into unknown motifs—as well to track them. These new findings offer good starting points for innovative, user-centered solutions. The concept formation: A principle or concept is derived from the reflection that serves as a concrete guideline in other situations. Whenever it is a matter of exploring the deeper reasons for a certain behavior to find new solutions to problems (whether with colleagues, family members, or criminal offenders), the principle of empathy offers a successful formula. Experimentation: Other related questions are now dealt with and problems solved with this principle. These mark the beginning of the learning cycle as concrete experience. In this way, strategic principles for acting in new contexts are gradually developed over time. The direct experience as a basis for learning and the active role of the learner are two principles that are deeply anchored in the design thinking approach. The innovators themselves go into the field to research the needs of the users. They analyze what they have found in a team and develop solution hypotheses, which they test as quickly as possible with the users as rough prototypes. From the users’ reactions, they learn how the solution should be further developed to finally give a valid answer to a real user after several iteration loops.

278

A. Kerguenne et al.

Design thinking can thus also be described as an experience-based learning process, applying methods to synthesizing concepts or principles that are then transferred to new contexts when working on complex challenges.

3.2

Our Perspective on Digital Transformation

When we talk about digital transformation, we refer to a practitioner’s perspective, which is characterized by an integrative and process-related view on the technology aspects of digitization, digitalization, and digital transformation. In practice, these processes are considered interconnected stages that occur seamlessly within organizational transformation. Verhoef et al. (2021) define digitization, digitalization, and digital transformation as follows: Digitization is the translation of analog information into digital information. While mainly aimed at changing an organization’s documentation process, it does not influence the value creation. Digitalization uses digital technology to support and optimize an organization’s business process. Digital transformation describes change across the whole organization, affecting all areas of business, ultimately resulting in new business models (Verhoef et al., 2021). Therefore, digital transformation builds on the effects of digitization and digitalization and their distinct impact on the organization. In the digital transformation process, a highly complex network involving different stakeholders, processes, products, and business models (Schallmo et al., 2017, p. 5), the main driver is not the technology, but the integration of social, technological, and business-related goals and actions into a strategy (Kane et al., 2015). Unruh and Kiron (2017) describe the mechanism of getting from digitization to digitalization and to digital transformation in practice as an iterative and interconnected evolution process. Digitization triggers digitalization which inspires digital transformation. The process then results in impulses for further changes. This cycle mirrors the experiences of practitioners in daily business: You digitize a paper process like the circular folder in administration. Other areas, from further sectors in administrative work through human resources management, can then be digitalized. The former HR “black box” transforms into a transparent and comprehensible system that motivates new governance processes on system level, for instance, the formal establishment of diverse career pathways and transparent compensation concepts that follow a binding development logic. Hence, the technology-driven transformation becomes the driver of transformation in the connected human (social) and organizational strategy systems (Kane et al., 2015), which unleashes dynamics on a larger system level that lead to increasing “wickedness” (Rittel & Webber, 1973) of digital transformation processes. When practitioners position digital transformation projects into strategic planning typologies, such as the Cynefin framework (Snowden, 2000) or the Stacey matrix (Stacey, 2007), they are usually classified into the “complex” domains of the frameworks. These are connected to strategies based on learning by experimenting forward (Cynefin) or design thinking (Stacey).

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

279

As an approach to grasp the essence of complex concepts and abstract concepts like innovation and transformation, a legitimate starting point is the etymological roots: Etymology dictionaries inform us that in the middle of the fifteenth century, the word innovacion meant “restoration, renewal” and that the Late Latin innovationem (nominative innovatio) was a noun of action from the past-participle stem of innovare meaning “to change; to renew.” This word came from in- or “into” + novus or “new,” meaning “a novel change, experimental variation, new thing introduced in an established arrangement.” The interesting aspect here is that the integration aspect—often neglected when we use the word innovation—is already anchored in the word roots (Innovation | Etymology Dictionary, n.d.). Let’s have a look at the historical provenience of transformation: It relates to the Old French transformation and directly from Church Latin transformationem (nominative transformation), meaning “change of shape.” It is a noun of action from the past-participle stem of transformare or “change in shape, metamorphose” (Transformation | Etymology Dictionary, n.d.). Building on these definitions, transformation can be described as a metamorphose of an established system due to the introduction of experimental variations (=innovations).

This is a description that reflects the characteristics of other transformation processes in nature, psychology, or biological evolution. The pattern of transformation seems to follow the integration of new elements in an established system. This is where we get first hints about the hurdles of digital transformation processes. For the successful integration of new elements, there must be a connection point between the system and the new element. In addition, when compared to isolated innovation, these processes take time. As stated in 2017 in an article by Forbes Magazine, transformation describes the ongoing processes that happen “after innovation has entered the scene.” Transformation processes take time and require establishing new organizational processes and structures and strategic planning as a result of moving from one state to another. Where innovation means challenging the norm, transformation means the drive toward establishing new norms (Newman, 2017). Reflecting on research work that takes a meta-perspective on transformation with sustainability-based concepts, a successful transformation process requires an interplay of technology/innovation, economy, culture, and institutions. This implies that navigating a transformation process requires a specific set of capabilities, knowledge, and mindset—it thus requires a certain transformative literacy (Schneidewind, 2018). Accordingly, we describe digital transformation from an application perspective as a cultural transformation task and ongoing learning process (Tabrizi et al., 2019). This work focuses on a specific perspective: Digital transformation is a transformative learning process located in a complex and dynamic system of interrelated social (human), technological, and organizational value creation systems.

Precisely, this systemic complexity is why we assume design thinking, with its human-centered and system-related approach, may have an untapped positive

280

A. Kerguenne et al.

impact worth being excavated and specified. We ask the following specified research question: How do we use design thinking methods and concepts in a workable, learning-based transformation process that maps both parts of the metamorphose, the experimental variations (=innovations), and their integration in an existing system?

4 A Theoretical Framework for Digital Transformation Our chosen perspective considers each digital transformation process as a learning process that involves both the creation of the experimental variation and the integration in an existing system. Transformation models focus on either the digital perspective or the social perspective (Haas et al., 2022; Nagel & Wimmer, 2014; Schneidewind, 2018; Wade et al., 2020; Westerman et al., 2011). Schneidewind’s model has been built with reference to Kolb’s experienced-based learning cycle and thus has the capacity to represent the main stages of a learning flow: (1) understanding of present system, (2) envisioning of desirable transformed future, (3) creation and testing of experimental variations, and (4) integration of system changes. Created as a structured approach to overall societal change toward a sustainable economy, the framework emphasizes the present necessity of change to achieve a sustainability-centered future. Schneidewind emphasizes the capabilities of politics, civil society, business, science, and pioneers of change to address necessary transformation processes from a vision and mindset of sustainability. He states that transformation processes need to integrate different types of knowledge, that is, knowledge about the current system, goal/desired state (a vision), and further possible futures, as well that about the transformation itself. This includes knowledge created through experiments. On an individual level, a mixture of knowledge, mindset, and capabilities is needed for a successful learning process. This concept refers to the multilevel perspective needed to understand the complexity of transformation processes, which can be transferred to organizations. The landscape level describes overarching structures and developments, the regime level describes institutional structures, and the niche level describes the specific “place” where innovation and change takes place. Schneidewind takes a systems thinking approach on change levers (meadow) in the sense of that many distinct changes on the small scale (niche) might add up to transform regimes (structures) and landscape (the whole); vice versa, changes on the small scale are strongly supported by a fitting support on the other levels. The drivers of change in Schneidewind’s model are experiments: They allow us to learn and acquire the necessary transformational knowledge in the down part of the cycle and represent real interventions as “realworld laboratories” (Schäpke et al., 2018; Schneidewind, 2018). Transferred to the context of digital transformation in organizations, these real-world laboratories can take over the function of creating transformation knowledge in the interrelated human (social), technological, and organizational system perspectives. Unlike the standard design thinking process that represents a structured way to find innovative

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

281

solutions starting with understanding a specific problem in a system to experiment forward a human-centered solution, the transformation framework offered by Schneidewind structures the complete transformation cycle taking a systems perspective. We took this process concept as a reference and starting point for further research in HPDTRP literature, practice-focused publications, and case research. It is a transformation process with four stages, each representing one action area: analysis, vision, learn, and diffusion. To integrate the different social, technology, organizational, and context systems that are involved in digital transformation, we decided to build on the well-known innovation diagram that represents the three factors of successful innovation. These are desirability, feasibility, and viability, introduced by Brown and Katz (2011), and we talk about the human, the technology, and the organizational system perspective, completed by the context perspective that represents the larger meta-systems (ecological, political, economic). In this early research step, one preliminary assumption was made that was based on the theoretical reflection above, on the result of the first expert workshop with innovation and transformation specialists (strategic design, engineering, IT and media, organizational development, AI research), as well as on our practical experience in the coaching of transformation projects. This assumption was that the classical design thinking innovation process works well for running the transformational experiments in the real-world laboratories located in the learn stage (Fig. 3).

4.1

The Process Framework: An Integrative Modular System for Ambidextrous Navigation

For specific questions, a digital transformation strategy kit shall be able to provide answers, and we analyzed 186 practitioner challenges brought into design thinking executive education programs from 2014 to 2022. Learning 1: Digital transformation is diverse—scope, motivations, goals, and starting points vary from organization to organization.

Digital transformation in organizations manifests itself in new ways of value creation. It can be defined as new ways of value creation in the organizational ecosystem with new services/products, new internal infrastructure, and new relationships to partners and clients (Foley et al., 2017; Weinreich, 2016). Completing the analysis of the practitioners’ challenges, we considered the data of the interviews we conducted with leaders in digital transformation (research case interviews and leader interviews in further companies) and saw that the entry points and motivations for digital transformation activities could come from different value creation perspectives. This means focusing on the human value perspective when transformation activities were driven, for instance, by the human resources department with a transformation goal in the area of new internal processes and infrastructure. The system value perspective, combined with a strong technology value perspective, is the driver for the leaders of corporate incubators or innovation labs aiming to

Fig. 3 Innovation vs. transformation process (own representation, based on HPI D-School and Schneidewind, 2018)

282 A. Kerguenne et al.

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

283

Fig. 4 Diversity of entry points and motivations of digital transformation activities

develop new products, services, or business models. The system value perspective is often the initial motivation behind creating new relationships to customers and partners. It then evolves and links with both other perspectives during the transformation process (Fig. 4). The practitioner challenges could be clustered in all four action areas of our chosen process framework, with stronger concentration in the analysis area. A double-review process ensured minimization of bias during the clustering process. Going deeper into the different challenge essences, we identified two overarching questions per phase that served as a subcluster for specifying assigned actions: Analysis Vision Learn Diffusion

How can we understand ourselves as a team and our specific system? How can we capture our driving and challenging dynamics? How can we see our (digitally enhanced) futures? How can we plan our exploration journeys? How can we explore to learn what works for us? How can we unleash our change dynamics? How can we cultivate our adaptive culture? How can we establish our flexible organizational and operational structure?

The following Table 1 shows exemplary challenges in each area: Learning 2: Digital transformation does not follow a linear step-by-step process—in real life, it can start anywhere and includes forward as well as backward steps.

The distribution of the project challenges as well as the “forward-and-backward” movement of the transformation projects over the entire process mirrored a real-life experience in transformation project coaching as well as in our three research cases. Organizations are very often already in the middle of transformation processes when they start to look for strategic support—they don’t start necessarily in the analysis stage. In some cases, we observed that several projects with transformational impact had been run during the several years, leading to the need of a strategy based on

284

A. Kerguenne et al.

Table 1 Exemplary practitioners’ challenges defined in transformation projects

Phase Analysis

Vision

Learn

Diffusion

Number of challenges 59

45

43

39

Practitioners’ transformation challenges—examples Having common understanding of the problem space as a team. Defining the status quo of our international collaboration. Understanding our stakeholders their challenges I see the biggest barrier in overcoming established behavioral patterns that do not support the company’s culture. . . the biggest opportunity is bringing people together despite cultural differences and regional distances The future consulting approach means understanding the expected change, find out the key levers to attract the target group, find out which expectations and needs might occur, match those with our opportunities Setting up a research plan for functional prototypes (feasibility). Evaluation of business model/consumer profile/experience briefing on the opportunity Potential improvements for the innovation process, desirability of the product program, safeguarding turnover Implementation of an innovation process—Shape the way forward; create ideas to make it successful; point out the real success factors Get a better understanding for the needs of our subscribers and develop a club/membership scheme for the multiplication of our subscribers To implement the change as smooth as possible; common design and commitment of a set of rules for cooperation

Industry European food company

Overarching question How can we understand ourselves as a team and our specific system?

Global robotics company

How can we capture our driving and challenging dynamics?

European consultancy

How can we see our (digitally enhanced) futures?

Global beverage company

How can we plan our exploration journeys?

Global automotive company

How can we explore to learn what works for us?

German financial services company

How can we unleash our change dynamics?

German publishing company

How can we cultivate our adaptive culture?

European consultancy

How can we establish our flexible organizational and operational structure? (continued)

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

285

Table 1 (continued)

Phase

Number of challenges

Practitioners’ transformation challenges—examples

Industry

Overarching question

und communication together with the staff, coordinated resource allocation and aligned monitoring criteria and processes Sum

186

synthesis and respective rollout plans for scaling and anchoring. In another case, organizational leaders would start with the redesign of their governance concept to set the basis to better respond to their original purpose and vision and realize that they would have to go back and understand the team dynamics in the leadership team. Managers might also have been assigned to build up a corporate innovation lab to structure the innovation and transformation projects for institutionalizing the company’s innovation capacity and culture. Although being in a learn action area, they would realize that essential business model-related factors and cultural characteristics had to be analyzed before being able to move on with implementation activities. Learning 3: Digital transformation needs ambidextrous navigation. The processes require both diverging and converging activities to move forward while overcoming transformation hurdles.

Strategy consultancies represent a valid content source for practice-related data about digital transformation hurdles, as their business models are built on producing answers and services for the needs of organizations in value creation processes. By screening selected business magazines and consultancy publications, we identified main hurdles of digital transformation failure that we could allocate to the different phases (Table 2). Taking a closer look through the design thinking lens at the required thinking and working modes for answering the overarching questions and overcoming the hurdles, we identified the same “rhythm” that drives the classical design thinking process. Consequently, we propose a transformation process with four action areas and eight phases—two per action area—that capture the exploration (divergence) and synthesis (convergence) needed within each action area. The phases capture the two crucial ways of dealing with knowledge and processing it, namely, a diverging phase of creating multiple insights, exploring, etc. and a converging phase to close, condense, synthesis, enable planning, strategy, and decision-making. This way, each action area offers room to explore and create knowledge as well as to enable decision-making and moving on in the process. The phases are informed by the practitioner’s transformation project challenges, the hurdles from a business magazine and consultancy perspective, and insights from qualitative interviews. The phases can be explained as follows:

286

A. Kerguenne et al.

Table 2 Overview of important hurdles in digital transformation processes as stated by consultancies and business magazines Phase Kickoff

Status quo

Envision

Plan

Hurdles—the business and consultancy’s perspective Poor diversity of leadership teams leads to less innovation and less investment in digital technology for digital transformation that generates revenue If middle managers are not on the same page with each other and with senior leaders from the start, digital transformation is 77% less likely to be successful than those instances in which leaders had cohesion When digital technology is not integrated into culture, transformation is more likely to fail If organizational status quo of mindset, structures, and processes isn’t identified before starting digital transformation, flaws can be enhanced by digitalization

Disagreement of management on goal is one of the main failure factors in digital transformation. Define opportunities and problem you want to solve and how the company will build the organization around the desired outcome before investing in implementation Organizations where management did not establish a clear change story for transformation are more than three times less successful than those that did. If a change story does not communicate clear targets for KPIs and the transformation’s timeline, it misses the most supporting factors Wrong KPIs doom digital transformation: KPIs related to technological capabilities just lead into digitization of the same business model, but not to digital transformation of culture One-size-fits-all solutions in the name of “best practices” brought from the outside do not answer specific challenges of organizations; and

Source Lorenzo, R., Voigt, N., Tsusaka, M., Krentz, M. & Abouzahr, K. (2018, January 23). How Diverse Leadership Teams Boost Innovation. The Boston Consulting Group Werner, R., Streubel, H., Lovich, D. & Halverson, J. (2021, December 07). When Leaders Say They Are Aligned—But aren’t. The Boston Consulting Group Tabrizi, B., Lam, E., Girard, K., & Irvin, V. (2019, march 13). Digital Transformation Is Not About Technology. Harvard Business Review Digital Article Hemerling, J., Kilmann, J., Danoesastro, M., Stutts, L., & Ahern, C. (2018, November). It’s Not a Digital Transformation Without a Digital Culture. The Boston Consulting Group Sutcliff, M., Narsalay, R., & Sen, A. (2019, October 18). The Two Big Reasons That Digital Transformations Fail. Harvard Business Review de la Boutetière, H., Montagner, A., & Reich, A. (2019, October 19). Unlocking success in Digital Transformations. McKinsey.

Tabrizi, B., Lam, E., Girard, K., & Irvin, V. (2019, march 13). Digital Transformation Is Not About Technology. Harvard Business Review Digital Article Forth, P., Reichert, T., de Laubier, R., & Chakraborty, S. (2020, October 29). Flipping the Odds of Digital Transformation Success. The Boston (continued)

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

287

Table 2 (continued) Phase

Hurdles—the business and consultancy’s perspective isolated transformation strategies are among the main failure factors in digital transformation

Experiment

Implementation

Cultivate

Governance

Transformation is less successful, if groups are not enabled to apply the scientific method, using observation, hypothesis, and experimentation and speed and effectiveness are blocked Without feedback to guide change, data-driven decision-making during transformation projects is blocked. Promising proofs of concepts don’t gain traction in implementation because of the “transformation chasm”—Due to underestimation of transformation flaws in the interrelated areas of strategy, operating model, innovation agility, technology capability, and most importantly people and culture Technology is important, but the people dimension (organization, operating model, processes, and culture) is the determining factor. Organizational inertia from deeply rooted behaviors is a major impediment. Organizations that did not support transparency by implementing digital tools for more accessible information were only half as successful than those who did this Organizations that did not support participation by establishing digital self-service technologies for employees and partners’ use were only half as successful as those who did

The 70% of businesses that do not transform successfully do not leverage fast—And simplified—Decision-making. They use the wrong KPIs to monitor progress toward outcomes

Source Consulting Group Schrage, M., Muttreja, V., & Kwan, A. (2022, March 8). How the Wrong KPIs Doom Digital Transformation. MITSloan Management Review Rethinking digital Transformation— New Data Examines the Culture and Process Change Imperative in 2020. (2020). Harvard Business Review Analytic Services

Rethinking Digital Transformation— New Data Examines the Culture and Process Change Imperative in 2020. (2020). Harvard Business Review Analytic Services Long, G. (2021, October 13). The answer is scale. . .what’s the question? The industry X Magazine (Accenture). Forth, P., Reichert, T., de Laubier, R., & Chakraborty, S. (2020, October 29). Flipping the Odds of Digital Transformation Success. The Boston Consulting Group

de la Boutetière, H., Montagner, A., & Reich, A. (2019, October 19). Unlocking success in Digital Transformations. McKinsey Blount, S., & Carroll, S. (2017, May 16). Leading Teams Overcome Resistance to Change with Two Conversations. Harvard Business Review Digital Transformation Refocused: New Goals Require New Strategies. (2022, May 12). Harvard Business Review Analytic Services Schrage, M., Muttreja, V., & Kwan, A. (2022, March 8). How the Wrong KPIs Doom Digital Transformation. MITSloan Management Review

288

A. Kerguenne et al.

Analysis Action Area • Diverging phase: kickoff to set off the transformation team; analyze infrastructures, stakeholders, and value creation processes in the organization; explore organizational purpose and values; as well as assess the digital diffusion. • Converging phase: status quo to consolidate the transformation journey, define internal and external drivers and blockers for digital transformation and evolution, and evaluate cultural maturity. Vision Action Area • Diverging phase: Envision to explore desirable futures, craft compelling goals and visionary stories for the transformation process, understand potential impacts and interrelations to define the transformation vision, and elaborate strategic spaces of opportunities. • Converging phase: Plan to prioritize what to learn and how, define a clear process and conditions for transformation experiments, and agree on desired outcomes and measurements for success. Learn Action Area • Diverging phase: experiment by finding relevant human-centered sources of inspirations, ideate, and prototype for testing and learn from feedback. • Converging phase: implement by conceiving a strategy that evaluates desirability, technological feasibility, organizational viability, and overarching opportunity of your experiments and plan your pilot projects, metrics, and capability development for implementation. Diffusion Action Area • Diverging phase: Cultivate to grow a self-energizing, participatory culture; set up a feedback system for scaling decisions and run and pivot transformational pilot projects. • Converging phase: Governance to build a transparent organizational structure that supports ongoing learning; defines clear processes, functions, and roles with vertical and horizontal connections; establishes a control, measurement, and reward system that supports driving cultural transformation factors and fuels an adaptive common behavior and culture (Fig. 5). The final process framework now integrates divergent and convergent phases in a structured yet flexible pathway that enables ambidextrous navigation in digital transformation. It is designed in a circular shape so that the ongoing nature of digital transformation is supported.

To further define actions that would use the impact of design thinking to energize the digital transformation process, we recurred to the HPDTRP knowledge pool. We identified 85 articles within the HPDTRP reports from 2011 to 2021 with potential contributions to answering the overarching questions and overcoming the hurdles.

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

289

Fig. 5 The process framework for flexible ambidextrous navigation of digital transformation

During an expert workshop at the HPDTRP community building, we extended the literature base with more recent suggestions from HPDTRP researchers. We identified design thinking drivers for digital transformation actions for each of the eight phases. In the area of analysis, we mostly found relevant resources regarding setting up teams (e.g., Balters et al., 2021; de Paula et al., 2022a; Dow et al., 2012; Noweski et al., 2012; von Thienen et al., 2012). Overall, less literature provided answers on how to analyze and understand the status quo (e.g., Elsbach & Stigliani, 2018; Ney & Meinel, 2019). In the action area of vision, the biggest gap could be identified in the phase of Envision with the lowest amount of relevant literature (e.g., Gabrysiak et al., 2011). In the phase of Plan, more resources could be identified on planning and measuring design thinking experiments (e.g., de Paula et al., 2022b; Guentert et al., 2014; Haskamp, 2021; Sheppard et al., 2021). In the learn action area, the highest number of relevant literature was found, especially in experimentation (e.g., Björk et al., 2010; Dow et al., 2010; Hölzle & Rhinow, 2019; Jobst & Meinel, 2014; Royalty et al., 2021; Traifeh et al., 2019). This phase relates the most closely to design thinking practices. Successful implementation has grown in importance to design thinking in the recent years, shown by the growing number of relevant research in the field of experimentation (e.g., Haskamp, 2021; Haskamp et al.,

290

A. Kerguenne et al.

2021; Lorson et al., 2022; Marx et al., 2022; Mayer et al., 2021; Royalty et al., 2021; Royalty & Roth, 2016). In the action area of diffusion, the amount of relevant literature has also grown in recent years. Relevant sources have grown, particularly in the phase of Cultivate with a focus on scaling (e.g., Haskamp, 2021; Haskamp et al., 2021; Marx et al., 2022; Mayer et al., 2021; J. P. A. von Thienen et al., 2021). Aspects of the Governance phase remain gaps in the literature landscape of design thinking research, as only few articles related to the questions of this phase (e.g., Beyhl et al., 2014; Santuber, 2019; von Thienen et al., 2021) exist.

5 Design Thinking Opportunities and Implications for a Modular Digital Transformation Strategy Kit Before building the comprehensive strategy kit, we analyzed the in-depth qualitative semi-structured interviews conducted with leaders from case partner organizations and other organizations. Those leaders had attended one of the design thinking transformation and leadership programs in the last 8 years. Learning 4: Methods are less likely to be transferred into the organizational culture and value creation processes than design thinking mindset elements and concepts/principles (as individual lessons learned) unless they are used for institutionalizing innovation processes.

One year after attending the design thinking training, the alumni would most likely remember methods with catchy names like the “Drama-Hero-Curve” (a storytelling method) or the “Oops-a-Daisy-Framework” (a method for structuring interview data). Most interviewed leaders referred to design thinking principles like “storytelling to involve the collective mind,” “empathy to step in the shoes of a user for inspiration,” or “prototyping to accelerate decisions.” Exception: Leaders who were in immediate need of a structure for innovation as part of the corporate transformation strategy directly implemented the design thinking process with its methods along with other agile approaches, like Lean and Scrum (Rhinow, 2018). Based on this learning, we connected research-based leverage points to define the transformative actions. The leverage points were derived from existing literature of the HPDTRP research base and evidence from adjacent research fields. The following figure shows how the elements of the strategy kit are connected: Starting with the process phase and respective tasks, we defined the challenges and hurdles, as well as leverages and optional actions. On a more in-depth action level, we provide the concrete description of the levers linked to a relevant research insight followed by a step-by-step process to execute the action (Fig. 6).

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

291

Fig. 6 The adaptive digital transformation strategy kit: anatomy of a research-practice bridge (example for Implementation phase)

292

A. Kerguenne et al.

6 Toward an Adaptive Digital Transformation Process Framework for Navigating Digital Transformation in Organizations In our concept testing phase, we focused on the design of tailor-made transformation action sets for distinct transformation tasks of our research case partners, which we used in digital transformation research workshops. Beforehand, we had defined the specific position of the teams within the digital transformation process framework and had developed the agendas according to the respective phases, defined case partner challenges, and goals while considering the matching research-practice bridges (Table 3). Our analysis of three cases looked at the factors we believe to be constitutional for identifying the opportunities and limitations of design thinking in navigating digital transformation. On a more informed stage of the research endeavor, we wanted to understand the blockers and drivers of digital transformation in contexts of different organizational cultures to test design thinking concepts/principles and respective action set. We assumed these to have a positive impact on overcoming the lockers and magnifying whether the drivers of digital transformation would work or not work. To create a starting point for defining the organizational status quo, we used the adaptive cultural maturity model (ACM) for navigating digital transformation in organizations proposed by Ney et al. (2022). The framework is based on a synthesis of the maturity factors in current digital maturity models (Buvat et al., 2017; Hölzle & Rhinow, 2019) which are combined with the anthropologist perspective of organizational cultures (Douglas, 1986). Five “generic factors” were presented for self-assessment of the leaders in our interviews: 1. Human centricity, manifested by how important user/customer/employee perspectives are for the running of the organization and how user/customer/ employee needs impact the organization (product and services, administration, informally, etc.) 2. Collaboration, manifested by how important collaboration is for the organization, what collaboration means for the organization, how collaboration has changed in the organization over the past 5 years, and what factors ensure successful collaboration. Table 3 Sources of evidence used in case research Documents Interviews Research workshops

Internal and external communication and publications (press releases, intranet, social media posts, website) Interviews with leaders Spoken and written manifestations of members of case organizations in form of workshop documentations

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

293

3. Iteration-driven, manifested by the importance of experiments and feedback in the product/service life cycle and how the leaders know when they have solved a problem or responded to a challenge. 4. Digital responsiveness, manifested by the importance of technology to the mission of the organization, the most relevant technological developments, the responsibility regarding the social impact of IT implementation, and how the role of technology in the organization has changed over the past 5 years. 5. System minded, manifested by the relevance of wider economic, societal, or environmental considerations to the organization and importance and relevant awareness of activities in other parts of the organization for the work and how the leader’s work impacts the overall organizational value creation. Furthermore, the leaders were asked about company beliefs and cultural values that were implicitly or explicitly referred to by decision takers and top management to legitimize rules, structures, roles, and processes of decisions and resources attribution. In the interpretation of the interview data by researchers, the cultural theory categories (“archetypes”) were taken as reference to deepen the understanding of the organization specificity and to give orientation in the strategic design of transformational action sets in the testing workshops. According to Ney et al. (2022), “Cultural Theory locates four viable organizational archetypes along the grid and group dimension. Each of these four ways of organizing consists of mutually reinforcing social relations, beliefs, and practices. Hierarchies feature highly stratified social structures supported by elaborate rules-systems as well as measures to sanction transgression and reward obedience. These organizational cultures help bring about ‘reliability.’ Markets or individualism provides individuals with the liberty to determine and negotiate transactions with others based on mutual selfinterest. Individualist organizational cultures are geared toward bringing about ‘validity.’ Sects or egalitarianism offers individuals closed communities based on values of equality and justice. Isolation or fatalism is a way of organizing in which trust in others is rare; therefore, interaction is likely to be short-lived and strictly opportunistic. In this sense, organizations are not merely patterns of transactions or resource networks but also ways of perceiving and acting upon the world.” (Fig. 7). Our research data shows that the cultural factor was mentioned repeatedly as one of the main digital transformation hurdles stated by business consultancies and business magazines. Such is reflected in a Capgemini survey from 2019: “62 percent of respondents consider culture as the number one hurdle to Digital Transformation” (Bohn et al., 2018). We had explored this field as a connected research project during the period October 2021–June 2022 (Table 4). In addition to the cultural maturity scan, the following criteria were considered constitutional for the dynamics of digital transformation for each case: 1. The area in which digital transformation in the company system first surfaced (department, line, function) and initial growth objective regarding the three basic perspectives we chose to consider in digital transformation processes: human value, technology value, economic value.

294

A. Kerguenne et al.

Fig. 7 Cultural Map: cultural theory categories according to Douglas (1986)

2. Transformation challenge. 3. The self-assessment of cultural identity and maturity for digital transformation according to the ACM model with generic maturity factors as well as individual cultural core factors (Ney et al., 2022). 4. The interpreted cultural worldview (interpretation by researcher based on Douglas, 1986). 5. The stated transformation vision. 6. The transformation process phases. 7. Main transformation results on human (social/cultural) level. 8. Main transformation results on technology level. 9. Main transformation results on organizational system level. 10. Perceived hurdles and drivers of taken transformation process. 11. Applied design thinking principles/concepts and perceived impact. 12. Design thinking limitations/completing thought schools, disciplines, and methods (Fig. 8). The case partners and their respective transformation challenges are described as follows (Fig. 8).

1. National for-profit organization insurance services 2. National publicly owned organization energy 3. European for-profit organization safety and hygiene

NF/IS NPO/EN EF/SandH

>14.000 >400 >11.000

Fig. 8 Overview cases regarding initial overarching question and main phases of adaptive digital transformation framework

Opportunities and Limitations of Design Thinking as Strategic Approach. . . 295

296

A. Kerguenne et al.

The following Table 4 shows the three companies’ transformation cases in comparison. Table 4 Overview of case studies in comparison

Case/criteria Area of entry/initial goal for digital transformation

Challenge

Main cultural identity factors

National publicly owned organization energy NPO/EN > 400 • Planning and consultancy department. • Internal processes and infrastructure. • Intersection of human value and system value: Unleashing the self-motivation and intrapreneurial mindset of teams for avoiding resource waste in project management.

European for-profit organization safety and hygiene EF/SandH >11.000 • IT department. • Internal processes and infrastructure. • System value: Digitalization to secure and grow business today and tomorrow.

How can we understand ourselves as a team and our specific system to enable collaborative and resultfocused adaptivity and efficient project run in extreme organizational growth (rapidly increasing number of employees and size of projects with high level of complexity)? Generic factors • Human centered: 35% (externally and internally). • Collaborative: 40%. • Iteration-driven: 25%. • Technology responsive: 75%. • System minded: 80% (world), 20% (internally). Company-specific factors • Performance driven: 70%.

How can we establish an agile governance concept (while running daily business projects)?

Generic factors • Human centered: 35% (internally). • Customer centered: 25% (externally). • Collaborative: 50%. • Iteration-driven: 25%. • Technology responsive: 30%. • System minded: 60% (world), 30% (company). Company-specific factors • Cost efficiency: 100% stability—20%.

National for-profit organization insurance services NF/IS >14.000 • Corporate innovation and transformation lab. • Internal processes and infrastructure connected to services and product portfolio. • Intersection of human value and system value: Transformation of business model to secure survival of company in digitally driven highly competitive market dynamics. How can we explore/ learn what works for establishing a corporate incubator and lab to create future value for the company?

Generic factors • Human centered: 80% (internally). • Customer cenetered: 25% (externally). • Collaborative: 60% (internally), 40% (externally). • Iteration-driven: 50%. • Technology responsive: 80% (security of data). • System minded: 80%. Company-specific factors (continued)

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

297

Table 4 (continued)

Case/criteria

National publicly owned organization energy NPO/EN > 400

European for-profit organization safety and hygiene EF/SandH >11.000

• Best quality: 80%. • Team spirit: 80%.

Cultural worldview and resulting strategic need

Individualism with egalitarian characteristics, market-driven Strategic need for controllable collaborative work processes and team structures

Individualism, strongly hierarchy- and market-driven Strategic need for participative transformation actions, user-need understanding and teambuilding routines

Transformation purpose/vision

“Effective climate protection is a matter of bundling forces efficiently” (organization) “We believe that climate protection needs easy-to-apply processes and a good functional team play” (strategy department)

“Recognized leader in sustainable, innovative, and digital solutions for all customers” (organization) “Carrying EF/SandH safely to a data-driven future “(IT department)

Transformation process phases

• Kickoff and Envision: Teambuilding workshop, definition of team’s transformation mission, vision, and “hot topics” of transformation. • Plan: Upskilling training leadership team and team members in strategic design thinking and agile project management methods (OKR). • Experiment: Selection of strategic key projects and

• Kickoff: Teambuilding workshop, definition of IT team’s mission and “hot topics” of transformation. • Status quo: Cultural self-assessment workshop based on maturity model. • Envision: Vision workshop to design the ideal future IT structure and processes. • Plan: Exploration research workshop to define prototypes

National for-profit organization insurance services NF/IS >14.000 • Security: 90%. • Continuity/reliability: 80%. • Community spirit: 90%. Egalitarian, hierarchical orientation Strategic need for repeatable work processes that enable both self-direction of lab teams and focus on measurable experimentation results “We want to be the voice of social cohesion and mutual responsibility because confraternity achieves more No defined digital purpose/vision on organizational level “Challenging the status quo (of classical automotive insurance category) and evolving from an insurance company to a solver of daily problems for our clients” (lab) • Envision: Definition of department’s mission, values, vision. • Experiment: Running several innovation projects, one for 5 years. • Implement: Co-creation workshop to learn from experiments, spot process gaps, and plan optimized lab process for ongoing and future projects.

(continued)

298

A. Kerguenne et al.

Table 4 (continued)

Case/criteria

Results human level

National publicly owned organization energy NPO/EN > 400

European for-profit organization safety and hygiene EF/SandH >11.000

development as exploration labs to learn for scaling. • Implement: Visualization of transformation factors and roadmap to create a common orientation artifact that helps navigate through huge complex projects. • Co-creation of team purpose and vision for common orientation and as a basis for selfdirection. • Establishment of monthly milestone team workshops as a combination of design thinking method training and project work. • Integrating high frequency feedback mechanisms in daily work for continuous learning and optimizing.

stages and testing periods. • Experiment: Unplanned rapid experimentation in context of organizational crisis. • Implement: Participative redesign of IT governance concept.

Results technology level

• Implementation of digital collaborative project work management systems matching, with selected agile project management methods.

Results organizational system level

• Implementing agile project management methods (OKR). • Implementation of regular knowledge

• Common definition of starting point and goal; co-creation of transformation vision in workshops with cross-functional leadership teams from technology and business perspective. • Internal and external storytelling about strategic goal, collaborative process, and status quo of transformation. • Participative development and testing of hybrid linear and agile governance concept prototypes. • Integration of new intersection role to enable efficient communication between IT and business. • Change of IT processes from technology focus to user and user-product focus.

• Establishment of strategic key visual to communicate interrelation of people, processes, and technology

National for-profit organization insurance services NF/IS >14.000

• Team alignment on corporate lab process strengths and weaknesses. • Collaborative mapping existing project on digital transformation process framework and decision about future redesign needs.

• Corporate lab’s scope is focused on digital services for business model transformation. So far, five digital services projects are work-inprogress. • Work-in-progress: Development of proposed adaptive digital transformation process and action sets (continued)

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

299

Table 4 (continued)

Case/criteria

Hurdles

Design thinking leverage impact Limitations of design thinking; completing thought schools, disciplines, and methods

National publicly owned organization energy NPO/EN > 400

European for-profit organization safety and hygiene EF/SandH >11.000

National for-profit organization insurance services NF/IS >14.000

sharing and feedback formats. • Integrating “synapses”-roles into overall team structure.

systems for organizational transformation. Design of usercentered IT governance concept including structures, roles, processes, KPIs • Complexity of IT restructuring while running daily business. • Communication hurdles between business and IT department team members. • Team dynamics. • Cyber-crisis.

for standardizing corporate lab workflows.

• Empathy, teambuilding, diversity, analogizing. • Structural governance concept redesign with classical frameworks (RACI) that were adapted with the help of design thinking-based user research and prototyping.

• Empathy, iteration, systems thinking.

• Leadership conflict of rapidly growing project scopes and team sizes on one hand and time needed for redesign of team governance concept on the other hand. • Orientation gap of team members in change period of restructuring and implementation of new project management methods. • Overall time pressure that would not allow additional time investment for collaboration and learning phases. • Empathy, teambuilding analogizing, systems thinking. • Operationalizing and measurement of vertical and horizontal alignment for timeefficient teamwork. • Management-byobjectives method (OKR) used to steer collaboration work and measure teamwork results.

• Lack of structure and standards for corporate lab processes.

• Classical design thinking process does not map the innovation and the integration process. • The necessity of stakeholder involvement from the beginning on. • Design thinking does not help in legal questions of venture development. • Business frameworks (e.g., SWOT, Cynefin); areas like foresight and strategic project management.

300

6.1

A. Kerguenne et al.

Case Study 1: From Siloed Best Quality Performer to Collaborative Knowledge Provider for Complex Need Systems (National Publicly Owned Organization Energy NPO/EN > 400)

Organization and Background Motivation NPO/EN had grown from approx. 200–400 employees within the past 2 years. Operating as expert hub for energy consultancy, project development, and network building at the intersection between public, industry, and private stakeholders, the organization had to cope with its increasing importance, new dynamics, and growing work scope complexity. The director of one of the largest departments in the organization, the consultancy department, aimed to unleash self-motivation and the intrapreneurial mindset of teams in the context of this dynamic growth. The upper middle management actor (I3), who started the digital transformation activities, attended a design thinking leadership program at the HPI Academy in 2019. Transformation Entry Area in Company System The consultancy department started their transformation program by focusing on the internal infrastructures. Initial Goal for Digital Transformation The main reason for the transformation efforts, as quoted by the department director, was “. . .to connect the knowledge and motivation of each of us to form a connected think and do tank with maximum capacity to solve our multifaceted challenges. What we sell is knowledge, expertise, and solutions for a highly complex topic. To increase the value-for-money benefit for all our stakeholders, we want to and we must bundle our forces.” This translated into a mixed initial growth objective of human and organizational system (business) value. Challenge The leader described the challenge the organization faced as follows: “With smaller teams and projects, our structures and processes were working perfectly. But growing at this high pace and facing continuously new challenges, we were confronted with the fact that we could not scale our existing organizational architecture. We would move on and deliver—but the price we had to pay was more work, less time to deliver, and an increasingly critical impact on our team spirit. We already used digital collaboration and project management tools like Google Teams and Stackfield, but that didn’t really help. We did not even have the time to properly understand and implement them.” They started with the aim to get a clear understanding about themselves as a team and their specific organizational system. Questions they asked themselves were, for instance, “which stakeholders are involved in our value creation processes and how do the different team roles and interactions in project work respond to the defined deliveries?” or “what part of our work can we plan in advance and where do we have to act on demand?” Generic Cultural Factors The self-assessment of the five cultural factor values (by the director) showed the lowest values in iteration-driven (25%) and human centered (35%). “We expect high resolution results, close to perfection. Although

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

301

‘first small seeds’ would appear with a C-level actor showing an innovator’s mindset, it still lacks a consistent rollout with a method or a strategy.” Human centeredness was rated the same value internally and in the work with stakeholders, yet for different reasons. “We have been working for such a long time with our partners, that we think that we always know what their needs are. But we don‘t really ask ‘why,’ when things don‘t work out as planned.” In her directly controllable field of responsibility, the director mentioned positive experiences in testing solutions and receiving feedback from customers and partners: “Internally, we do have a lot of surveys about employee satisfaction—what lacked in the past was the synthesis and the resulting actions. We are getting better at this, as current results such as in the ‘Great Places to Work’ rating show.” The collaboration factor (40%) was evaluated as being “on a good way: We evolve from classical hierarchical and in-transparent decision mechanisms to more collaboration and sharing of knowledge. Yet, there is still a rather strong competition mindset. But when we anchor collaboration in a project structure, it works.” Technology-responsive (80%) and system-minded (externally 75%) were both rated high. The strong impact of technologies for the organizational purpose of climate protection and the system character of the topic were mentioned as the reasons. The highest contradiction in the generic cultural factors was seen in the gap between the external and internal (20%) system-minded rate. The director explained the connection with a competitive mindset between the departments and the complexity of the interrelations between the different energy impact factors. Company-Specific Cultural Core Factors The three specific cultural factors added were best quality (80%), performance (70%), and team spirit (80%). “We may take longer than others when we craft our recommendations, but we really have the best experts in our organization, and we check everything before we present our solutions.” The business model of the organization is based on regulated catalogs of requirements and predefined result deliveries connected to respective time investments. As the director points out: “We do perform; we deliver and start again to perform and deliver best quality. We do not take the time to breathe or to recapitulate what went well or wrong.” The positive team spirit, however, was seen as emotional buffer in times of extreme pressure. Cultural Worldview and Resulting Strategic Need The egalitarian perceptual lens on the world came across in statements like “team spirit,” “talking on eye level,” and informal communication habits (using the informal German address “Du” across hierarchical levels). In the workshops, especially the teambuilding interventions were quickly adopted and transferred to daily business processes. Mixed teams with participants of different hierarchal levels did not block team dynamics related to the different positions. Important parts of the interview with the director were dominated by topics such as boundaries through strict regulations as well as being unable to define clear deliveries in constantly evolving projects. This corresponds to a hierarchical worldview in the top right-hand quadrant on the cultural map (Fig. 7). We assumed the strategic need for controllable collaborative work. This guided the

302

A. Kerguenne et al.

selection of action sets for the research workshops in different transformation phases. Transformation Purpose/Vision “We believe that climate protection needs easyto-apply processes and a good functional team play among all stakeholders.” This purpose defined by the leadership team relates to their commitment to added value and to the urgency of changing their current way of teamwork.

6.1.1

Transformation Process Phases

Kickoff and Envision Teambuilding Workshop, Definition of the Team’s Transformation Mission, Vision, and “Hot Topics” of Transformation. As a first step, the director initiated a design thinking workshop to set up the first collaboration activity with her leadership team. The aim of the workshop was to define the team profile, purpose, and vision. The workshop was also used as a space to openly discuss the internal problems in project work and to develop first solution concepts. The questions of “how to prioritize projects under time pressure” and “how to enable participative decisions in diverse teams” were chosen as key challenges in the context of an optimization of processes and teamplay. The created ideas served as steppingstones for the different transformational action described further below. Plan Upskilling Training Leadership Team and Team Members in Strategic Design Thinking and Agile Project Management Methods (OKR). To establish a solid knowledge basis and confidence in application, the director and her leadership team decided to invest in two main method trainings: strategic design thinking, to enable the crafting of human-centered transformation strategies and operational implications. In addition, the Objectives and Key Result (OKR) project management methodology supported self-direction and collaboration in teamplay. The team was able to plan and track the success of their key transformation experiments with these two methods. Experiment Selection of strategic key projects and development as exploration labs to learn for company-wide scaling. To create their first blueprints for new structures and processes, the team decided to focus on two leverage points. One project would establish an institutionalized network of key users for continuous feedback on the organization’s services, and the second project aimed at building a so-called project navigator. This is a (physical and virtual) container of all constitutional elements and principles that have to be built up for supporting climate protection with “easy-to-apply processes and a good functional team play.” Both projects were planned to include several learning loops and iterations to deliver solutions that would pay into the performance and best quality culture. Implement Visualization of factors and roadmap to navigate huge, complex projects. After some prototyping cycles, the team decided to visualize their new structure and processes to document their learnings as a temporary draft. It was a tool to be used in all strategic meetings and presentations as starter and reference.

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

303

The visual roadmap was used as a living artifact, which would be updated every few weeks before the stakeholders’ meetings.

6.1.2

Results Human Level

Establishment of Monthly Milestone Team Workshops as a Combination of Design Thinking Method Training and Project Work “We do not have the time for trainings only.” The high-performance focus in the organizational culture has a blocking effect on “pure trainings” without connection to day-to-day business. Accordingly, the strategic decision was taken to combine key milestones in project work with the application training of fundamental design thinking concepts such as empathy, diversity, or iteration. The leaders and team members trained in the design thinking courses were slowly moving into becoming facilitators of the milestone works sessions. The result was that the teams moved on in their project work and created results; they did not have the feeling of losing work time with “fancy design thinking games” and could grow their knowledge on a cognitive- and skills-based level. Integrating High Frequency Feedback Mechanisms in Daily Work for Continuous Learning and Optimizing The first direct change in the work routine was the establishment of a compact feedback mechanism in the daily standup meetings. This was inspired by the classical design thinking feedback grid: “What works well in my project?” “Where do I need help?” and “What is a good new solution for a challenge?”. This format allowed a time-efficient added value for growing collaboration and knowledge. Results Technology Level The leadership team had struggled with the implementation of digital collaborative project work management systems. This was expressed in the comment: “The time we need to transfer all our Excel content into a digital system that we have to learn to handle simply does not exist.” When the team realized that PM systems could be matched with the selected agile project management method OKR, the prioritization changed. The team leaders quickly got information on project achievements without losing time in meetings. Additionally, the “one-click” horizontal alignment between the teams within project flows was an attractive aspect that balanced the time investment in learning how to use the digital tools.

6.1.3

Results Organizational System Level

Implementing Agile Project Management Methods (OKR) The implementation of an agile project management method called “Objectives and Key Results” (OKR) helped teams to work in a self-organized way by connecting on horizontal and vertical levels. Teams and their leaders follow a structured workflow that starts with the overall business objectives, which is translated into concrete results and

304

A. Kerguenne et al.

work packages. Synchronization of all actors takes place in defined frequencies. The teams and their team leads quickly appreciated the benefit of knowing how the specific work packages were connected to the bigger picture and how these would impact the other team’s work. Although the implementation of the OKR method was time-consuming, everybody actively contributed to it. Implementation of Regular Knowledge Sharing and Feedback Format The second main change was the integration of different exchange formats between the key stakeholders to ensure a complete and constant understanding of respective need structures. The aim was to create a monthly sharing format in a small team representing main stakeholders to openly share knowledge and discuss issues beyond daily business. We/the team built the user panel mentioned above to get closer to the core user-multiplicator group. Integrating “Synapses” into the Organizational Structure The third important change was implemented in the organizational structure of the consultancy department. They established a coordination team in addition to the different expert teams for specific topics and classical service teams, such as legal consultancy, communication, and digital tools and applications. Their job was to create “the synapses of the organizational brain” by making sure that temporary teams from suitable areas could easily form on demand, efficiently perform, and finally evaluate their results of a specific subproject. This transformation represented the urgently needed structural element that would make collaboration a “built-in” part of every work package. Hurdles on the Digital Transformation Way The main hurdle expressed by the director was the conflict of rapidly growing project scopes and team sizes on one hand and the time needed for redesign of team governance concept and respective new ways of working on the other hand. The overall time pressure would not allow additional time investment for collaboration and learning phases; therefore, “transforming while performing” was a challenge. The team also had to deal with the orientation gap in the period of restructuring and implementation of new project management methods. Design Thinking Leverage Impact Especially teambuilding- and empathy-related design thinking actions helped to optimize the time vs. result ratio. These included team check-ins and checkouts and regular stakeholder feedback sessions. Knowing better how to meet specific objectives would lead to better results and less rework. In addition, stepping into the shoes of her team, the director integrated the active communication of transformation benefits to her team in daily work. Among these were good feedback from top management or clients and a time-saving way to prioritize knowledge topics. Using the analogy of a map helped the team to make the complex information and stakeholder system they operate tangible and manageable. The team defined interrelations as “highways” for frequent alignment and “streets and country lanes” for flows with smaller amount of information and less “travel frequency.”

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

305

Limitations of Design Thinking and Completing Thought Schools or Methods Used For a structured operationalizing and measurement of vertical and horizontal alignment to enable time-efficient teamwork, design thinking did not offer suitable strategies or methods. Here, the business management method based on the Management-by-Objectives approach (Drucker 1954)—the OKR method— was implemented to steer collaboration work and measure teamwork results. As it fits to the design thinking principle of systems thinking, the OKR method connected easily to other actions results, for instance, to the vision the team had defined.

6.2

Case Study 2: From IT as a Cost Factor to an Integrated Partnership for Organizational Future-Fitness European for-Profit Organization Safety and Hygiene (EF/SandH >11.000)

Organization and Background Motivation Becoming the “recognized leader in sustainable, innovative, and digital solutions for all customers” in their market segment is the defined overall strategic goal of EF/SandH, a family-equity company rooted in Germany. The company has 11,000 employees and operates across Europe. EF/SandH was founded as a sales and service company in the 1950s and later merged with a textile service company in the 2000s, to be finally fully integrated into a European family-equity portfolio in 2019. The top management actor (I2), who started the digital transformation activities, attended a design thinking leadership program at the professional development entity of HPI Academy in 2019. After changing to her current position as CIO (Chief Information Officer), the decision-maker started a comprehensive restructuring of the IT department, including functional, personnel, structural, and processual changes. Design thinking was chosen as a strategy in combination with classical change management methods to drive the process. The entire organization had just been reorganized from a centralized to a divisional structure to better meet customer needs in areas that operate in different markets. Transformation Start Area in Company System The transformation program started in the IT department, which was responsible for the implementation of digital transformation activities. This department cooperated closely with the digitalization department, responsible for the development of digital transformation activities. Initial Goal for Digital Transformation The main reason for the transformation efforts by the CIO as well as the leadership team was “for business reasons. . .to survive and continue to grow our business in times of increasing competition and a quickly changing environment.” The main activities of the transformation program had been defined on different levels. These included harmonization of current IT systems, development, and implementation of a comprehensive IT security system as well as integration of data- and AI-based customer relationship systems. This

306

A. Kerguenne et al.

transformation also required the establishment of new agile governance structures to augment customer focus, speed, and performance. Challenge According to the CIO, the problem was that “[they] got stuck.” Classical organizational governance frameworks like RACI [author’s note: RACI is an abbreviation for defining who is responsible and accountable, must be consulted, or must be informed in different project phases] would not be sufficient to take decisions to create a suitable structure. Nor could they facilitate requisite processes and job profiles to bring about customer centricity and flexibility while meeting the given performance and resource objectives. “There are many different organizational setups that we consider, but we can’t decide which one is the best for us. . .also, we can’t just stop and build from scratch, but we have to keep on running the business while restructuring and implementing new processes.” Generic Cultural Factors The self-assessment of the five cultural factor values by the leadership team showed low values in almost all generic cultural factors. The factor HUMAN CENTERED (35%) was interpreted as internal factor relating to the employees; whereas CUSTOMER CENTERED (25%) was related to the (internal and external) customers using products and services. The factor ITERATIONDRIVEN was evaluated with 25%. SYSTEM-MINDED was again subdivided in internal company SYSTEM-MINDED (30%) and external world or market SYSTEM- MINDED (60%). The latter, higher value, was discussed in the team as “natural” and linked to the organization’s purpose and market operation field safety and hygiene. Collaboration was rated with a value of 50% that was explained by “great conversations across hierarchies and disciplines happening,” yet “the situation of siloed working of the past is still present” despite digital collaboration tools that although supporting functional—“forced”—collaboration, would not translate into content-related teamwork. The low scores for iteration-driven and human/customercentered factors led to intensive conversations about the missing failure and learning culture as well as the perceived urgency of changing the perspective from “IT as a tool to IT as a partner.” Technology-responsive was rated with 30%. The common evaluation agreed on the knowledge and focus of the functional benefit of new technologies, without being aware of implications from a human-centered or cultural perspective. Company-Specific Cultural Core Factors The two specific cultural factors that the team added were stability/safety and cost efficiency. Stability (20%), a factor rooted in the company’s heritage in the safety and hygiene industry, received a particularly low rating. This was explained with the ongoing change initiated by the transformation program (“we are still changing, even though the direction is clear”). Cost-efficiency (100%) was rated the highest in the entire factor combination. This also led to an intensive team discussion about manifestations such as a strong focus on key performance indicators (KPIs), the priority on cost reduction, delivery on financial results, a general perspective through the “cost lens,” and measurability of activity success. The team also reflected on the relation between the overall sustainability objectives of the company and the financial stability as one aspect to achieve.

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

307

This is the so-called Enkelfähigkeit—a concept the family-equity company commits to as described by the economic philosopher Anders Indset. “Being ‘enkelfähig’ means living in a value-oriented way and reconciling this with entrepreneurial thinking. Enkelfähigkeit is economic sustainability over time—not in time—where ecology and economy are not contradictory.” The high rating of the factor systemminded (60%) for the external system supported this deeper reflection. Interpreted Cultural Worldview and Assumed Strategic Need The conversations and verbal notes on Post-it’s around the cultural factor prompts showed a concentration of expressions like “hierarchy,” “rules,” “stability vs. instability (due to transformation),” “top-down approach,” “too much politics,” “cost-driven,” and “silo structure,” which can be interpreted as an emphasis on strong regulation, given rules and structures. This corresponds to a hierarchical worldview in terms of cultural theory (Thompson et al., 1990). We found this perspective frequently in medium-sized companies with classical family-foundation roots. The team identified the strategic need for participative transformation actions, user-need understanding, and teambuilding routines. Transformation Purpose/Vision The IT team analyzed their skills, passions, and the needs of the organization regarding the transformation process. This resulted in their transformation purpose as “carrying the company safely to a data-driven future.” This statement describes the team’s commitment to its r service and support role as well as its connection to the overall company purpose with its safety core. Transformation Process Phases The CIO initiated a multistep process with her leadership team and decided to base the milestones on collaborative, partly crossfunctional workshops. To “unleash and connect the intelligence in the team,” she opted for working sessions facilitated with design thinking methods. Kickoff Setting up collaboration, teambuilding workshop, definition of the IT team’s mission, and “hot topics” of transformation. The 11 IT leadership team members met to form a transformation team. Their aim was to analyze their skills, discuss their personal passion regarding the transformation, and define their purpose as well as the team values to which they would commit. They discovered nine “hot topics” that were rated as critical for the success of their transformation program, the top 3 being “governance,” “overcoming silos,” and “a common understanding of the IT role for the company-wide transformational success.” The topic of governance was selected as a priority area. Status Quo Expanding collaboration, cultural self-assessment workshop base. In the second step, a collaborative workshop for cultural self-assessment based on the maturity model took place including colleagues from different business areas. In three mixed teams, the workshop participants evaluated the cultural profile of their company based on concrete manifestations like projects, processes, remuneration systems, or crisis behaviors. The three teams consolidated their evaluations into one organizational cultural profile, which would serve as reference for planning the transformation activities.

308

A. Kerguenne et al.

Envision Increasing customer centricity, vision workshop to design the ideal future IT structure and processes. In the third step, the team invited their direct internal customers from the different business areas to join in a collaborative concept workshop to prototype the ideal IT structure and processes. In teamwork co-creation sessions and one-to-one interviews, functional and emotional customer needs were collected and synthesized to serve as inspiration for three different approaches to the organizational setup of the future: 1. Creating “built-in” connection points between IT and business departments. 2. Organizing an efficient change with service focus, defining products and projects consequently from a user’s point of view (responsibility for tech-onboarding of new employees) instead of from a technology’s perspective (responsibility for the printers). Plan Prototyping for testing and iteration, exploration research workshop to define prototypes stages and testing periods. The fourth step consisted of deciding on a concrete organizational setup, prototyping archetypical use case scenarios for the future IT services, and crafting the exploration strategy to plan testing, iteration, and implementation. The core IT team met to share and synthesize their new insights, build the new IT structure and governance concept, as well as define the respective implications for exemplarily job families and job profiles. Based on the cultural profile, concrete, cultural value-related KPIs were defined for all planed experiments, for example, “raise the perceived collaboration between IT and business department to at least 50%.” An exploration strategy containing plans, time periods, and concrete actions for three levels was defined for three activity levels. These levels were team and collaboration activities, prototype, testing and measurement activities, as well as communication activities. Experiment The unplanned crisis as living lab, rapid experimentation and learning during a cyberattack. During the transformation process, just before running the defined experiments, the company suffered a cyberattack. Within an extremely short time, systems had to be stopped, secured, and rebuilt. “Decisions and changes that would have taken months to implement were now put to work within 3 weeks.” The team was forced to work around the standard processes and structures to secure the quick reestablishment of basic value creation processes. From a transformational point of view, this crisis can be considered an “extreme experimentation space.” Decisions and solutions would be put into work immediately and iterated on the run. As the team could not follow—and be hindered by—usual procedures, the agile way of experimenting forward was a welcome strategy that turned out to be efficient. Implement Participation and team conflicts; social dynamics management becomes priority in participative design of new IT government concept. After having coped successfully with the cyber-crisis, the team joined forces in one of the research workshops to continue work on the governance model and to consider the learnings from their “extreme experimentation.” When the concept proposed by the leader was physically put on the table, a team conflict became tangible that had been hidden by the cyber-crisis. These upcoming team dynamics were blocking the teamwork, so

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

309

that they had to be managed immediately during the workshop and transformed into new team rules before the governance concept could be finalized and communicated to middle management. Results Human Level The CIO of the company argued as follows: “Two things struck me about design thinking. First, it has the power to unleash team intelligence and to connect it to something new and bigger. Second, it is a systems thinking perspective on complex challenges and focuses on visual communication in a way that helps to tell your transformation story.” She decided to build her strategy presentations around a simple visual that communicates the interrelation of three key factors of the intended restructuring: the people, processes, and technology factor. This helped her to explain the complexity of the task and to create clarity about the area of decision-making. It also made it easy for the leadership team to tell one consistent story to their teams and to craft clarity about the interrelations of different transformation areas. Internal and External Storytelling about Strategic Goal, Collaborative Process, and Status Quo of Transformation The CIO also increased transparency and frequency of information, going beyond the internal presentation and update activities. She started an internal and external storytelling series, openly talking about strategic goals, collaborative processes, achievements, and the status quo of transformation activities. Results Technology Level The transformation of the IT services changed fundamentally. Putting the focus on user and user products (services) resulted in a redesign of entire IT processes, the quicker implementation of new IT systems and platforms, as well as the establishment of new technological skill profiles.

6.2.1

Results Organizational System Level

Participative Development and Testing of Hybrid Linear-Agile Governance Concept Prototypes Before the crisis, the leadership team had phrased the main question as follows: “The biggest challenges lie in front of us—how will we be able to explore and learn by testing our prototypes for agile governance in a culture where cost efficiency and stability are the most important values?” Taking the identified cultural profile as a reference, the team created a precisely crafted test design with different stages and clear test hypothesis. Emphasis was on measurable indicators for successful solutions as well as the interrelation of those cultural factors identified as “uncontestable.” These were the factors customer centricity and cost efficiency. The main key narrative in organizational storytelling by the leadership team was now “cost-efficient transformation = failing cheap to succeed sooner.” What surprised the leadership team and their superior was how appreciative middle management would react to being asked for creative participation in the execution and rollout of the new concept. The main change in the governance concept was the integration of a new intersection role, “digital Dolmetschers,” speaking both the languages of IT

310

A. Kerguenne et al.

and business, meaning the establishment of an additional function to enable efficient communication between IT and their internal customers from the business divisions. On a process level, all IT services and each archetypical IT product were precisely structured regarding the linear and agile phases. The result was a clear governance structure and processes that enable both running the IT services efficiently and implementing new IT services while learning quickly. Hurdles on the Transformation Way At the beginning, the most important hurdles during the research workshops were stated as the complexity of IT restructuring while running daily business. Later, this was the communication hurdles between business and IT department team members. The cyberattack crisis represented a blocker that was transformed into an accelerator. The team dynamics were a hurdle that could have made “the whole thing fail,” according to the CIO. Design Thinking Leverage Impact To handle the complexity, the team used principles of systems thinking and prototyping. The key visualization of interconnected people, technology, and process systems helped to deconstruct and cluster the elements and topics that had to be worked on. An important milestone for the restructuring meant exploring the needs of users in interviews and integrating a respective role in their structure. Analogizing energized the process in the different phases of getting started and defining a tangible vision (“we are the carrier”) addressing critical topics (“hot potatoes”) and unexpressed conflicts (“elephants in the room”). The most important impact, though, was created by the application of principles such as Empathy and Teambuilding. Using the team dynamics cycle (Tuckman, 1965) and related methods of nonviolent communication, the team managed to solve conflicts and was equipped with tools to handle those in the future. Limitations of Design Thinking and Completing Thought Schools or Methods Used The structural governance concept redesign was done with an adaptation of classical frameworks that were combined with design thinking-based user research and prototyping approaches.

6.3

Case 3: Toward a Structured Yet Flexible Process for a Corporate Incubator—National for-Profit Organization Insurance Services (NF/IS >14.000)

Organization and Background Motivation NF/S is a national insurance company that emerged from a fusion to a cooperative organization in the 1930s. Later, financial services joined, and the economic growth was largely driven by the expanding mobility market. The innovation department had started with single innovation projects to explore new business opportunities in their automotive insurance market 7 years ago. Feeling the increasing pressure and decreasing legitimation of “selling insurance face to face, and just relying on [their] good brand image,” actors in different areas of the company were given budgets to use digital technology

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

311

to create innovations for the service portfolio. One of the projects turned out to be particularly successful: an app for truck drivers that made the search for parking lots easier. The upper middle management actor (I5) who started the digital transformation activities attended a design thinking leadership program at the HPI Academy in 2017. Transformation Entry Area in Company System The project team established themselves as drivers of organizational innovation and became the first members of an official corporate innovation lab with transformational goals. With a team of 10, they handle approximately five to seven projects in parallel, which are chosen to transform the company’s services and product portfolio. Initial Goal for Digital Transformation The main reason for the transformation efforts is quoted by the leader of the lab as follows: “The insurance business as we know it today will not exist anymore tomorrow. This is due to different social and technological developments (. . .) we now must discuss topics like security aspects when nobody is actively driving a car anymore (. . .) the classical paper insurance card with classical insurance parameters is about to die—which is also the case for our specific distribution channels: physical banks. The “every-day-problem-solving” for commercial clients that had been defined as the core of the internal narrative for future success positions the organization’s interest of transformation within the intersection of human value and organizational system value creation.” Challenge Initially, the leader defined the challenges of the lab in the internal teambuilding area and in the project portfolio and in relationship to partners and customers areas. “How do I make sure that I do not demotivate my team in daily business?” “How can we scale our successful pilot project?” or “How can we help our clients come up with future-oriented business models?” Synthesizing the various questions and needs, the team realized that after 2 years of official “lab operations,” the entity would need a tailored, user-centric, effective, and repeatable process with respective basic tools for the development of transformative innovations. In fact, the overarching question was “How can we explore to learn what works for the establishment of a corporate incubator and lab to create future value for the company? Generic Cultural Factors In the five interviews that we conducted with leaders inside and outside the lab, the self-assessment of the five cultural factor values showed the lowest values in customer centered with 25% (externally) and collaborative with 40% (externally). This was explained by the perceived relationship gap between the organization and their partners and customers. “Feedback and the exploration of needs are not built in our value creation processes.” One of the written rules is, “if necessary,” clients can be consulted. “When we want to know something, we usually ask the sales rep—assuming that he would know what the client needs, but we don’t have any centralized space to collect user data. In the lab, we work in close exchange with a user panel that we built up in our first innovation project.” The form of the organization is based on a cooperative model, and the internal human centered factor rated accordingly high with 80% and collaborative (internally) with

312

A. Kerguenne et al.

60%. The reason given for a medium iteration-driven rating of 50% was surprising: “We have a positive culture regarding failure—nobody is blamed for making mistakes—but we don’t have mechanisms to learn from failures, and more importantly, the problem of failure lies in our product management processes. Very often, the projects that we work on are too big to fail. We move on, because a high budget has been invested, even if we know with common sense that it probably won’t work out.” The high rating of technology responsive (80%) is connected to the organization’s core value “security.” All capabilities of digital technology that would help increase safety and security are continuously explored and applied. The nature of insurance business processes as well as the cooperative values led to a high rating of the factor system minded with 80%. “When we introduce something new, we consider the consequences: What will change for the client? How does it affect the employees in their daily work? Do they have the necessary skills for it to succeed? We have a strong worker’s council; they care a lot about the consequences of change.” Company-Specific Cultural Core Factors Three specific cultural factors are related to the organization’s core value creation concept as an insurance company and cooperative form: security, rated with 90%; continuity/reliability, with 80%; and a strong community spirit, which explains the tolerant handling of failure mentioned above, with 90%. Interestingly, community spirit did not find its way into the lab’s purpose, although representing a relevant user benefit in the market category and, more importantly, a core business model factor in some of the innovation projects. “Community spirit is deeply anchored in our culture. It translates in our leadership guidelines, hierarchy structure, and guidelines for individual development paths, for instance, with a long-term view on life success instead of a short-term view of individual success.” The downside appears in decision processes or in the fact that decisions take a long time to finally be taken. “Our consensus culture simply blocks the speed. Also, uniqueness or courage is not fed by consensus.” When “the hierarchy joins the game, these qualities slow down,” one leader pointed out. Cultural Worldview and Resulting Strategic Need The focus on an egalitarian worldview with a distinctly hierarchical orientation would help the organization to kickoff human-centered collaboration processes yet block the design and establishment of new formal structures and processes for innovation and transformation processes. Mindset-wise, there is a fertile ground that cannot be easily transformed in structured and repeatable actions. Therefore, after synthesizing the interviews, we defined the strategic need for repeatable work processes that enable both the creative self-direction of lab teams and the measurement of experimentation result achievements. Transformation Purpose/Vision The lab team had developed their vision based on the “golden circle” (Sinek, 2009), a systems thinking-based framework putting the “why” (purpose) of an organization in relation to the “how” (value guidelines) and the “what” (concrete activities). The aim was defined as “challenging the status

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

313

quo (of classical automotive insurance category) for a better tomorrow and evolving from an insurance company to a solver of our clients’ daily problems.”

6.3.1

Transformation Process Phases

Envision Definition of the lab’s mission, values, and vision. With the formal establishment of the lab, the team leader had started in a participative way to define their reason for being. “Challenging the status quo for a better tomorrow” was completed by values such as love of problem, solving for change, customer centricity, diversity, and open mindness. The scope of the lab was defined in “building great mobility solutions that no one would expect from an insurance company.” “These are our guidelines and decision criteria for everything we do,” the team leader stated. Experiment Running several innovation projects. After 2 years of lab work and approximately five to seven work-in-progress projects, the team had accumulated a broad and deep knowledge about what would work or not work. However, in the phase of growing with new team members, it was now important to synthesize the learnings and to integrate the knowledge into an effective lab process. Especially aspects like stakeholder buy-in and marketing of new solutions represented an unsolved area with many question marks. “The service is well accepted by our pilot group, but scaling is a problem, and internally, the decision takers start to hesitate in moving on with the investment,” the team told us. Implement Co-creation workshop to learn from experiments, spot process gaps, and plan optimized lab process for ongoing and future projects. The team used the case research workshop to connect their knowledge and to do a structured analysis of their projects focusing on the most recently completed one that was about to be scaled. Using the design thinking feedback grid structure for analyzing strengths, weaknesses, open questions, and existing improvement ideas, they first unpacked the knowledge data. In a second step, the team matched the data around our proposed digital transformation process framework phases. It was now easy to have focused discussions about the drivers and blockers in the different process phases, because their own case served as a concrete example on one hand and could be deconstructed for easier reflection. The reflection results were later aligned with the process to create a complete picture with clearly identifiable gaps. The team decided to use the proposed framework as a basis for the redesign of their lab processes. The most interesting discussion during reflection focused on the early involvement of stakeholders in the process to avoid resistance in later implementation and scaling phases. The team decided to integrate a step where they would discuss the starting point of projects regarding the exploration level, that is, exploring user needs at the start vs. exploring user needs connected to an existing solution idea as a starting point. Results Human Level For the first time, the team reflected on the strengths and weaknesses of the corporate lab process and decided to transform it into their tailored

314

A. Kerguenne et al.

process. A team participating at one of the next strategic design thinking programs at the HPI Academy will work on this project. With the collaborative mapping of their group knowledge regarding the existing project on the proposed digital transformation process framework, the team created a common reference of future redesign needs and rules. Results Technology Level The corporate lab’s scope strongly focuses on digital services for business model transformation. The team decided to integrate a stronger user focus into the future process. Results Organizational System Level Work-in-progress is the further development of the proposed adaptive digital transformation process with action sets for standardizing the corporate lab workflows. Hurdles on the Digital Transformation Way Even though team members were experienced project managers, many with design thinking skills, it was not easy for the team to effectively define the governance concept with roles, skills, and processes for their workflows that would help them in: 1. Enabling lab project strategies that would consider the dynamics of their company culture. 2. Integrating stakeholders for buy-in in early stages. 3. Developing project rationales/narratives for broad internal communication in early stages. 4. Setting metrics for decisions about moving on or stopping experiments. 5. Collecting, documenting, and synthesizing their learnings from experiments. Design Thinking Leverage Impact The “lessons learned” about empathy, iteration, and systems thinking helped to transfer design thinking principles into other domains of transformational actions such as leadership and emerging transformation guidelines (e.g., “we accept a 70% solution”). Connection points of design thinking with the organizational culture were done on principle level. “Systems thinking is part of our culture; it fits to our community spirit.” The design thinking rhythm of diverging and converging, with which we underpinned the digital transformation process, helped the team in defining the concrete phases of independent work as a lab (diverging) and collaborative decision taking as a part of the organization (converging). Limitations of Design Thinking and Completing Thought Schools or Methods Used “I thought that there are things missing in the design thinking process. When I built up the lab, I could not just take the process and map it.” This is the evaluation of the lab team leader. In the workshop, team members addressed following gaps: Classical design thinking process does not map: • Both the innovation and the integration process of transformative solutions. • The necessity of stakeholder involvement and storytelling from the beginning on. • Design thinking does not help in legal questions of venture development.

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

315

Business frameworks like SWOT analysis, Cynefin framework, and disciplines like foresight and strategic project management are not anchored in the design thinking toolbox or process.

6.4

Lessons Learned from the Cases: Opportunities and Limitations of Design Thinking for Navigating Digital Transformation

Learning 5: The cultural digital transformation maturity can serve as strategic decision reference. It helps to define the starting point and metrics to track achievements and it creates individual identification.

We can summarize the overall learnings from the cases as follows: • In all three cases, the active evaluation of cultural beliefs and values and the reflection about the interrelation of generic cultural factor values and specific cultural factor values created a tangible reference for transformation decisions and actions taken by both the actors and the facilitators of the research workshops. • Active reflection about the value of the generic cultural maturity factors in relation to the specific cultural core factors helped to create a starting point and measurement references for transformation strategies and results. On a leadership level, this translated into decision confidence about choosing education programs that would grow factors with perceived weakness as well as about defining respective transformation actions. • In all three cases, the leader or the leader and her team perceived the organizational culture profile as an important help to “zoom out and synthesize everything after having zoomed deep into the projects.” Learning 6: Transformation processes are vulnerable and fragile. Teams get caught in the “paralysis trap” in times of priority conflicts with daily business or unplanned events and need “kick-starters” to continue. (NPO/EN)

Starting in an organizational area with increasing importance and size, the main strategic need within the transformation process was a new setup of workflows, work division, and work collaboration. Active participation in the transformation process was intended, supported by the leader, and took place, yet the intensity and frequency still varied between the team members, due to daily business time resources as well as work priorities. Whenever unplanned additional work packages would call for action, the priorities of transformation activities were pushed to a lower importance level. Then when transformational actions were continued, the team had to invest time in remembering the status quo where they had stopped. (Note: This motivated us to come up with canvases that condense the results of the different action areas and phases to have a precise starting point to continue). Learning 7: Digital transformation requires constant integration of human, technology, and system perspective to be successful. (EF/SandH)

316

A. Kerguenne et al.

The transformation process was initiated in the IT department, an organizational area with interconnection to all other areas. The transformation strategy focused on creating small prototypes that were tested with users before being implemented in daily work. Influenced by their hierarchy-driven culture, the sudden “responsibility of creating a structure that affects the entire organization” was a strong driver for the team for considering all main company perspectives. One of the most important insights that was gained in the research workshops was the communication gap between IT and business teams, due to their fundamentally different perspectives. This led to the decision of integrating the “digital Dolmetschers,” specific roles in the intersection areas of IT-business project work into the organizational governance model. Learning 8: Integrated transformation strategies with clearly structured process that guides analysis, vision, learn, and diffusion actions (NF/IS) are more likely to succeed because they • Give everybody in the team orientation and structure. • Give the team and lead confidence for early collaboration and communication with stakeholders in the organization. • Provide the team lead with a reference of how to measure team performances in selfdirected, explorative work processes. • Provide the team lead with a basis for defining specific skills, profiles, roles, and respective development plans for team members.

Starting in an organizational area dedicated to digital transformation and the creation of new business models, the lab team’s strategy focuses on developing and implementing new services to drive the organization’s evolution from an insurance provider to solver of daily problems connected to mobility in the wider context of security. The team felt like pioneers in defining the future agenda of the organization. Coming from an interpreted egalitarian worldview with a strong hierarchical focus, our assumption was confirmed: Our results show a culture-related strategic need for a formal, repeatable structure that would organize the explorative way of working of the lab and create connections between the lab projects and the organizational system along the way.

6.5

Case Conclusion

The preceding sections have argued—both at the conceptual and empirical levels— the case for extending and integrating organizational culture into maturity models of digital transformation. For one, as the cases have demonstrated, culturally sensitive maturity models enable analysts as well as would-be organizational transformers a more nuanced and profound understanding of where the organization stands, what potential trajectories are appropriate and feasible, as well as what existing transformation potential is ready for mobilization. In this way, digital transformation efforts are more likely to resonate with extant organizational structures, norms, and practices, which significantly increase the chance of success. In sum, the evidence from

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

317

these cases seems to indicate that a culturally sensitive maturity model may exhibit the following features and functions: 1. Externalization of perceived cultural beliefs and values as a reference for starting a transformation process. 2. Growing of individual and collaborative understanding about opportunities and hurdles in a transformation process. 3. Orientation and confidence for decision-making when crafting a transformation strategy. 4. Decision support for design of success measuring metrics of the transformation activities. 5. Reference for integration, anchoring, and scaling of transformation activities in the organizational system. 6. Participation catalyst—in raising the entrepreneurial perspective. 7. Synthesis tool that helps zooming out within the transformation process after zooming in phases.

7 Summary and Outlook: Design Thinking for the Digital Transformation’s Challenges of Tomorrow After a year of practice-oriented action research, the opportunities and limitations of design thinking as a strategic approach for navigating digital transformation can be described in reference to the three levels of design thinking. The approach is a mindset, a method, and a set of concepts or principles in the sense of lessons learned from reflected experiences. Navigating digital transformation throws the leading actors into a field characterized by dynamic complexity, which leads to a high level of ambiguity, volatility, and uncertainty. They engage in a long-term journey where they must manage the human, technology, and organizational systems that are naturally interrelated and constantly influence each other. Therefore, on the way, the unexpected becomes the standard, and the ability to creatively deal with unplanned events is key. On the other hand, transformation extends over a longer period and needs to follow parallel paths: the exploration paths for the new and the integration path for connecting the new with the existing system. The best prevention against “disruption paralysis” is to keep the cycle of exploration and integration going. On a mindset level, design thinking can be described as a motivation motor for leading actors who want to actively drive transformation processes. With design thinking, it is easier to dare to do things differently in huge organizational systems with high levels of rigidity. This means building up new entities and starting transformation processes that will impact the entire organization with a strong inner belief that embracing collaboration, human centricity, or learning by experiments are the best ways to move forward in that complex journey. Any limitations

318

A. Kerguenne et al.

within the concrete process were seen in the lack of operationalization tools for distinct transformation tasks, especially on the integration paths. On a method level, design thinking is an impactful toolbox, applied to get the right things done in the right way. These are tasks in the transformation process areas that are elements of the classical design thinking approach: for instance, teambuilding, stakeholder needs exploration, and learning processes. The limitations mirror the ones mentioned on the mindset level. This means the design thinking toolbox is less suitable in guiding and structuring action, specifically for essential tasks in the transformational action areas of analysis, vision, and diffusion. On a concept or strategic principle level, we think that design thinking offers an untapped opportunity field with high potential. Design thinking crossed the innovation boarder to become a helper in transformation, when practitioners could transform their mindset and method experiences to lessons learned for the transfer into the diversity of their distinct application contexts. On a concept level, design thinking can be combined with concepts of other disciplines like foresight, organizational development, or governance modeling. This led us to our conclusion: We see the future contribution of design thinking for digital transformation processes in its deconstruction. Design thinking concepts along with the scientific explanation of their transformational impact could become the ingredient for an adaptive strategy kit that every actor can use. Through the use of this kit, the actor can develop it in self-direction. Design thinking should continue to be open and develop itself as a “cocktail of intelligences” coming from various disciplines to evolve to transformation thinking. From a practitioner’s view, it should become easy to mix and match design thinking concepts with existing concepts of business thinking and foresight to infuse human centricity, quick learning, and collaborative diversity into tailored transformational strategies and actions. Thus, our motivation is to expand our adaptive digital transformation strategy kit, more specifically with further research-practice bridges that build the basis to design specific action sets. For the future development of design thinking as a strategy for digital transformation, our research has led to more open questions that could translate into possible follow-up projects. The importance of culture and governance concepts for successful transformation is the main field that we want to investigate further. How can we understand specific cultures with the cultural theory to craft transformation strategies that avoid the waste of resources and value on the way? And how can we craft governance concepts that enable both a waste-avoiding value creation and continuous learning for adaptivity to the future challenges? These are the topics that will contribute to the evolvement of design thinking, creating a link to digital technology as an integrated, impactful approach to transform the reality of today into desirable futures. Acknowledgments First and foremost, we would like to thank Prof. Dr. Christoph Meinel for supervision and the Hasso Plattner Foundation for funding this research project within the Hasso Plattner Design Thinking Research Program. Thanks go to our colleagues in the research community who provided input and feedback and to Sharon Nemeth for editing.

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

319

References Balters, S., Mayseless, N., Hawthorne, G., & Reiss, A. L. (2021). The neuroscience of team cooperation versus team collaboration (pp. 203–217). Springer. https://doi.org/10.1007/9783-030-62037-0_9 Beyhl, T., Berg, G., & Giese, H. (2014). Connecting designing and engineering activities. In Design thinking research (pp. 153–182). Springer International Publishing. https://doi.org/10.1007/ 978-3-319-01303-9_11. Björk, J., Boccardelli, P., & Magnusson, M. (2010). Ideation capabilities for continuous innovation. Creativity and Innovation Management, 19(4), 385–396. https://doi.org/10.1111/j.1467-8691. 2010.00581.x Bohn, U., Forbach, M., Käppler, C., & Wähler, L. (2018). Culture first! How your business benefits from a digital culture change. Capgemini Invent. Brown, T., & Katz, B. (2011). Change by design. Journal of Product Innovation Management, 28(3), 381–383. https://doi.org/10.1111/J.1540-5885.2011.00806.X Buvat, J., Crummenerl, C., Kar, K., Sengputa, A., Solis, B., Aboud, C., & el Aoufi, H. (2017). The digital culture challenge: Closing the employee-leadership gap by the digital transformation institute. Capgemini. Davenport, T. H., & Westerman, G. (2018). Why so many high-profile digital transformations fail. Harvard Business Review. https://hbr.org/2018/03/why-so-many-high-profiledigitaltransformations-fail de la Boutetière, H., Montagner, A., & Reich, A. (2018). Unlocking success in digital transformations. McKinsey. https://www.mckinsey.com/capabilities/people-and-organizational-per formance/our-insights/unlocking-success-in-digital-transformations de Paula, D., Hahn, D., Matthies, C., & Uebernickel, F. (2022a, January 4). InnoPulse: a Tool to support collaborative reflection in creativity-driven engineering projects. Proceedings of the 55th Hawaii International Conference on System Sciences. https://doi.org/10.24251/HICSS. 2022.005. de Paula, D., Marx, C., Wolf, E., Dremel, C., Cormican, K., & Uebernickel, F. (2022b). A managerial mental model to drive innovation in the context of digital transformation. Industry and Innovation, 1–25. https://doi.org/10.1080/13662716.2022.2072711 Design Thinking DTINGRE. (2012). Larry Leifer on design thinking & the “Hunter” metaphor. Retrieved from https://www.youtube.com/watch?v=VhzrxPgDHbY Douglas, M. (1986). How institutions think. In Journal of Public Policy (Issue 1). Syracuse University Press. Retrieved from https://www.cambridge.org/core/product/identifier/S0143814 X00004360/type/journal_article Dow, S. P., Fortuna, J., Schwartz, D., Altringer, B., Schwartz, D. L., & Klemmer, S. R. (2012). Prototyping dynamics: Sharing multiple designs improves exploration, group rapport, and results. In Design thinking research (pp. 47–70). Springer Berlin Heidelberg. https://doi.org/ 10.1007/978-3-642-31991-4_4. Dow, S. P., Glassco, A., Kass, J., Schwarz, M., Schwartz, D. L., & Klemmer, S. R. (2010). Parallel prototyping leads to better design results, more divergence, and increased self-efficacy. ACM Transactions on Computer-Human Interaction, 17(4), 1–24. https://doi.org/10.1145/1879831. 1879836 Drucker, P. F. (1954). The practice of management. Harper & Brothers. Edelman, J. A., Owoyele, B., Santuber, J., & Talbot, A. V. (2021). Designing as performance: Bridging the gap between research and practice in design thinking education. 75–101. https:// doi.org/10.1007/978-3-030-62037-0_3. Elsbach, K. D., & Stigliani, I. (2018). Design thinking and organizational culture: A review and framework for future research. Journal of Management, 44(6), 2274–2306. https://doi.org/10. 1177/0149206317744252 Foley, C. F., Mayfield, E. S., & Boland, F. K. (2017). The transformation of microsoft. Harvard Business School Case, 218-048.

320

A. Kerguenne et al.

Forth, P., Reichert, T., de Laubier, R., & Chakraborty, S. (2020). Flipping the odds of digital transformation success. The Boston Consulting Group. https://www.bcg.com/de-de/publica tions/2020/increasing-odds-of-success-in-digital-transformation Gabrysiak, G., Giese, H., & Seibel, A. (2011). Towards next generation design thinking: scenariobased prototyping for designing complex software systems with multiple users. In Design thinking (pp. 219–236). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-64213757-0_13. Guentert, M., Luebbe, A., & Weske, M. (2014). Sharing knowledge through tangible models: Designing kickoff workshops for agile software development projects. In Design thinking research (pp. 203–218). Springer International Publishing. https://doi.org/10.1007/978-3-31901303-9_13. Haas, O., North, K., & Pakleppa, C.-B. (2022). Transformation Tiefgreifende Veränderungen verstehen, ermöglichen und gestalten. Vahlen. https://www.vahlen.de/haas-north-pakleppalust-transformation/product/32317708 Haskamp, T. (2021). Performance measurement of design thinking: Conceptualisations, challenges and measurement approaches. In H. Plattner & L. Leifer (Eds.), Design thinking research. Understanding innovation (pp. 273–295). Springer. https://doi.org/10.1007/978-3-030-763244_15 Haskamp, T., Dremel, C., Marx, C., & Uebernickel, F. (2021). Understanding inertia in digital transformation: A literature review and multilevel research framework. Proceedings on the International Conference on Information Systems (ICIS). Hölzle, K., & Rhinow, H. (2019). The dilemmas of design thinking in innovation projects. Project Management Journal, 50(4), 418–430. https://doi.org/10.1177/8756972819853129 Innovation | Etymology Dictionary. (n.d.). Retrieved December 20, 2022, from https://www. etymonline.com/search?q=innovation Jalonen, H. (2011). The uncertainty of innovation: A systematic review of the literature. Journal of Management Research, 4, 1. https://doi.org/10.5296/jmr.v4i1.1039 Jobst, B., & Meinel, C. (2014). How prototyping helps to solve wicked problems. In Design thinking research (pp. 105–113). Springer International Publishing. https://doi.org/10.1007/ 978-3-319-01303-9_8. Kane, G. C., Palmer, D., Phillips, A. N., Kiron, D., & Buckley, N. (2015). Strategy, not technology, drives digital transformation—Becoming a digitally mature enterprise. Retrieved from http:// sloanreview.mit.edu/projects/strategy-not-technology-drives-digital-transformation Kolb, D., & Fry, R. (1974). Toward an Applied Theory of Experiential Learning. In Theories of group process. MIT Sloan School of Management. Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development. Prentice Hall. Kolko, J. (2015). Design thinking comes of age. Harvard Business Review. Retrieved from https:// hbr.org/2015/09/design-thinking-comes-of-age Lorson, A., de Paula, D., Dremel, C., & Uebernickel, F. (2022). Beyond the fast lane narrative—A temporal perspective on the unfolding of digital innovation in digital innovation units. Thirtieth European Conference on Information Systems. Retrieved from https://www.researchgate.net/ publication/360355243 Marx, C., Haskamp, T., & Uebernickel, F. (2022). The Nexus of design thinking and intrapreneurship: Insights from a large-scale empirical assessment. The Nexus of Design Thinking and Intrapreneurship. Retrieved from https://hdl.handle.net/10125/79992 Mayer, S., Haskamp, T., & de Paula, D. (2021). Measuring what counts: An exploratory study about the key challenges of measuring design thinking activities in digital innovation units. Proceedings of the Annual Hawaii International Conference on System Sciences, 2020January, 4951–4960. https://doi.org/10.24251/HICSS.2021.602. Nagel, R., & Wimmer, R. (2014). Systemische Strategieentwicklung (6th ed.). Schäffer-Poeschel.

Opportunities and Limitations of Design Thinking as Strategic Approach. . .

321

Newman, D. (2017). Innovation vs. transformation: The difference in a digital world. Forbes. Retrieved from https://www.forbes.com/sites/danielnewman/2017/02/16/innovation-vs-transfor mation-the-difference-in-a-digital-world/?sh=fad5fdd65e8b Ney, S., Kerguenne, A., & Meisel, M. (2022). Towards an adaptive cultural maturity model for navigating digital transformation in organizations. SSRN. Retrieved from https://papers.ssrn. com/sol3/papers.cfm?abstract_id=4265743 Ney, S., & Meinel, C. (2019). Putting design thinking to work. Springer International Publishing. https://doi.org/10.1007/978-3-030-19609-7 Noweski, C., Scheer, A., Büttner, N., von Thienen, J., Erdmann, J., & Meinel, C. (2012). Towards a paradigm shift in education practice: developing twenty-first century skills with design thinking. In Design thinking research (pp. 71–94). Springer Berlin Heidelberg. https://doi.org/10.1007/ 978-3-642-31991-4_5. Pello, R. (2018). Management consultancy industry upgraded with design. Medium. Retrieved from https://medium.com/@pello/management-consultancy-industry-upgraded-with-design-b1 f323e51945 Potsdam Eins. (2012, September 20). Hasso Plattner über Design Thinking. Retrieved from https:// www.youtube.com/watch?v=CdUe1WHrUsA Rhinow, H. (2018, May 24). Observations on design thinking, lean startup and scrum (1): Common Ground. Medium. Retrieved from https://medium.com/@holger.rhinow/observations-ondesign-thinking-lean-startup-and-scrum-1-common-ground-fd7e9ea93e91 Rittel, H. W. J., & Webber, M. M. (1973). Dilemmas in a general theory of planning*. Policy Sciences, 4, 155–169. Royalty, A., Chen, H., Roth, B., & Sheppard, S. (2021). Developing a tool to measure the transfer of design practice from training contexts to applied contexts (pp. 103–121). Springer, Cham. https://doi.org/10.1007/978-3-030-62037-0_4. Royalty, A., & Roth, B. (2016). Mapping and measuring applications of design thinking in organizations (pp. 35–47). Springer. https://doi.org/10.1007/978-3-319-40382-3_4 Santuber, J. (2019). A framework theory of legal design for the emergence of change in the digital legal society. Rechtstheorie, 50(1), 41–58. https://doi.org/10.3790/rth.50.1.41 Schallmo, D., Williams, C. A., & Boardman, L. (2017). Digital transformation of business models—Best practice, enablers, and roadmap. International Journal of Innovation Management, 21(08), 1740014. https://doi.org/10.1142/S136391961740014X Schäpke, N., Stelzer, F., Caniglia, G., Bergmann, M., Wanner, M., Singer-Brodowski, M., Loorbach, D., Olsson, P., Baedeker, C., & Lang, D. J. (2018). Jointly experimenting for transformation? Shaping real-world laboratories by comparing them. GAIA - Ecological Perspectives for Science and Society, 27(1), 85–96. https://doi.org/10.14512/gaia.27.S1.16 Schneidewind, U. (2018). Die große Transformation: eine Einführung in die Kunst gesellschaftlichen Wandels. S. Fischer Verlag. Sheppard, S. D., Chen, H. L., Toye, G., Kempf, F., & Elfiki, N. (2021). Measuring the impact of project-based design engineering courses on entrepreneurial interests and intentions of alumni (pp. 297–313). Springer. https://doi.org/10.1007/978-3-030-76324-4_16 Sinek, S. (2009). Start with why: How great leaders inspire everyone to take action. Penguin. Retrieved from https://books.google.de/books?hl=de&lr=&id=iyw6ul1x_n8C&oi=fnd&pg= PR9&dq=simon+sinek+golden+circle&ots=UNG-o-wEY_&sig=swJMaNw6x7cflR2 yPpocNezecp8&redir_esc=y#v=onepage&q=simon%20sinek%20golden%20circle&f=false Snowden, D. (2000). The social ecology of knowledge management. In Knowledge horizons (pp. 237–265). Elsevier. https://doi.org/10.1016/B978-0-7506-7247-4.50015-X Stacey, R. D. (2007). Strategic management and organisational dynamics: The challenge of ... Ralph D. Stacey - Google books (5th ed.). Pearson Education Limited. Retrieved from https:// books.google.de/books?hl=de&lr=&id=PENbGZDXTU4C&oi=fnd&pg=PR12&dq=Strate gic+Management+and+Organisational+Dynamics:+The+Challenge+of+Complexity+3&ots= IFf9Bh2QhN&sig=L6lk6EYCHGcMutoLkdU7LhG-9Kg&redir_esc=y#v=onepage&q=

322

A. Kerguenne et al.

Strategic%20Management%20and%20Organisational%20Dynamics%3A%20The%20Chal lenge%20of%20Complexity%203&f=false Tabrizi, B., Lam, E., Girard, K., & Irvin, V. (2019). Digital transformation is not about technology. Harvard Business Review Digital. Retrieved from https://store.hbr.org/product/digitaltransformation-is-not-about-technology/h04to3?sku=H04TO3-PDF-ENG Thompson, M., Ellis, R., & Wildavsky, A. (1990). Cultural theory. Westview Press. Traifeh, H., Staubitz, T., & Meinel, C. (2019). Improving learner experience and participation in MOOCs: A design thinking approach. 2019 IEEE Learning With MOOCS (LWMOOCS), 165–169. https://doi.org/10.1109/LWMOOCS47620.2019.8939623. Transformation | Etymology Dictionary. (n.d.). Retrieved December 20, 2022, from https://www. etymonline.com/search?q=transformation Tuckman, B. W. (1965). Developmental sequence in small groups. Psychological Bulletin, 63(6), 384–399. https://doi.org/10.1037/H0022100 Unruh, G., & Kiron, D. (2017). Digital transformation on purpose. MITSloan Management Review. Retrieved from https://sloanreview.mit.edu/article/digital-transformation-on-purpose/ Verhoef, P. C., Broekhuizen, T., Bart, Y., Bhattacharya, A., Qi Dong, J., Fabian, N., & Haenlein, M. (2021). Digital transformation: A multidisciplinary reflection and research agenda. Journal of Business Research, 122, 889–901. https://doi.org/10.1016/j.jbusres.2019.09.022 von Thienen, J., Noweski, C., Rauth, I., Meinel, C., & Lang, S. (2012). If you want to know who you are, tell me where you are: The importance of places. In Design thinking research (pp. 53–73). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-21643-5_4. von Thienen, J. P. A., Clancey, W. J., & Meinel, C. (2021). Theoretical foundations of design thinking. Part III: Robert H. McKim’s visual thinking theories (pp. 9–72). https://doi.org/10. 1007/978-3-030-62037-0_2 Wade, M., Bonnet, D. C. L., & Shan, J. (2020). The new elements of digital transformation. MIT Sloan Management Review, 62(2), 495–501. https://doi.org/10.1108/JSMA-05-2020-0116/ FULL/HTML Weinreich, U. (2016). Lean digitization. In Lean digitization. Springer Berlin Heidelberg. https:// doi.org/10.1007/978-3-662-50502-1. Westerman, G., Calméjane, C., Bonnet, D., Ferraris, P., & McAfee, A. (2011, November). Digital transformation: A road-map for billion-dollar organizations, report. Capgemini Consulting & MIT Center for Digital Business. Yin, R. K. (2017). Case study research and applications (6th ed.) In Paper knowledge. Toward a media history of documents (6th ed.). SAGE Publishing.

Designing Innovation in the Digital Age: How to Maneuver around Digital Transformation Traps Carolin Marx, Thomas Haskamp, and Falk Uebernickel

Abstract Companies that embrace human-centered digital innovation approaches like Design Thinking are better equipped to master the challenges of digital transformation. Still, the number of failed digital transformation activities is rising, and companies struggle more than ever to turn their transformation initiatives into a success to remain competitive in the digital age. One promising avenue toward detangling this phenomenon is to identify the “traps” that, if not identified or addressed, can cause an organizational digital transformation to run into distress and potentially fail. In this chapter, we discuss the potential traps organizations can fall into during their digital transformation activities. We then synthesize them based on their difficulty to identify in a three-layered framework. The framework consists of execution traps, reaction traps, assessment traps, and recognition traps. Based on an in-depth case study with a large incumbent firm in the banking sector, we validate the trap categories and provide guidelines on how to use the framework in practice. This not only adds to the “demystification” of failing digital transformation endeavors but also serves as a foundation for understanding how adopting humancentered digital innovation approaches or other mitigation strategies can help to maneuver around the traps of transformation.

1 Introduction Companies that embrace human-centered digital innovation approaches like Design Thinking are evidently better equipped to master the challenges of organizational transformation. Adopting Design Thinking creates higher customer proximity in IT departments (Vetterli et al., 2016), improves requirement engineering (Hehn & Uebernickel, 2018), and accelerates digital transformation (Magistretti et al., 2021; Marx, 2022). There exists a profound understanding in academia of how specific Design Thinking elements can be linked to the building blocks of digital C. Marx (✉) · T. Haskamp · F. Uebernickel Hasso Plattner Institute, University of Potsdam, Potsdam, Germany e-mail: [email protected]; [email protected]; [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_15

323

324

C. Marx et al.

transformation activities (Marx, 2022). However, we cannot explain exactly how the adoption of Design Thinking might contribute to the gap between companies managing to succeed through digital transformation compared to those that fail this challenge. One promising avenue toward detangling this relationship is to take a step back and look at the “traps” that can cause an organizational digital transformation to fail—unless they are recognized and addressed. The digital transformation of organizations is a topic that has been hyped for years and declared dead by many (Gale, 2021). After thousands of articles, a plethora of models, discussions, and special issues on the subject, and a whole new digital acceleration brought to the fore by the global pandemic, there is most certainly no executive left in the world who has not heard about digital transformation. The Google Trends interest score for the search term “digital transformation” has more than doubled within the past 5 years (Wade et al., 2020). Yet, at the same time, the number of failed digital transformation activities remains high and rising. How can it be that after all these years of knowledge and experience-gathering, firms are still struggling to turn their digital transformation initiatives into a success and remain competitive in the digital age? Taking a closer look, one can see that managing and leading this “continuous, complex undertaking that can substantially shape a company and its operations” (Matt et al., 2015, 341) has become one of the most challenging tasks in today’s business world (Vial, 2019). To remain competitive, companies need to integrate digital technologies into products, processes, and business models to successfully develop and implement digital innovations (Fichman & Dos Santos, 2014). This imposes challenges throughout the whole process of innovating. Given this high relevance, it is not surprising that academic research and practitioner reports on challenges, barriers, and failure causes of digital transformation activities are plentiful, but also highly fragmented. While the focus has been on challenges in general, it would be more interesting to study those situations and conditions where a negative outcome is very likely but still avoidable. Hence, considering that the goal is to eventually turn distressed digital transformation initiatives around, it is most promising to search for potential “traps”—things that companies should try to avoid or, if already too late, at least be aware of— instead of looking at challenges or barriers in general. Thus, this chapter aims to identify and synthesize the potential traps organizations could fall into during their digital transformation journey. Looking at these pitfalls not only adds to the “demystification” of failing digital transformation endeavors but also serves as a foundation for understanding how adopting Design Thinking or other mitigation strategies can help us maneuver these minefields. We, therefore, pose the following research question:

Designing Innovation in the Digital Age: How to Maneuver around. . .

1.1

325

What Are the Biggest Traps for Organizations when Digitally Transforming?

To address this question, we developed an integrative trap framework for digitally transforming organizations. Regarding research design, our framework builds on two steps. In the first step, we conducted a literature review on digital transformation traps following our conceptualization of “traps.” As part of this literature review, we analyzed the literature on digital transformation in management, information systems, and psychology regarding potential traps, their potential impact, and how difficult it is to identify them in practice. The resulting list of traps from the interviews and the literature served then as a basis for several research workshops in which the resulting framework was developed. We further validated our frameworks and findings through expert discussions with MIT Center for Information Systems Research members in Boston, USA, and experts from the Haas School of Business at the University of California, Berkeley. In a second step, we applied our framework to practice. For this, we ran an in-depth case study (Yin, 2011) with a large, established firm in the banking sector that is in the middle of an agile transformation. As part of this framework validation, we applied the identified traps to the case and used the framework to assess the biggest potential traps. The trap framework developed in this study is based on two main assumptions. First, we are convinced that traps are not generic but specific to the activities that organizations employ when digitally transforming. We will therefore start with a conceptualization of digital transformation activities and so-called explosions, which will serve as the foundation for identifying traps. Second, our conceptualization of traps implies that stepping in a trap likely, but not necessarily, leads to negative outcomes. Hence, a core factor for synthesizing the traps is identifying their existence and assessing their probability. After elaborating on the different digital transformation activities that serve as the context for traps to unfold, we will therefore discuss the traps based on their difficulty to be identified in practice.

2 Digital Transformation Activities and Explosions The term digital transformation has been used extensively in the past years to describe the overall impact digital technologies have on organizations and societies (Baiyere et al., 2017). With regard to the organizational layer, we understand digital transformation as “the combined effects of several digital innovations bringing about novel actors (and actor constellations), structures, practices, values, and beliefs that change, threaten, replace or complement existing rules of the game within organizations, ecosystems, industries, or fields” (Hinings et al., 2018, 53). Thus, the boundary-encompassing characteristics of digital technology seem to trigger a set of fundamental changes on the societal and organizational levels (Appio et al., 2021). In particular, incumbent companies are challenged by digital transformation,

326

C. Marx et al.

as they—in contrast to born-digital companies—need to change their entire business model to secure organizational survival (Vial, 2019; Sebastian et al., 2017). This includes strategic changes that companies experience due to the impact of digital technology. Hence, digital technologies threaten existing value-creation paths or afford new paths for organizations to pursue. Especially large incumbent organizations are subject to strong inert forces (Haskamp, Dremel, et al., 2021a). To ensure organizational survival, they need to engage in digital transformation activities that aim to renew the existing structures, as well as processing and adopting a digital culture and mindset to avoid organizational death.

2.1

Digital Transformation Activities

Current research refers to digital transformation activities as a broad set of actions and activities that companies pursue in order to drive digital transformation. To kickstart digital transformation, companies launch dedicated digital transformation strategies or roadmaps and derive initiatives, programs, and projects (Barthel, et al., 2020b). These activities may unfold in different organizational domains. In terms of specific activities, a set of instruments has emerged over the past years to drive digital transformation. We briefly highlight two exemplifying activities that in particular incumbents pursue. To increase speed and overcome high inertia in terms of structure and processes, many incumbent organizations try to adjust their organizational setup by, for example, setting up a digital transformation office, often closely affiliated with the organization’s CEO, CIO, or CDO (Barthel, et al., 2020b). Further, companies set up dedicated organizational units—so-called digital innovation units—to develop new digital business models or to digitize existing processes (Barthel, et al., 2020a). These units are granted greater freedom for action and are often used as speedboats to develop new digital innovations (Barthel, et al., 2020a). While their success is hard to assess (Haskamp, et al., 2021c) and their specific design varies a lot depending on the context, they have become an integral part of the digital transformation efforts in companies (Dremel et al., 2017). Further, building digital capabilities often comes with adopting new ways of working (van der Meulen et al., 2020). This usually includes the company-wide adoption of agile methods and frameworks such as Scrum and SAFe (Fuchs & Hess, 2018) or the adoption of human-centered design methods such as Design Thinking to develop digital products or upskill existing employees (Wang, 2022). Initially born in digital software development, these methods now diffuse into other organizational areas such as marketing, human resources, operations, or product development.

Designing Innovation in the Digital Age: How to Maneuver around. . .

2.1.1

327

Status Quo: Dealing with Spaghettis and Silos

In transforming their existing organization, many firms need to redesign their current structure and processes (Sebastian et al., 2017). In many cases, existing approval or budgeting processes are too slow to cope with changing environmental factors and with the required speed to develop new products and services. Thus, tensions are likely to emerge between the requirements for digital product and service development and the already existing analog products and processes (Haskamp et al., 2023). In practice, successfully transforming a company’s business requires dealing with the existing legacy that organizations have accumulated over the past years. This legacy can become very sticky, with organizational inertia the reason behind the high failure rate of many digital transformation activities (Forth et al., 2020). Inertia appears in different forms as part of an organization’s digital transformation journey (Haskamp, et al., 2021a): • Structural inertia refers to existing processes and structures in place that have served in the past successfully but now emerge as a barrier to the successful development of digital innovation. • Socio-cognitive inertia refers to cultural traditions and values that have secured success in the past but have now become a challenge in implementing digital technology. • Socio-technical inertia refers to existing IT infrastructures, such as enterprise resource planning systems, which may become a barrier with hard-to-access data or missing interfaces undermining new digital efforts. • Political inertia refers to vested interests that companies need to overcome to launch new products and services. • Economic inertia refers to resource decision conflicts between exploration and exploitation undermining change efforts. Taken together, firms must find ways to deal with these issues, as existing products and services are often established in organizational silos, which are connected through a set of complex processes and practices, also referred to as spaghettis (Woerner et al., 2022; van der Meulen et al., 2020). In order to be able to launch new digital products and services, established organizations need to find ways to detangle these spaghettis and overcome silo-thinking within the company. While different pathways exist to deal with the existing legacy, companies all share the same goal: become future-ready (Weill & Woerner, 2018).

2.1.2

The Objective: Becoming Future-Ready

Future-ready companies can overcome these issues by mastering the multiple tensions involved in pursuing digital transformation and acting ambidextrously. This means that they are innovating with new products and services and continue to serve changing customer needs while simultaneously focusing on cost reduction

328

C. Marx et al.

and efficiency, ensuring an effective and efficient operational backbone (Weill & Woerner, 2018). Research shows that companies that can master both outperform their peers by far, not only regarding financial results but also in championing employee satisfaction and customer scores (van der Meulen et al., 2020). Hence, becoming future-ready is imperative for successful transformation.

2.2

Organizational Explosions

However, becoming future-ready is a challenging endeavor, as it requires serving existing customers while changing the business model at the same time. Depending on how companies try to become future-ready, they experience so-called organizational explosions (Weill et al., 2018; Woerner et al., 2022). Organizational explosions refer to general issues companies must confront as part of their digital transformation journey. Often these issues are rooted in tensions grounded in the existing way of doing things optimized for the effectiveness and efficiency of current operations which are now being challenged with new requirements to develop digital products or services (Haskamp et al., 2023). Thus, as companies pursue digital transformation, these conflicts emerge as organizational explosions, which executives are forced to deal with (Weill et al., 2018). Current research (van der Meulen et al., 2020) on these explosions distinguishes between four types: changing decision rights, adopting new ways of working, performing organizational surgery, and adopting a platform mindset.

2.2.1

Changing Decision Rights

Companies often need to introduce new processes, projects, and products to realize digital innovation and carry out a successful transformation. This requires a change in decision rights, affecting business practices such as funding, reporting, and hiring. With these changes comes the question of whether the decision rights of existing management members are extended or whether new joining management members are driving digital efforts to take over responsibility. These decisions often involve questions around funding and budgeting, but also about what products the company is supposed to sell, which customers to target, and how to position the company in the market. A core answer to these questions is found in the changing roles of information technology within organizations. Whereas formerly, the management of digital technologies was assigned to CIOs and their teams, with the all-encompassing character of digital technology, this management has become pervasive in all functions and products of organizations. Thus, aligning business and IT is not enough but is a prerequisite to developing new digital products. This also requires rethinking what structures need to be in place to support the development of digital products and who is in charge of the development. Therefore, changing decision rights becomes a major organizational explosion that executive

Designing Innovation in the Digital Age: How to Maneuver around. . .

329

teams must handle over the course of their digital transformation efforts. These changes are related to changes in performance metrics, incentives, budgeting practices, and approval processes, which challenge the organization’s power and status relationship (van der Meulen et al., 2020).

2.2.2

Adopting New Ways of Working

Another relevant organizational explosion as part of digital transformation comes with the adoption of so-called new ways of working. These often refer to the adoption of methods such as Scrum, SAFe, Lean Startup, and Design Thinking that are implemented and scaled in organizations to make organizations more customer-centric and to increase speed within product development (Fuchs & Hess, 2018; Magistretti et al., 2021). These methods are often closely related and applied for digital products and are then transferred to other contexts and departments such as operations, marketing, and HR to foster a more collaborative, customer-centered, and faster approach to work. While this transfer to other organizational areas is often challenging and becomes subject to discussions (Fuchs & Hess, 2018), there is nevertheless a fundamental shift in terms of organizations’ required speed and customer-centricity. To achieve this, these methods aim to increase customer proximity, often involving customers directly in the co-creation or co-development of products and services.

2.2.3

Performing Organizational Surgery

The explosion of organizational surgery refers to activities around the restructuring of business processes, reworking corporate hierarchies, and changing corporate functions and roles (Weill et al., 2018). These restructuring activities have been pursued by organizations for many years and belong to the bread and butter of executives. Still, their extent in the context of digital transformation often exceeds prior experiences. For example, elements of digital transformation such as reorganizing product and service delivery based on organizational structures or establishing a new product order and delivery process are often hard to avoid. Further, installing a new digital innovation unit raises power conflicts between existing and new stakeholders. It also raises the question of product and development responsibilities and, in the end, questions about where digital technology is supposed to be located in the organization. Hence, “organizational surgery” is not an easy undertaking and often involves the management of numerous conflicts and tensions. Still, it becomes necessary to prevent the organization from becoming subject to organizational inertia and to ensure efficiency, effectiveness, and smooth integration of new and ongoing operations.

330

2.2.4

C. Marx et al.

Creating a Platform Mindset

Digital transformation further requires companies to deal with the explosion of creating a platform mindset. Developing digital products and services requires a fundamental change in how the organization is operated and how employees think about their daily lives. Whereas companies previously focused on providing a specific product or service to one specific customer group, today’s successful companies, such as Amazon or Spotify, are platforms providing a digital infrastructure that creates an ecosystem for several stakeholders and customer groups. Thus, thinking in terms of a platform requires building modular technology that is implemented quickly and in which application program interfaces (APIs) ensure seamless integration of other services. Therefore, adopting a platform mindset affects the product and service infrastructure and is about moving toward a mindset of continuous delivery. While organizations previously relied on projects as a vehicle of product development, with the adoption of digital technology, product development became a constant task, requiring continuous resource supply for digital product teams.

3 Digital Transformation Traps Based on our literature review and the conceptualization of activities and explosions, we have developed an integrated framework synthesizing the different kinds of traps organizations should look out for. The categories are execution traps, reaction traps, assessment traps, and recognition traps. We have further identified eight subcategories and grouped the traps according to their difficulty to be identified in practice into three layers. The resulting framework of digital transformation traps is displayed in Fig. 1 and will be explained in detail within the following subchapters.

3.1

Recognition Traps

Recognition traps refer to biases in the very beginning of thought processes and attitudes regarding the organization’s digital transformation activities. Biases describe systematic faulty tendencies in perceiving, remembering, thinking, and judging (Tversky & Kahneman, 1974). These tendencies usually remain unconscious and are based on cognitive heuristics. Hence, this trap category can be difficult to identify in practice. While remaining unnoticed, the single effects can cascade into a deluge that threatens the very core of a successful digital transformation endeavor.

Designing Innovation in the Digital Age: How to Maneuver around. . .

331

Fig. 1 The three layers of digital transformation traps

3.1.1

Biased Goal Setting

Short-term thinking/protection of profitability: Unfortunately, digital transformation programs are often viewed as a destination rather than a journey by executives driving the effort. This results in choices and investments that enable short-term success but shackle the program from evolving meaningfully once the business operations kick in. This lack of long-term commitment and impatience can easily threaten the success of the organization’s digital transformation and is one of the major underlying traps we have identified. Pursuit of perfection: While in general, there is a positive connotation with the word perfection, there exists, however, a significant difference between striving for excellence and insisting on perfection. Insisting on perfection and overemphasizing details can be a major barrier to dynamically adapting through experimentation. Especially in the context of digital transformation initiatives, perfectionism can severely backfire. Insisting on perfection when striving for organizational transformation can put an incredible strain on employees and other stakeholders. It adds to the risk of setting too high expectations and can potentially slow change processes down. In practice, however, pursuing perfection is a common phenomenon frequently standing in the way of successful digital transformation activities.

332

C. Marx et al.

Copying others: As a potential response to uncertainty, organizations within a similar field face the general tendency to adopt related behaviors, hence becoming increasingly similar to each other, for instance, in terms of processes and internal structure (DiMaggio & Powell, 1983). By modeling the success of others, the imitating organization hopes to obtain success for itself. However, given the complex and dynamic nature of digital transformation where recognizing threats and opportunities under uncertainty is key, the process of copying others can eventually impede change. In most cases, the combination of external and internal factors shaping the need and the conditions for digital transformation is not the same between companies. Moreover, copying can become a trap, as the actual factors that have led the leading company to its success are often based on actions that are invisible to outside companies. Hence, the underlying more intangible changes a company goes through to achieve successful transformation are extremely challenging if not impossible to imitate.

3.1.2

Biased Decision-Making

Fear of failure and experimenting: In a digital, ever-changing environment, organizations have to constantly adapt and strive for flexibility in their decision-making. However, with high levels of uncertainty and rapidly changing conditions, the path toward success is extremely difficult to plan in advance and nearly impossible to predict. Hence, experimenting and dynamically adapting are key actions. In practice, however, one can frequently observe low experimentation and risk-averse behavior in organizations aiming to transform digitally. This tendency can be linked to the fear of doing things “the wrong way,” which is often rooted in the organization’s enforced pursuit of perfectionism. Escalation of commitment: Given the complex, uncertain, and dynamic nature of digital transformation endeavors, many initiatives continue to run over budget, extend past schedule, and deliver less than or different outcomes than anticipated. While persistence is intuitively associated with success, it can also be applied to distressed or generally troubled digital transformation initiatives. In this case, being overly persistent and committed without recognizing negative signs can become a severe trap. Escalation of commitment—the failure to withdraw from negative courses of action (Staw, 1981)—is a phenomenon that can be frequently observed in practice. The failure to address escalation of commitment and to turn distressed initiatives around can lead to declining managerial performance and firm profitability, productivity loss, and increased organizational inertia, which eventually threatens the very core of a potentially successful digital transformation (Marx & Uebernickel, 2022). Excessive caution: Although it goes without saying that a business should not be reckless in the decisions it makes or how it operates, when a company undertakes a digital transformation, it is easy to find organizations at the other extreme: those who are overly cautious. To some degree, it is possible to plot out the system and process changes that need to be made, but with that change there ought to be an appreciation

Designing Innovation in the Digital Age: How to Maneuver around. . .

333

of the element of risk. Implementing incremental changes is unlikely to result in the transformation desired. Hence, a certain comfort with risk in order to implement bold and innovative strategies is essential when trying to change the status quo. Unfortunately, excessive caution is a frequently observed behavior and a major risk when putting digital transformation strategies into action.

3.2

Assessment Traps

Assessment traps refer to inaccuracies in evaluating opportunities and challenges related to digital transformation activities and resulting organizational change. We have identified a plethora of potential traps organizations could fall into when making sense of potential challenges and risks and assessing market opportunities or existing capabilities. While assessment on the individual level is a cognitive process involving rational thinking, there are major parts in the process of assessing challenges and opportunities that rely on intuition and remain unconscious. Hence, the traps in this category are considered as difficult to identify in practice. When remaining unnoticed, underestimating challenges and the overall complexity of a transformation in combination with overestimating opportunities and existing capabilities can easily snowball and threaten the success of the organization’s digital transformation.

3.2.1

Underestimating Challenges and Complexity

Forgetting about infrastructure: In a survey study about small and medium-sized enterprises, Shevtsova et al. (2020) found that challenges related to supporting technical infrastructure are key technical barriers when it comes to implementing digital transformation initiatives. Changing processes or implementing new technology as part of a digital transformation project requires in most cases the foundation of a suitable technical infrastructure. In practice, however, we frequently observe that those dependencies and the necessity to build suitable product architectures and infrastructure before implementing technological changes are overlooked. For instance, many organizations struggle with the technical integration of digitally extending an analog product because architectural challenges are underestimated. In general, many organizations underestimate the complexity and the dependencies related to digitizing processes, reorientating their business model, or leveraging data which can jeopardize the entire initiative (Vogelsang et al., 2019). Underestimation of complexity and dependencies of changing processes: Digital transformation is often referred to as a never-ending mammoth project that is likely to affect every single aspect of an organization. Still, many organizations underestimate the complexity and the dependencies related to organizational change. Underestimating cultural barriers, for instance, is frequently brought up as a major reason for poor digital transformation performance (Wade et al., 2020).

334

C. Marx et al.

Underestimating existing IT legacy: Given the complexity of average IT architectures in organizations, underestimating existing IT legacy is another major trap organizations can fall into when digitally transforming. The existing landscape of IT systems may impose limits on the integration of new digital innovations. Remaining unaddressed with substantial integration efforts, those existing structures can impede transformation efforts, potentially slowing down or even hindering the success of integrating new digital innovations. For these integration efforts to be successful, it is crucial that the initial assessment of potential IT legacy barriers is accurate. The high number of IT projects running over budget and time, however, indicates that this assessment is far from being accurate in most cases. Underestimating regulatory issues: Along the same lines of underestimating the general complexity of changing processes and threats, such as existing IT legacy slowing down integration projects, organizations often struggle with regulatory issues related to digital transformation activities. Barriers in the form of legal policies are a major cause of transformation failure (Orzes et al., 2018). Underestimating regulatory issues that are particularly present in data-related activities can impede the change process and endanger the success of the overall digital transformation. Underestimating potential of daily business to hijack DT agenda: Entities, functions, roles, and established stakeholder constellations have to be reorganized to realize digital transformation. As a result, new internal units emerge that try to find their role within the organization. During these complex changes associated with digital transformation, there is a significant risk that employees are still “trapped” within old roles and responsibilities, while the transformation efforts require substantial resources. Along the same lines, a common conflict in practice is a battle for resources, where the daily business often “hijacks” the digital transformation agenda (Heavin & Power, 2018).

3.2.2

Overestimating Opportunities and Capabilities

Overestimation of existing capabilities: While the underestimation of challenges is a severe threat to the success of transformation efforts, so is the overestimation of opportunities and capabilities. On an individual level, inaccurate capability assessments due to overestimation as a form of overconfidence or ignorance are a common trap (Moore & Healy, 2008). The tendency to overestimate the existing capabilities and resources of an individual but also of the entire organization can severely backfire, especially in the context of digital transformation. In an uncertain and disrupted environment, organizations have to build dynamic capabilities to remain competitive through digital transformation (Marx et al., 2021; Teece et al., 1997). For this, an accurate self-assessment is essential. Hence, overestimating existing capabilities is a common trap for organizations that aim to digitally transform. If unaddressed, this can significantly endanger the transformation’s success. Falling for hyped topics: Digital transformation is on every executive’s agenda and one of the most popular topics in the management press. This hype, however, also has its downsides as it creates buzz and excitement about specific technologies

Designing Innovation in the Digital Age: How to Maneuver around. . .

335

or topics. Companies believed they had to do something—and fast. In many cases, falling for hyped topics and technology without carefully creating a digital roadmap has led to expensive efforts that did not always produce transformative results. Hence, it is a major challenge to carefully separate hype from reality and to assess the actual fit and advantages of specific actions instead of “doing things just for the sake of doing them.” Overemphasizing technology over people: With technological advancements and changing market dynamics come changing requirements and needs regarding successfully operating a business in the digital age. Unfortunately, many organizations assess these new requirements and needs only through the lens of technology while not giving much thought to what these choices mean for the employees who need to adopt it. Too often digital transformation programs are exclusively focused on digital technologies (Wade et al., 2020). As digital transformation is as much about people as it is about technology (Marx et al., 2023; Kane, 2019), recognizing the challenges the transformations pose for the individuals involved, and explicitly addressing these issues is crucial. Acknowledging the importance of the human factor in digital transformation, academia, and practice identifies human-centricity as an essential foundation to build when digitally transforming (Marx, 2022; Magistretti et al., 2021). Starting with step 100: Many organizations face the dilemma of whether to increase the efficiency of current operations as a top priority or to emphasize customer-centricity and address needs (Heavin & Power, 2018). Setting the right priorities and focusing on the plethora of potential digital transformation activities are a crucial task to successfully maneuver around digital transformation traps. In practice, however, many companies have adopted a “let a hundred flowers bloom” philosophy. Such an approach generates excitement and fosters experimentation, but, if not carefully managed, it can also be self-defeating. Ignoring the necessity of building foundations, for instance, suitable IT infrastructure, and running too many competing initiatives are success-threatening results of this lack of focus and priorities.

3.3

Reaction Traps

Digital transformation at its core is about change. Reaction traps refer to unfavorable reactions caused by those changes that potentially impede digital transformation. These reactions can be rooted in two main areas. First is in the inability to change due to a lack of resources, miscommunications, or support deficit from management or from within the organization. Second, the reactions can be tied to conscious or unconscious unwillingness to change. When remaining unaddressed, reaction traps can jeopardize otherwise smoothly running transformation initiatives and reinforce inert forces. While the “symptoms” of reaction traps are comparatively easy to identify, like employee resistance to change, it is still extremely challenging to notice the complex interdependencies and influencing factors that might cause

336

C. Marx et al.

unfavorable reactions to digital transformation activities within organizations. Hence, we have placed this trap category in the middle layer of identification difficulty within our trap framework.

3.3.1

Inability to Change

Lacking capabilities and resources: Most companies underestimate how long it takes to build new or alter existing capabilities. Lacking organizational capabilities and resources and lacking individual-level skills like technology and process knowledge are typical traps within the digital transformation journey of organizations that can lead to an inability to change (Hille et al., 2023; Vogelsang et al., 2019). For instance, a lack of resource capacity to deal with data by relevant actors can represent a major bottleneck. Regarding the individual level, factors comprising human capital, such as the number of employees or the presence of specialized personnel, are particularly important for changes that embed a high degree of new knowledge. In the case of big data, for example, the adequacy of IT-literate personnel and the availability of data science expertise are essential success factors (D’Ignazio and Bhargava, 2021). Lacking support from management: Top management support is essential throughout the whole transformation process since digital transformation strategies affect the entire organization (Matt et al., 2015). However, lacking management support is one of the most frequently experienced barriers in digital transformation initiatives (Nambisan et al., 2019). Digital innovation activities, for instance, are often very difficult to steer and evaluate as innovation outcomes require new ways of measuring (Haskamp, et al., 2021b). Hence, those activities rely particularly on top management support to create legitimacy and provide sufficient financial resources. Further, lacking management support can have negative spillover effects on the reaction of employees and other stakeholders to change initiatives. Unfortunately, this trap is reinforced by existing structures and traditional ways of measuring performance and success. Hence, it is particularly challenging to execute a digital roadmap if the top management does not fully support the initiatives. While dedicated roles like the chief digital officer can counteract these mechanisms, in practice, only a very small fraction of organizations have engaged a chief digital officer to support their transformations. Lacking support from employees: When having recognized the need to transform, many organizations try to force it to happen by pushing for top-down compliance to new processes and systems. Often, a lack of enthusiasm and evolving skepticism about the necessity to digitize can be identified as barriers resulting from insensitivity. Given that digital transformation is a challenging organizational task affecting the very core and almost every aspect of an organization, lacking support from employees can be fatal. Misunderstanding changes/miscommunication: Another form of inability to change as a reaction trap that organizations fall into when digitally transforming is miscommunication. One of the key challenges of digital transformation is that the

Designing Innovation in the Digital Age: How to Maneuver around. . .

337

whole organization needs to be involved. In this regard, formulating a digital transformation strategy and communicating its implications across the enterprise are of utmost importance as miscommunication might lead to resistance among the employees and decreased legitimacy. When introducing new processes or technologies, many organizations make the mistake of neglecting the implications these changes can have on existing working habits. Frequently observed consequences of poorly communicated changes are “workarounds” where employees manage to remain within their old habits while conditions and structures around them have changed. Those workarounds are often inefficient and potentially result in negative spillover effects making it increasingly challenging to implement changes within the organization.

3.3.2

Unwillingness to Change

Change fatigue: Sustaining the momentum of a major transformation effort against inert forces is a major challenge. Hence, it is not surprising that organizations put tremendous resources and efforts into transformation initiatives. However, employees can easily become overwhelmed by the scale and complexity of the transformation and the strain the change imposes. This makes change fatigue a major threat to success (Vial, 2019). Change fatigue often evolves slowly as teams are worn down by process changes, new learning curves, and workflow disruptions. In practice, visible symptoms of change fatigue include employee apathy, confusion, frustration, and stress (Bernerth et al., 2011). Rejection and fear of change: Change is an emotional and diffuse issue. One of the major reaction traps implies fears and acceptance problems of individuals—like skepticism or missing acceptance (Vogelsang et al., 2019). The extent to which people are willing to change depends very much on the management and the communication of change. In the context of organizational transformation, fear of change can be linked to negative anticipations of consequences around the changes. For instance, employees might fear changing how they typically do things or fear becoming irrelevant, losing power, or becoming lower in status within the organization. Fearing change is a common and very natural tendency that, under certain circumstances, leads to rejection of or even resistance to change. The latter can fracture teams and erode trust between employees and management. Along these lines, research has emphasized the role of inertia as “a form of rigidity that is determined by past experiences and actions and that manifests itself on multiple levels as part of digital transformation activities”—as a key barrier and potential trap in the context of digital transformation (Haskamp, et al., 2021a, 7).

338

3.4

C. Marx et al.

Execution Traps

Execution traps refer to the symptoms directly visible when something in the execution of the digital transformation strategy goes wrong. We have clustered those symptoms in internal symptoms, including, for instance, the perception of cannibalization effects and digital divide, and external symptoms like lacking desirability of newly developed services or products. We call these traps symptoms, as they are usually linked to a complex underlying structure of influencing factors. This makes execution traps comparatively easy to recognize in practice but also more difficult to avoid.

3.4.1

Internal Symptoms

Cannibalization effects: The perception of potential cannibalization effects, when introducing new digital products or services as part of digital transformation, is a major trap for organizations. Cannibalization effects refer to the fear that new digital services or products will undermine sales of existing core products and services. In line with this often irrational but very common inference is the aim of protecting profitability. This idea is reflected in the argument that existing products guarantee certain profits that the executive team wants to keep. In contrast to this tendency, research has shown that willingness to cannibalize might explain why certain organizations are better at developing radical innovations than other organizations (Nijssen et al., 2005). Hence, while “the extent to which a firm is prepared to reduce the actual or potential value of its investments” (Chandy & Tellis, 1998, 475) is crucial for developing new and more radically original products and services, in practice it focuses on fearing cannibalization and protecting existing profits. These effects can also be transferred to a departmental level, where departments fear losing their access to existing customers. Digitizing “garbage” processes: For many companies, digital transformation starts with digitizing analog processes. While these activities might seem less complex than other digital transformation-related initiatives, they hide a common trap. In line with the challenges of providing focus and a clear roadmap to digital transformation activities, many organizations make the mistake of transferring existing inconsistencies and flaws of analog processes to the digital world. This results in process redundancies, where processes are conducted several times within the organization with similar steps and goals without being synergized. Digital divide: Market fragmentation and ambiguity challenge the existing logic of incumbent organizations. Firms that undergo digital transformation often change existing structures, processes, roles, and even business models in order to adapt to external disruptions. These profound changes require clear priorities and dedicated top management support to succeed. However, ingroup-outgroup employee situations can arise from a lack of sensitivity about the far-reaching effects of these changes. Likewise, perceptions of interorganizational digital divide can result. The

Designing Innovation in the Digital Age: How to Maneuver around. . .

339

initial definition of digital divide refers to “the separation between those who have access to digital information and communications technology (. . .) and those who do not” (Dewan & Riggins, 2005, 289). This becomes a transformation trap: Presumed and actual digital divide can cascade into frustration and resistance on the individual level and diminish efficiency on the organizational level. Hence, recognizing and bridging the interorganizational digital divide is of utmost importance when maneuvering around digital transformation traps. Identity loss: While digital transformation involves a new organizational identity, IT-enabled transformation enhances an existing organizational identity (Wessel et al., 2021). This is a key differentiating factor. This factor can be linked to changes in the value proposition that come with digital transformation. A famous example of organizational identity change is Netflix, a former provider of rental movies that became a streaming platform. Changing organizational identity, however, can be challenging for employees and customers who are used to the status quo and fear a loss of identification with the organization. Furthermore, a strong identity that has been built up over a long period of time needs to change, and this change affects the very core of an organization. Utesheva et al. (2016) found that in order to survive digital disruptions, an ongoing strategic renegotiation of the identities of all the actors involved is required. Hence, identity renegotiations and changes must be handled very carefully. Identity loss is a dangerous trap that organizations should be aware of and ideally avoid when transforming their value propositions.

3.4.2

External Symptoms

Developing something that you can’t (feasibility): Surveying the skills of employees, a study found that 80% of employers question the adequacy of their workforce training in the context of digital transformation (Koshal & Natarajarathinam, 2019). Similarly, many organizations overestimate their readiness in terms of resources and capabilities when it comes to changing their value proposition and developing new products and services. In combination with falling for hyped technologies accelerated by the pressure to keep up with the market developments and competitors, this tendency can lead to overconfidence and unrealistic expectations, eventually threatening the success of the digital initiative. Developing something that nobody wants (desirability): Human-centricity is a crucial part of successful digital transformation (Micheli et al., 2019; de Paula et al., 2022). It has been shown that adopting Design Thinking and other human-centered approaches is directly and indirectly connected to the mechanisms of generating rapid adaptation and supporting changes in the organizational setup. These are ideal prerequisites for digital transformation (Marx, 2022). In practice, however, many organizations struggle to identify changes in customer needs caused by technological advancements and general changes in customer expectations. Fueled by hyped technologies and mechanisms, such as copying successful competitors, “developing something that nobody wants” is a dangerous trap that can cascade into transformation failure if not adequately addressed.

340

C. Marx et al.

Developing something that is not viable (viability): Many digital transformations run into difficulties because costs explode while savings or revenue growth takes longer than expected. Innovation and rapid experimentation are costly, and it is extremely challenging for organizations to balance desirability and economic viability, especially when it comes to developing new products and services. Viability in this context refers to whether the business model and the way companies build new solutions along the way of digitally transforming are profitable.

4 Application to Practice In this chapter, we aimed to identify and synthesize the potential traps organizations could fall into during their digital transformation journey. Given the high practical relevance of this endeavor, we decided to translate our theoretical trap framework into a useful tool for practitioners. Additionally, the application to practice functions as a validation of the developed framework. We urge practitioners involved in transformational activities to be aware of the traps which, if not recognized or unaddressed, can prove fatal to the transformation’s success. To use the framework, we suggest two essential steps: First, as traps are sensitive to context, practitioners should reflect on which specific digital transformation activities and explosions are relevant to their transformation journey. This adds scope and helps to focus on the most relevant traps. In a second step and based on the selected activities, practitioners should go through the potential traps and assess their relevance, likelihood, and the extent to which falling into the trap would threaten the transformation activity. Here, both the recognition of traps that the company already fell into and those that are likely to be relevant in the future should be considered. For the assessment, the various layers of identification difficulty also help to raise awareness for those difficult to identify traps. In the following, we will describe the in-depth case study that we conducted and the traps applicable to this specific case.

4.1

Adopting New Ways of Working at a Financial Service Provider

This case study deals with a “new ways of working” initiative of an operations department of an incumbent financial service provider. The department headed by a managing director has five teams with five to ten team members, each including a team lead per team. This adoption took place in the operations department, which (1) serves as a coordinating unit of activities within the company, (2) runs processwise end-to-end activities, and (3) pilots and incorporates new digital technologies within the operations of the company for other departments within the company.

Designing Innovation in the Digital Age: How to Maneuver around. . .

341

Specifically, they focused on adopting agile methods within the department as a new way of working. This agile transformation was initiated in 2020 with the managing director’s appointment, as she is deeply convinced about organizing work based on agile principles. The adoption of agile practices was understood as a means for adopting a new mindset that relies on values of human-centeredness, teamwork, iterative working, client-centeredness, open-mindedness, and openness to change. Here, agile practices referred to organizing work based on tribes, squads, and chapters, including cross-functional staffing of teams. Practically, it was about adopting new ways of working, empowering teams to make decisions, giving teams the freedom for iteration and reprioritization at the end of each sprint, and focusing on clients. Adopting agile ceremonies was a vital part of the process, including dedicated sprint planning, daily team huddles, a sprint retrospective, backlog refinements, and demos. Regarding the implementation approach, the idea of agility was embraced. According to the department: “We want to introduce agile in an agile way—this means a continuous improvement of our approach to achieving a robust agile maturity level (numerous iterations) and best-fit agility model for the team.” This meant that rather than deploying all methods and concepts fully, individuals and teams were supposed to adopt agile practices step-by-step, constantly evolving. Thus, it was acceptable for teams to “fail” as long as they were seriously committed. In terms of the implementation, the managing director decided to run all operations agile after an initial successful pilot. The team leads were to execute this and rolled out agile practices supported by a few knowledgeable people in the team. Jira was implemented and used as a tool for tracking activities and running sprints to embed agile within the organization.

4.2

Transformation Traps in the Case

As part of the introduced case study, the department encountered different transformation traps based on the introduced framework. We will shortly introduce the top 3 traps encountered and what the managing director and the team did to handle them. Fear of failure—the blind leading the blind: A critical challenge that the managing director encountered was fear of failure from the team leads. As one of the team leads explained from his perspective: “The way the managing director described it was, ‘I want to introduce this as an agile way, so you try out these tools and we gradually build from there.’ But the problem is that we’re like the blind leading the blind. I have no idea how to use these tools. I have no idea whether we are doing it right or wrong and, therefore, whether we are moving in the right direction or not.” The lack of knowledge on the side of the team leads, including the fear of failing in front of the team members, made the team leads feel insecure. The department’s leadership team tried to mitigate these feelings by providing some initial training, but there emerged an additional conflict based on different mindsets. Whereas the managing director wanted people to try things out and learn from them, the team leads were afraid to expose themselves too much to their employees.

342

C. Marx et al.

Lacking resources/capabilities—no budget for training: Contributing to the issues previously presented were some structural shortcomings and a lack of resources and capabilities. While within the department, the adoption rate and the level of expertise differed among the people involved, due to the COVID crisis, the financial service provider decided to cut training budgets immediately. Although the company ran an agile transformation hub, which included coaches to support departments going agile, it was impossible to access the funding for training. The managing director, a certified coach herself, was trying to train and share her knowledge and expertise, yet the lack of budget for external training was still an immense barrier. This amplified some of the team leaders’ skepticism. They challenged the implementation approach by asking for more training before adopting agile practices, which would have delayed the entire transformation as training budgets were frozen then. At a later time—when the budget freeze was over— intensive training was offered, but the initial skepticism was still hard to overcome. Complexity and dependencies—the others don’t run agile: Another key trap in this case that needed to be overcome was related to significantly challenging dependencies on other departments. The department’s activities and projects were deeply embedded within the overall organizational structure and processes. Depending on their role within each project, they had limited abilities to impose changes on work practices on projects that other departments led. A team lead explained the underlying problem: “Agile typically works within a project where you are all trying to achieve the same goals. If you now have a broader organization with different teams working on different goals and objectives, and they are not aligned with each other, then the sprint ceremonies and plans concept does not quite fit as well.” This led to problems in specific projects. For instance, employees needed to report progress in different formats, which led to frustration: “Currently, we have to report it in three different ways. You do a presentation, you create a storyboard, and it also is part of the audit tool set. So you end up with an audit tool set, presentations, and a storyboard. This means you’re doing about three different ways of communicating the information.” Hence, the fact that members of the department did not own the projects entirely posed a major challenge and undermined agile practices. This example shows how structural dependencies threaten the ongoing change and make it hard for organizational members of the department to fully adopt agile practices.

5 Conclusion A major challenge in today’s business world is to turn distressed digital transformation activities around. While adopting Design Thinking might be one very promising strategy to increase the chances of successfully transforming, until now little has been known about the underlying traps that can cause failure if neither identified nor addressed. Our digital transformation trap framework addresses this opportunity. Based on a literature review, an analysis of different digital transformation activities,

Designing Innovation in the Digital Age: How to Maneuver around. . .

343

and the application to practice through an in-depth case study, we identified the trap categories recognition, assessment, reaction, and execution traps each consisting of two subcategories and six to eight specific traps. Hence, in total, we could identify and describe 27 major traps. Moreover, the three layers of our trap framework account for the fact that different traps vary in difficulty to detect. The application to practice through the case study demonstrates how to use the trap framework in practice and provides real-world examples of three major trap areas that were relevant to the case. This chapter contributes to the “demystification” of failing digital transformation endeavors and can serve as a foundation for understanding how adopting Design Thinking, or other mitigation strategies, can help navigate around digital transformation traps. Regarding future research, we intend to further validate the trap framework with real-world case studies. Moreover, we plan to scale and ease the application of the framework in practice by translating it into a self-assessment tool for practitioners. Acknowledgments We appreciate the funding provided by the Hasso Plattner Foundation as part of the HPI-Stanford Design Thinking Research Program that enabled this research. We further thank our research partner from practice, as well as our collaboration partners from the MIT Center for Information Systems Research and the Haas School of Business at the University of California Berkeley.

References Appio, F. P., Frattini, F., Petruzzelli, A. M., & Neirotti, P. (2021). Digital transformation and innovation management: A synthesis of existing research and an agenda for future studies. The Journal of Product Innovation Management, 38(1), 4–20. Baiyere, A., Grover, V., Gupta, A., Woerner, S., & Lyytinen, K. J. (2017). Digital ‘X’–a new tune for IS research or old wine in new bottles? In ICIS Proceedings. Barthel, P., Fuchs, C., Birner, B., & Hess, T. (2020a). Embedding digital innovations in organizations: A typology for digital innovation units. In WI Proceedings. Barthel, P., Stark, N., & Hess, T. (2020b). Exploring new areas for project portfolio managementevolving practices for digital transformation projects. In ECIS Proceedings. Bernerth, J. B., Jack Walker, H., & Harris, S. G. (2011). Change fatigue: Development and initial validation of a new measure. Work & Stress, 25(4), 321–337. Chandy, R. K., & Tellis, G. J. (1998). Organizing for radical product innovation: The overlooked role of willingness to cannibalize. JMR, Journal of Marketing Research, 35(4), 474–487. Dewan, S., & Riggins, F. J. (2005). The digital divide: Current and future research directions. Journal of the Association for Information Systems, 6(12), 298–337. D’Ignazio, and Bhargava. (2021). Approaches to building big data literacy. Proceedings of the Bloomberg Data for. http://www.kanarinka.com/wp-content/uploads/2021/01/DIgnazio-andBhargava-Approaches-to-Building-Big-Data-Literacy.pdf DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48(2), 147–160. Dremel, C., Wulf, J., & Herterich, M. M. (2017). How AUDI AG established big data analytics in its digital transformation. MIS Quarterly Executive.

344

C. Marx et al.

Fichman, R. G., Dos Santos, B. L., & Zheng, Z. (2014). Digital innovation as a fundamental and powerful concept in the information systems curriculum. MIS Quarterly, 38(2), 329–A15. Forth, P., Reichert, T., de Laubier, R., & Chakraborty, S. (2020, October 29). Flipping the odds of digital transformation success. BCG (blog). https://www.bcg.com/de-de/publications/2020/ increasing-odds-of-success-in-digital-transformation Fuchs, C., & Hess, T. (2018). Becoming agile in the digital transformation: The process of a LargeScale agile transformation. In ICIS Proceedings. Gale, M. (2021). Why digital transformation is dead and It’s too late to cross the chasm. Leadership strategy, Forbes. https://www.forbes.com/sites/michaelgale/2021/12/14/why-digital-transfor mation-is-dead-and-its-too-late-to-cross-the-chasm/ Haskamp, T., Dremel, C., Marx, C., Rinkes, U., & Uebernickel, F. (2023). The new in the old: Managing inertia and resulting tensions in digital value creation. In Digital Value, edited by Kathryn Broman. Haskamp, T., Dremel, C., Marx, C., & Uebernickel, F. (2021a). Understanding inertia in digital transformation: A literature review and multilevel research framework. In ICIS Proceedings. Haskamp, T., Lorson, A., de Paula, D., & Uebernickel, F. (2021b). Bridging the gap–an analysis of requirements for performance measurement systems in digital innovation units. In WI Proceedings. Haskamp, T., Mayer, S., Annalena, L., & Uebernickel, F. (2021c). Performance measurement in digital innovation units–an information asymmetry perspective. In ECIS Proceedings. Heavin, C., & Power, D. J. (2018). Challenges for digital transformation–towards a conceptual decision support guide for managers. Journal of Decision Systems, 27(sup1), 38–45. Hehn, J., & Uebernickel, F. (2018). The use of design thinking for requirements engineering: An ongoing case study in the field of innovative software-intensive systems. In 2018 IEEE 26th International Requirements Engineering Conference (RE), 400–405. Hille, Z. M., Marx, C., Perscheid, M., & Uebernickel, F. (2023). Leading digital innovation units: A repertory grid study about key skills for the digital age. In HICSS Proceedings. Hinings, B., Gegenhuber, T., & Greenwood, R. (2018). Digital innovation and transformation: An institutional perspective. Information and Organization, 28(1), 52–61. Kane, G. (2019). The technology fallacy: People are the real key to digital transformation. Research-Technology Management, 62(6), 44–49. Koshal, A., & Natarajarathinam, M. (2019). Workforce training and industry 4.0 adoption in warehouses at SMEs. 2019 ASEE Annual. Magistretti, S., Pham, C. T. A., & Dell’Era, C. (2021). Enlightening the dynamic capabilities of design thinking in fostering digital transformation. Industrial Marketing Management, 97(August), 59–70. Marx, C. (2022). Design thinking for digital transformation: Reconciling theory and practice. In C. Meinel & L. Leifer (Eds.), Design thinking research: Achieving real innovation (pp. 57–77). Springer International Publishing. Marx, C., de Paula, D., & Uebernickel, F. (2021). Dynamic capabilities & digital transformation: A quantitative study on how to gain a competitive advantage in the digital age. In ECIS Proceedings. Marx, C., de Paula, D., Haskamp, T., & Uebernickel, F. (2023). A cognitive perspective on digital transformation: Literature review and research framework. In HICSS Proceedings. Marx, C., & Uebernickel, F. 2022. How to turn around: Escalation of commitment in the context of ISD project distress. In ICIS Proceedings. Matt, C., Hess, T., & Benlian, A. (2015). Digital transformation strategies. Business & Information Systems Engineering, 57(5), 339–343. van der Meulen, N., Weill, P., & Woerner, S. L. (2020). Managing organizational explosions during digital business transformations. MIS Quarterly Executive, 19(3), 4. Micheli, P., Wilner, S. J. S., Bhatti, S. H., Mura, M., & Beverland, M. B. (2019). Doing design thinking: Conceptual review, synthesis, and research agenda. Journal of Product Innovation Management, 36(2), 124–148.

Designing Innovation in the Digital Age: How to Maneuver around. . .

345

Moore, D. A., & Healy, P. J. (2008). The trouble with overconfidence. Psychological Review, 115(2), 502–517. Nambisan, S., Wright, M., & Feldman, M. (2019). The digital transformation of innovation and entrepreneurship: Progress, challenges and key themes. Research Policy, 48(8), 103773. Nijssen, E. J., Hillebrand, B., & Vermeulen, P. A. M. (2005). Unraveling willingness to cannibalize: A closer look at the barrier to radical innovation. Technovation, 25(12), 1400–1409. Orzes, G., Rauch, E., Bednar, S., & Poklemba, R. (2018). Industry 4.0 implementation barriers in small and medium sized enterprises: A focus group study. In 2018 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM), 1348–1352. de Paula, D., Marx, C., Wolf, E., Dremel, C., Cormican, K., & Uebernickel, F. (2022). A managerial mental model to drive innovation in the context of digital transformation. Industry and Innovation, May, 1–25. Sebastian, I., Ross, J., Beath, C., Mocker, M., Moloney, K., & Fonstad, N. (2017). How big old companies navigate digital transformation. MIS Quarterly Executive. Shevtsova, H., Shvets, N., Panychok, M., & Sokolova, H. (2020). Digitalization of small and medium-sized enterprises in Ukraine. In 2020 61st International Scientific Conference on Information Technology and Management Science of Riga Technical University (ITMS), 1–5. Staw, B. M. (1981). The escalation of commitment to a course of action. Academy of Management Review. Academy of Management, 6(4), 577–587. Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509–533. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131. Utesheva, A., Simpson, J. R., & Cecez-Kecmanovic, D. (2016). Identity metamorphoses in digital disruption: A relational theory of identity. European Journal of Information Systems, 25(4), 344–363. Vetterli, C., Uebernickel, F., Brenner, W., Petrie, C., & Stermann, D. (2016). How Deutsche Bank’s IT division used design thinking to achieve customer proximity. MIS Quarterly Executive, 15(1). Vial, G. (2019). Understanding digital transformation: A review and a research agenda. The Journal of Strategic Information Systems, 28(2), 118–144. Vogelsang, K., Liere-Netheler, K., Packmohr, S., & Hoppe, U. (2019). Barriers to digital transformation in manufacturing: Development of a research agenda. In HICSS Proceedings. Wade, M., Shan, J., & International Institute for Management Development. (2020). Covid-19 has accelerated digital transformation, but may have made it harder not easier (pp. 213–220). MIS Quarterly Executive. Wang, G. (2022). Digital reframing: The design thinking of redesigning traditional products into innovative digital products. The Journal of Product Innovation Management, 39(1), 95–118. Weill, P., van der Meulen, N., & Woerner, S. L. (2018). Becoming future ready requires organizational explosions (Vol. XVIII-8). MIT Center for Information Systems Research. Weill, P., & Woerner, S. L. (2018). Is your company ready for a digital future? MIT Sloan Management Review, 59(2), 21–25. Wessel, L., Baiyere, A., Ologeanu-Taddei, R., Cha, J., & Blegind-Jensen, T. (2021). Unpacking the difference between digital transformation and IT-enabled organizational transformation. Journal of the Association for Information Systems, 22(1), 102. Woerner, S. L., Weill, P., & Sebastian, I. M. (2022). Future ready: the four pathways to capturing digital value. Harvard Business Review Press. Yin, R. K. (2011). Applications of case study research. SAGE.

Facets of Hybrid Education Selina Mayer, Martin Schwemmle, Claudia Nicolai, and Ulrich Weinberg

Abstract Hybrid educational formats are on the rise, but there exists neither definitional clarity on the term “hybrid” nor has there been a deeper examination of approaches that lend themselves to highly collaborative education, such as that required for design thinking. This book chapter first presents a taxonomy of timeand place-based hybridity. It then discusses in-depth the settings that are particularly relevant to design thinking education and presents their key opportunities and challenges. Overall, this chapter raises awareness and understanding of teaching in hybrid settings and offers researchers inspiration for future research and educators issues to consider when teaching in hybrid formats.

1 Introduction The COVID-19 pandemic disrupted education worldwide, accelerating the transformation from physical spaces to virtual settings. While information technology has influenced education already strongly in various manners over the past 30 years, many institutions still relied heavily on traditional physical classroom teaching. The “screeching halt” to on-site education that was triggered by the COVID-19 pandemic (Ellis et al., 2020, p. 559) led to the emergence of a variety of hybrid education formats. Today, educators and researchers use the term hybrid education in many ways. For some, hybrid education refers to the streaming or recording of the traditional lecture with frontal input, allowing no or only limited interactions. In the literature, hybrid learning commonly refers to the combination of online and face-to-face learning activities (Olapiriyakul & Scher, 2006). While the former are

Selina Mayer and Martin Schwemmle contributed equally to this chapter. S. Mayer (✉) · M. Schwemmle · C. Nicolai · U. Weinberg Hasso Plattner Institute, University of Potsdam, Potsdam, Germany e-mail: [email protected]; [email protected]; [email protected]; [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_16

347

348

S. Mayer et al.

flexible, so that learners can access learning material online any time they want, the latter require learners and educators to meet at the same time. These understandings of hybrid education, however, run short in the context of highly collaborative learning settings, such as in design thinking education. Design thinking is highly collaborative and based on experiential learning; subsequently, hybrid education for design thinking requires suitable formats. Whereas, for example, grammar lessons might be easily transferrable into hybrid settings via videos and exercise sheets, collaborative problem-solving asks for more advanced hybrid settings. One concept honoring this demand is hyflex education, combining hybrid (e.g., the combination of online and face-to-face learning activities) with a high flexibility (flexible choice of attendance by the learner) (Beatty, 2014). We believe that this flexibility can be defined along two dimensions, time and location, thus allowing synchronous or asynchronous learning. This means, for example, allowing some learners to remain off-site while at the same time permitting others to flexibly join in on-site. In addition, depending on the learning content or group dynamics, some elements will require attendance, while others can decide between an on-site or off-site option. . We will provide more detailed examples in Sect. 2. Providing flexible education settings is crucial for modern education systems. Mayer gives three major reasons arguing for more flexibility (Mayer, 2023). First, the numbers of students and learners are constantly growing, bringing physical spaces in universities to their limits, already today (Twigg, 2003). Second, lifelong learning is on the rise as a central concept, reflected in the adult learners who form the majority of online learners (Ke & Xie, 2009). Adult learners have to balance various tasks, such as their work and their family life with their educational tasks, demanding a higher flexibility in order to profit the most from their education (Kara et al., 2019). And last, our world is becoming strongly interconnected, allowing and demanding highly international audiences (Mittelmeier et al., 2018). While information technology makes international learning settings possible, it also creates a highly diverse learner community with different needs and preferences (Wise et al., 2012). Flexible education settings can help better integrate international learners and their demands. In this chapter, we explore the various facets of hybrid education and discuss the major challenges and opportunities in this format.

2 Forms of Asynchronicity in Hybrid Education Flexibility in education can be defined along two dimensions: location and time. Each of these two dimensions has three broad categories. For location, we distinguish between “on-site,” “hybrid,” and “remote.” “On-site” refers to rather traditional classroom education where educators and learners come to the classroom or lecture hall. “Remote” refers to complete online education, where educators and learners join from different physical locations. Hybrid combines the two mentioned

Facets of Hybrid Education

349

Table 1 Flexibility in education: nine settings

All same time (compulsory attendance)

Time slots (possibilities for joint attendance)

Not all same time (selforganized)

All on-site Traditional education 1. Traditional education with compulsory attendance.

4. Education with possibilities for joint attendance(time slots). Flexible presence, not forced attendance Different knowledge status, need ways to update 7. Self-organized education with physical presence. “Lernbüros” (learning offices) in schools In higher education, no examples to our knowledge

Location-hybrid 2. Hybrid education with compulsory attendance(time-synchronous). Mixed or separate times by location Remote and on-site group work Educator location:On-site or online 5. Hybrid education. With possibilities for joint attendance (time slots)

8. Self-organized. Hybrid education (timeasynchronous) Same as 7) No examples to our knowledge

All remote Online education 3. Online education with compulsory attendance. Live streaming Remote group work (not hybrid)

6. Online education with possibilities for joint attendance (time slots). Flexible presence, not forced attendance Different knowledge status, need ways to update, online 9. Self-organized online education. self-directed learning MOOCs Forum

options, containing elements for on-site interactions as well as purely online interactions. When speaking of time, we differentiate between “same time,” “hybrid,” and “not same time.” “Same time” means all learners learning simultaneously, which is usually the case for compulsory attendance. On the other end, in “not same time,” learners are completely flexible as to when they want to learn, which is usually the case for self-organized learning. Hybrid forms generally include formats with flexible learning times, but also some defined spots of group attendance. Together, these two dimensions with three categories create nine different learning settings. Table 1 displays all nine options. In this section, we will offer a brief overview of all nine learning settings. Setting 1: Traditional On-Site Education The traditional on-site education setting refers to all learners on-site at the same time (i.e., with compulsory attendance). This includes the classical lecture or face-to-face teamwork in the classroom. This setting still dominates today’s education system in high schools or universities. Similar to the increase of flexible work in companies

350

S. Mayer et al.

and the rise of new teaching formats, the relevance of this setting might decrease in the future. Setting 2: Hybrid Education with Compulsory Attendance (Time-Synchronous) In the hybrid classroom setting, flexibility is allowed for location, but not for time. This means, for example, that a classical lecture takes place with a portion of the learners on-site, while the lecture is also live-streamed to a remote attendance. Such models have already been applied in previous years, for example, when lecture halls were not large enough to guarantee a seat for each learner. Hybrid classrooms are now more regularly used, as they seem a good compromise between pre-COVID traditional classroom settings and the during-COVID remote classrooms. We offer more details for this setting and more interactive and collaborative learning modes in Chapter “Predicting Creativity and Innovation in Society: The Importance of Places, the Importance of Governance”. Setting 3: Online Education with Compulsory Attendance The setting describes online education with compulsory attendance that takes place fully virtually. While the educator and the learners are all in different remote locations, they meet at the same time in a virtual room. This setting allows a fast mix between live-streamed lecture elements and remote group work settings, such as in breakout rooms. This setting is becoming increasingly popular for new universities that don’t require large facilities and attract international students or those from more rural areas without a university nearby. Setting 4: On-Site Education with Possibilities for Joint Attendance (Time Slots) Setting 4 refers to education formats that center around on-site education but allow flexibility in attendance. This might include on-site teaching without compulsory attendance (and without live streaming, which distinguishes Setting 4 from Setting 2). Thus, learning material has to be provided to allow components of self-learning. This material does not have to be provided online but can also be physical reading material. This setting can also refer to on-site offers with mandatory attendance but which offer differing time slots and therefore provide some form of flexibility for learners. Another example could be exercises or colloquia, where educators offer learners rooms to clarify questions or have discussions about certain topics. Setting 5: Hybrid Education with Possibilities for Joint Attendance (Time Slots) Setting 5 includes educational formats with the maximal flexibility of time and location and thus combines the two other hybrid settings (2 and 8). Learners can choose whether to attend on-site or remotely and can flexibly choose time slots. In our experience and to guarantee the presence of students, most educational formats define specific time slots for attendance and leave other time slots open to choose for self-learning or teamwork. Generally speaking, this setting creates a high complexity for educators as they cannot plan the number of attendees and have to avoid the danger of learning deficits between different learning groups.

Facets of Hybrid Education

351

Setting 6: Online Education with Possibilities for Joint Attendance (Time Slots) This setting refers to formats where all learners are remotely located, but there are time slots for joint online attendance and time slots for self-learning. The online attendance can be compulsory or voluntary. Examples include a live-streamed lecture, which is also recorded and available for individual consumption at any point in time. It also refers to massive open online courses (MOOCs), in which the content is usually provided online via video or other learning material but where learners can meet for joint discussions or working sessions. Setting 7: Self-Organized Learning with Physical Presence Setting 7 implies full flexibility with regard to time but is combined with physical presence. To the best of our knowledge, such a format only takes place at schools in the form of “Lernbüros,” roughly translated as “learning offices.” Learners attend, as usual, their school on-site. However, they don’t have a strict timetable but can flexibly choose at what time they want to work on which subject. Educators provide learning content and discuss learners’ next steps and progress, and the learning itself can be both self-reliant or in teams. We could not find examples for higher education. This might be because physical presence might be more a question of surveillance and guidance, reasons which usually do not apply to adults. Setting 8: Self-Organized Hybrid Education (Time-Asynchronous) This setting is like Setting 7, but with an additional online component. While we can imagine that the learning offices listed under Setting 7 have a hybrid component (e.g., offer learning material online as well as on-site), we have not found examples for this in practice. Setting 9: Self-Organized Online Education Setting 9 refers to self-organized online education. Relying heavily on self-directed learning, this includes MOOCs without any joint attendance. Often, MOOCs have one dedicated “running cycle,” in which the format is run “live,” offering dedicated slots for joint attendance. After this running cycle, the MOOC is still available to learners, however without any support or supervision. While such purely selforganized online education is providing full flexibility for learners, collaboration is only possible indirectly, for example, through forums, where learners can pose questions and receive answers from other learners—but not in real time. Conclusion When it comes to collaborative learning, formats with no same-time learning opportunities are not suitable, as collaboration relies mostly on the (virtual) presence of other learners. Hence, formats 7–9 cannot be recommended for such learning formats. Therefore, hybrid forms of collaborative learning work best when all or some time slots are presence time (i.e., Settings 2 and 5).

352

S. Mayer et al.

3 Two Main Challenges and Opportunities In the following, we discuss two main challenges and opportunities of hybrid education settings. A more detailed description and the scientific approach to defining these challenges can be found in our conference paper (Mayer, 2023). We introduce the challenges and combine them with opportunities from our experience as researchers and educators.

3.1

Challenge 1: Fostering Fruitful Interactions

Facilitating interactions between the educator and learner, as well as between learners, is a crucial component of learning (Mayer et al., 2022). This is a challenge in hybrid settings, as learners are either not at the same location or not present at the same time. We differentiate two cases of this challenge and discuss the hybrid challenges for each of them.

3.1.1

Creating Personal Connectedness

This challenge lies at the core of ensuring interactions. How can educators create interactions and trust between learners in hybrid settings? Regarding space-hybrid settings, educators have two main options to handle this challenge. Either, they treat both groups separately, so that there are purely online and purely offline teams. Or, they form hybrid teams, so that learners on-site team up with remote learners. The first option implies that collaboration does not take place across media but that only virtual and only on-site learners will collaborate. This option comes with two main advantages. On the one hand, it is much easier to manage and handle for educators and learners, as those students online remain in an online space and those working on-site can work together in real teams. Presentations are streamed, but teams can independently work with the tools they prefer. On the other hand, it is also easier from a logistical and noise perspective, as teams on-site do not necessarily need screens and computers and don’t communicate via microphones and speakers. This keeps noise levels in a manageable range. At the same time, choosing this option has two main disadvantages. It prevents certain individuals from interacting and bears the danger of systemic discrimination. Remote learners, in particular, may come from international backgrounds, so that separating online and offline groups can lead to systemic biases. For example, local learners collaborate with local peers but miss out the collaboration with international remote students. To at least partly solve this issue, educators might schedule specific slots, where remote and non-remote students interact, for example, during check-in rounds in the morning or coffee table settings. The second disadvantage is that on-site students will not experience virtual collaboration, which, in light of increased

Facets of Hybrid Education

353

Fig. 1 Hyflex learning environment at the HPI School of Design Thinking (picture copyright “HPI School of Design Thinking/Jana Legler”)

global collaboration in organizations, is a skill that should be trained in higher education settings. The second option, consequently mixing teams so that remote and on-site learners are in different teams, bears several advantages—but also two main challenges. Obviously, mixed teams allow for high levels of diversity, with a greater possibility of learners coming from different countries. This added “team mixing” overcomes the challenges of participation and reduces the danger of discrimination. At the same time, students of such teams learn how to collaborate remotely and best use technology in teamwork. These skills prepare them for their future jobs, such as in international organizations. In contrast to these advantages, we see two major challenges. The first challenge is a logistical one: On-site students need technology to collaborate with their virtual peers. Each student may work in front of their own device with headphones on, which raises the question of why these students need to be in the classroom, at all. In the other case/otherwise, spaces need to be equipped with the right technology, so that students belonging to the same team can work together on-site. The team spaces at the HPI School of Design Thinking, for example, were designed for this situation. They are equipped with a conference camera that can show several people and automatically moves to the speaker. Moreover, this space has table microphones so that all on-site team members can be understood easily. Lastly, the space/area is equipped with two screens—one for seeing the remote participants and one to be used as a virtual whiteboard. Figure 1 depicts this setup. Depending on the room situation, increasing noise levels with multiple calls in parallel and learners next to each other can be a further issue here. The second challenge of such fully mixed teams is to integrate remote learners in the on-site team. In particular, activities that require physical interaction (such as prototyping) or conversations during lunch breaks or side conversations may lead to excluding remote students. A more technical solution is to use conferencing

354

S. Mayer et al.

robots, so that remote team members can “join” the coffee break. A further way of handling this challenge is to establish rules that avoid these kinds of exclusion. We will discuss this issue in more detail in the next section on providing equality. Creating connectedness in time-asynchronous settings is even harder, if not almost impossible, as learners do not necessarily have any “overlap” in terms of shared time. It is thus advantageous to at least schedule some social slots for getting to know each other or exchanging views in between. To make these social events more entertaining, educators can use networking platforms, such as Wonder, or invite their students to other virtual activities, such as a virtual exit game. A further way to establish social connections between time-asynchronous students lies in structured forms of self-presentations. For example, every learner fills out a virtual profile or creates a short video introducing themselves to the class. These videos can then be distributed via a learning management system or be sent as daily emails to the team. Furthermore, educators can support such endeavors through social networks, such as a LinkedIn group or internal forums provided by the learning management systems.

3.1.2

Ensuring Equality

From conversations with coaches, we learned that a further challenge of virtual settings is to ensure the equality between learners. We remember quite well the quote of one coach who considers full online education as making participants equal, as each learner is represented through a tile on the screen. However, this turns into a challenge for location-hybrid settings. Two main factors determine the specifics of this challenge—the location of the educator and the number of students on-site. If educators are on-site, they have to make sure to treat their remote learners equal to their on-site learners—as the latter can much more easily communicate through nonverbal communication and more easily join discussions. Educators can, for example, deliberately give space to online learners by actively addressing them. They might also mirror their experiences from the classroom to the online world, such as sharing that they sense insecurity in the classroom and ask if this holds also true for remote learners. The second factor refers to the proportion of students online and on-site, as this might determine what is considered to be the dominant classroom. If the majority of learners joins remotely, educators might automatically cater more to this audience. Ensuring equality then means to also create a meaningful educational experience for on-site learners, so that they still have benefits from being in the classroom. For both challenges, it might help to establish clear rules, such as raising real or virtual hands to ensure that everybody can join a discussion or repeating all questions or side discussions in the classroom so that remote learners can hear what is going on. Regular quick surveys about how included learners feel can give educators an indication of how well they manage the task of ensuring equality. Besides the challenge of providing equal participation in class, ensuring equality can touch further topics, of which we discuss two more points that particularly refer

Facets of Hybrid Education

355

to international learners. On the one hand, different time zones can cause inequality, so that learners from different continents might have to get up at night to join a session. While such inequalities cannot be avoided completely, educators should communicate timings upfront, should be aware of the special effort such students are willing to invest, or might even consider offering their courses at alternate times. On the other hand, the technical infrastructure in different countries differs greatly. For example, learners with low Internet bandwidth will be unable to join with their video or might have problems using tools with a high data consumption, such as Miro. The specific handling of such inequalities is a matter of the specific case, which seems to make general recommendations inappropriate. However, the first step should be to create awareness for such issues for both educators and fellow learners. Using such realities as creative constraints might be a more playful way to handle them. For example, educators could challenge their learners by simulating a server downtime of a provider, to demonstrate why one of the commonly used tools simply cannot be used anymore.

3.2

Challenge 2: Flexibility and Structure

The second challenge of hybrid education concerns learners’ flexibility of freely choosing their time and location, while at the same time structure is needed to ensure joint learning experiences. We discuss three issues of flexibility and structure in the following—creating group experiences, random encounters, and the diversity of learners.

3.2.1

Creating Group Experiences

The high flexibility of hybrid education provides each individual learner with the freedom to learn when and where they want. However, learning is far from being an individual task. In particular, more innovative learning formats that rely on group work and collaboration need some sort of presence. Moreover, if learners are completely free to choose whether they join a virtual or physical session, the educator or a few students might end up being alone in the classroom, which might cause frustration. Thus, educators should carefully balance the flexibility for their learners with the need to create a group experience. To this end, they might together with their students create rules for a learning culture, such as the default of being present, or use tools to make transparent who is there and who isn’t so that everybody can plan accordingly. This holds true for both time- and space-hybrid education. A further aspect of group experiences concerns questions of creativity and innovation. As a recent study shows, purely virtual collaboration can reduce the novelty of ideas or the behavioral collaboration itself (Balters et al., 2022). Hence,

356

S. Mayer et al.

especially for more innovative tasks, such as for the project-based learning in design thinking, planning time for physical presence might advance learning outcomes.

3.2.2

Enabling Random Encounters

Besides structured learning taking place through inputs, lectures, or group work, learning experiences also happen more randomly. For example, a learner approaches an educator during the break or after class to clarify a question. Similarly, learners have a conversation in the coffee break that raises new questions on the course topic, or they help each other in better understanding certain content. Both time- and spacehybrid education make such random encounters hard. Educators can at least compensate for such random encounters through creating fixed touchpoints. From our experience, using smaller groups, such as breakouts in video calls, tremendously helps to spur conversations. In space-hybrid settings, educators can create a setting with coffee table, where each table has a computer including some remote learners. In time-hybrid settings, creating such encounters is challenging. Without having tried out such formats on our own, we can only think of a world café-style, where students add thoughts to certain topics on a virtual whiteboard and others can “join” this discussion afterward or directly get in touch with other people whom they would like to talk to.

3.2.3

Different Life Conditions

Lastly, we want to point out that the need for flexibility and structure is quite different for learners who are in different life phases. Whereas high-school or university students might prefer to work late at night or not have lectures early in the morning, adult learners with a part-time job and a family have to schedule their week precisely in advance. They prefer flexibility to follow other obligations (such as work or family obligations), unlike students who associate flexibility with spontaneity and doing things at the spur of the moment. Planning their curricula and course formats for hybrid education, educators are well-advised to critically think about what their learners need, so as not provide flexibility where it does not make sense or is unnecessary.

4 Conclusion This chapter introduced time and location, each with three variations, as central dimensions to better understand the variety of existing hybrid education formats. In detail, we describe the flexibility of education formats across time and location with nine different settings. This structured approach offers a systematic way to evaluate

Facets of Hybrid Education

357

and compare existing education formats and thereby contributes to the fast growing discourse around hybrid (and hyflex) education. In addition, we describe two major challenges that arise from these various possibilities, specifically in hybrid education formats. By describing these challenges and the possible opportunities they bear, we hopefully enable educators and learners to (a) be aware of them and (b) to find ways to overcome these challenges. To support readers in this endeavor, we have presented examples we found in practice and describe certain methods and approaches that can give a first kickstart into successfully experience and design hybrid education formats, including asynchronous dimensions of time and location. While the limitations imposed by the COVID-19 pandemic might be reduced within the next months, we hope that the forced experiences of more hybrid education will lead to the emergence of new hybrid education experiences. Technological developments, such as augmented or virtual reality and the metaverse, will further accelerate such developments. As we emphasized in this chapter, the overarching goal should be the creation of optimal learning experiences, so that hybrid education is not technology-focused, but learner-centered. Acknowledgments The authors want to thank Dr. Sharon Nemeth for her copyediting support. A more scientific analysis of the challenges and opportunities of hybrid learning that served as an inspiration for this book chapter is described in the following conference paper: Mayer S. (2023) Understanding the Challenges and Opportunities of Hybrid Education with Location Asynchrony. Proc 56th Hawaii Int Conf Syst Sci.

References Balters, S., Miller, J. G., Li, R., Hawthorne, G. H., & Reiss, A. L. (2022). Virtual (zoom) interactions Alter behavioral cooperation, neural activation, and dyadic neural coherence. bioRxiv 2022.06.03.494713. Beatty, B. (2014). Hybrid courses with flexible participation. In L. Kyei-Blankson & EN (Eds.), Practical applications and experiences in K-20 blended learning environments. IGI global (pp. 153–177). Ellis, V., Steadman, S., & Mao, Q. (2020). ‘Come to a screeching halt’: Can change in teacher education during the COVID-19 pandemic be seen as innovation? European Journal of Teacher Education, 43, 559–572. https://doi.org/10.1080/02619768.2020.1821186 Kara, M., Erdoğdu, F., Kokoç, M., & Cagiltay, K. (2019). Challenges faced by adult learners in online distance education: A literature review. Open Prax, 11, 5. https://doi.org/10.5944/ openpraxis.11.1.929 Ke, F., & Xie, K. (2009). Toward deep learning for adult students in online courses. The Internet and Higher Education, 12, 136–145. https://doi.org/10.1016/j.iheduc.2009.08.001 Mayer, S. (2023). Understanding the challenges and opportunities of hybrid education with location asynchrony. In: Proceedings of the 56th Hawaii International Conference on System Sciences. pp 103–112. Mayer, S., Schwemmle, M., Nicolai, C., & Weinberg, U. (2022). Experiences of facilitating virtual design thinking: Theoretical reflections and practical implications. In Design thinking research (pp. 79–95). Springer.

358

S. Mayer et al.

Mittelmeier, J., Rienties, B., Tempelaar, D., Hillaire, G., & Whitelock, D. (2018). The influence of internationalised versus local content on online intercultural collaboration in groups: A randomised control trial study in a statistics course. Computers in Education, 118, 82–95. https://doi.org/10.1016/j.compedu.2017.11.003 Olapiriyakul, K., & Scher, J. M. (2006). A guide to establishing hybrid learning courses: Employing information technology to create a new learning experience, and a case study. The Internet and Higher Education, 9, 287–301. https://doi.org/10.1016/j.iheduc.2006.08.001 Twigg, C. A. (2003). Improving learning and reducing costs for online learning. Education Review, 1148–1154. https://doi.org/10.4018/978-1-60566-198-8.ch163 Wise, A. F., Perera, N., Hsiao, Y. T., Speer, J., & Marbouti, F. (2012). Microanalytic case studies of individual participation patterns in an asynchronous online discussion in an undergraduate blended course. The Internet and Higher Education, 15, 108–117. https://doi.org/10.1016/j. iheduc.2011.11.007

Design Thinking Transfer Gap: Differences Between Knowledge and Application of Design Thinking in the Organizational Environment Lena Mayer, Selina Mayer, Katharina Hölzle, Nikolaus Bönke, and Christoph Meinel

Abstract Design thinking has become a popular innovation approach in organizations globally. The request for design thinking (DT) training has increased in the past 10 years. However, the question remains if and how employees transfer their DT knowledge into their organization. In this study, we assess employees’ DT knowledge and DT application in an international company. We assume a gap between what employees know and apply, what we call the design thinking transfer gap. Furthermore, we assess employees’ perception of their own DT practice vs. the company’s DT practice. We find that on average, employees rate their knowledge of DT significantly higher than the extent of the application of DT in their work. Employees also perceive their individual practice of DT as higher than the company’s practice, indicating a potential mismatch between the organizational climate for innovation and employees’ capabilities. Our results call for further examination of the design thinking transfer gap. We discuss future research avenues and point out practical implications.

1 Introduction Whenever I’m faced with a tough business challenge, rather than trying to use some prescribed CEO logic, I tackle it as a design problem. That’s not an inborn ability, it’s a skill—OK, a mastery—learned over many years of doing.

L. Mayer · S. Mayer (✉) · C. Meinel Hasso Plattner Institute, University of Potsdam, Potsdam, Germany e-mail: [email protected]; [email protected]; offi[email protected] K. Hölzle IAT Universität Stuttgart und Fraunhofer IAO, IAT Universität Stuttgart, Stuttgart, Germany e-mail: [email protected] N. Bönke Albert-Ludwigs-Universität Freiburg, Freiburg, Germany e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 C. Meinel, L. Leifer (eds.), Design Thinking Research, Understanding Innovation, https://doi.org/10.1007/978-3-031-36103-6_17

359

360

L. Mayer et al.

—Tim Brown, August 27, 2015, in the Harvard Business Review, When Everyone Is Doing Design Thinking, Is It Still a Competitive Advantage?

Design thinking has moved from its early conceptualization of capturing and understanding designers’ cognition and work to a managerial approach for problemsolving in all fields of an organization (Brown, 2015; Martin, 2009; Badke-Schaub, 2010). Practitioners and scholars claim that design thinking (DT) is suitable when dealing with complex, uncertain environments, when solutions to a problem are unknown (see Rittel & Weber, 1973; Kröper et al., 2010; Johansson-Sköldberg et al., 2013). Especially well-established and large companies strive to integrate dynamic work modes to stay flexible and adapt quickly to fast changes in order to come up with new, innovative solutions (e.g., SAP, P&G, IBM). Design thinking has been one of the preferred instruments and processes in this respect over the last years (Schmiedgen et al., 2015). Its user-centered approach helps teams to focus on the needs and hidden motivations of users and to come up with user-centered solutions. Furthermore, its iterative and fast-paced approach focusing on trial-and-error and rapid prototyping enables companies to increase the pace of the innovation process and become overall more nimble and flexible. In the past 10 years, requests for DT and DT training in the business have substantially increased (Dorst, 2011; Seidel & Fixson, 2013). Companies are sending their staff to 1-day and multiday training to learn the DT process, the instruments, and the mindset (Royalty & Roth, 2016). The aims are manifold—better products, more creative employees, new revenue streams or business models, a better overall performance, or a more attractive and innovative company culture (Martin, 2009; Carlgren et al., 2016; Dell’Era et al., 2020; Nagaraj et al., 2020; Nakata & Hwang, 2020). Besides the mentioned positive aspirations, there are still many open questions when it comes to the actual implementation and application of DT in organizations apart from these trainings. Although many physical, digital, and hybrid learning formats exist, there is little research on the successful transfer of DT training in organizations (Schmiedgen et al., 2015). An open question is what happens to employees when they return to their organizational routine after a DT training. Looking at the training industry in general, companies invest up to $54 billion (in the USA in 2000) or 33.5€ million (in Germany in 2013) for employee training on an annual basis (Arthur Jr. et al., 2003; Seyda & Werner, 2014). Yet only ~10% of the skills, knowledge, and abilities taught in further education programs are actually transferred to the workplace the way organizations intended (Holton III et al., 2000). Therefore, in order to gain the assumed benefits from DT, organizations need to understand if and how this transfer gap is also happening for DT training. In addition, research has asked for a better understanding of which training and practices are needed to become a “design thinker” (Micheli et al., 2019; Razzouk & Shute, 2012). Investigations on how to assess the acquisition and effectiveness of DT skills are called for by Micheli et al. (2019). Researchers further suggest the investigation of contextual factors, e.g., organizational climate and culture, to better understand aspects that support and impede DT deployment in organizational environments (Beverland et al., 2016). This is in line with literature on Education

Design Thinking Transfer Gap: Differences Between Knowledge. . .

361

and Training Evaluation Research, which outlines an infinite list of factors that can affect or predict successful training transfer (e.g., Burke & Hutchins, 2007; Holton III et al., 2000). Overall, investigating the knowledge transfer gap for DT promises to bridge the educational and organizational DT literature. Organizations want to know how their practice of DT relates to what academic literature states and practitioners advise (Royalty et al., 2019). Not knowing if training leads to the envisioned application in the organization, organizations might reduce or stop their training investments (Salas et al., 2012). Organizations might even conclude that DT itself does not create the expected organizational outcomes, blurring the effectiveness of DT as an innovation approach with the success of training initiatives. More needs to be known about the actual transfer of knowledge into application after DT training. Therefore, this study investigates to what extent employees know about DT (DT knowledge) and to what extent they actually apply DT (DT application) in their organizational work. We assume a gap between employees’ DT knowledge and DT application. As this is the first study to explore this DT transfer gap, we have chosen an explorative research design to investigate to what extent DT knowledge and DT application are due to individual factors, and we additionally assess three types of employee expertise (Tenure, DT Expertise, Innovation Expertise). Finally, we compare how employees perceive their own DT practice (Employee DT Practice) vs. the company’s DT practice (Company DT Practice).

1.1

Theoretical Background and Research Questions

In order to investigate the DT transfer gap, we refer to DT literature as well as educational research. First, we cover literature on DT in organizations and education, with a focus on how DT is applied in the organizational context. We then move on to the broader educational research and embed this study in the scientific frame of training evaluation and knowledge transfer. We conclude with the derivation of all research questions.

1.1.1

Organizational DT and DT Education

DT process visualizations by institutions and agencies are numerous. For example, the Stanford d.school illustrates five hexagons, the Hasso Plattner Institute School of Design Thinking in Potsdam teaches the DT approach in six bubbles, and IDEO.org represents three DT phases (e.g., see d.school bootcamp bootleg, 2011; HPI School of Design Thinking, 2020; IDEO.org, 2015). Recently, practitioners have established a number of DT resources for application (Mayer et al., 2020; Osann et al., 2020; Lewrick et al., 2018), yet researchers still discuss the conceptualization of DT. Systematic literature reviews and case studies suggest frameworks to classify DT and attempt to come up with a mutual understanding and definition of the

362

L. Mayer et al.

phenomenon (Johansson-Sköldberg et al., 2013; Carlgren et al., 2016; Liedtka, 2017; Elsbach & Stigliani, 2018; Micheli et al., 2019; Dell’Era et al., 2020; Magistretti et al., 2021). Organizations have different manifestations of DT and ways to use it, with no unified version (Hoolohan & Browne, 2020). Schmiedgen et al. (2015) conducted a multi-method study to explore the effects that employees report in establishing the DT approach in their organization. Specifically, the outcomes of DT practice are taken into consideration. Other studies focus on the elements or challenges of DT application in organizations (Rauth et al., 2014). Based on qualitative company interviews, Carlgren et al. (2016) introduce a framework capturing a mutual understanding of DT elements across a multi-organizational sample. To date, qualitative investigations are numerous, yet quantitative investigations remain few in number—despite being called for multiple times (e.g., Elsbach & Stigliani, 2008). Quantitative studies to date focus on the impact and outcomes of DT (Jaskyte & Liedtka, 2022) and often contain student samples (Seidel & Fixson, 2013; Roth et al., 2020) or resemble practitioner reports by McKinsey and IBM (Sheppard et al., 2018; Forrester Research Inc., 2018). For example, first quantitative insights can be drawn from an IBM internal study, stating, for example, a ROI up to 301% and two times faster time-to-market for projects run with the DT approach (Forrester Research Inc., 2018). Specifically looking into the impact of DT training, Kurtmollaiev et al. (2018) investigated the effect of DT training in a quasi-experimental field study. The result showed a positive effect on teams’ dynamic capabilities and innovation outcomes (Kurtmollaiev et al., 2018). The following paragraph gives a more detailed overview of DT Education Research. After years of training and learning about DT outside the designers’ realm, in educational institutions such as the Stanford d.school (Palo Alto, USA), the HPI School of Design Thinking (Potsdam, Germany), OpenLab (Stockholm, Sweden), CUC Communication University of China (Bejing, China), and many other design thinking schools or affiliated training centers, the research literature on DT education continues to increase (e.g., Cross, 1982; Dunne & Martin, 2006; Lindberg et al., 2010; Dampérat et al., 2019; Taheri, 2022). The takeoff of DT in student-target educational systems was noticed by a business world facing complex and ambiguous problems. This led to an increased interest from companies looking for external DT training as well as in establishing internal DT capabilities for employees beyond their core design team (Dorst, 2011). Companies have adopted DT programs to train their managers to develop dynamic capabilities (Kurtmollaiev et al., 2018) or to prepare them for dealing with uncertainty (Schumacher & Mayer, 2018). Roth et al. (2020) conducted an empirical study on innovation projects run with a DT approach in a student-company setup. The authors found that DT practices increase project performance, mediated by psychological empowerment, indicating that DT education in organizational settings might have similar benefits. Consequently, the question arises as to how the gathered DT knowledge is transferred into practice. For organizations, this is a central question, as “transfer is

Design Thinking Transfer Gap: Differences Between Knowledge. . .

363

critical because without it, an organization is less likely to receive any tangible benefits from its training investments” (Salas et al., 2012, p. 77f).

1.1.2

Education and Training Evaluation Research

In the educational literature, researchers define education as “the acquisition of the art of utilization of knowledge” (Whitehead, 1968, as cited in Everwijn et al., 1993, p. 425). From this definition, it could be derived that the goal of education is not to acquire knowledge and be familiar with the topic at hand but to make the knowledge accessible in the application context. In connection with organizations, this implies that organizational learning is effective when employees apply their knowledge in their daily work. The training literature refers to transfer as “the extent to which learning during training is subsequently applied on the job or affects later job performance” (Salas et al., 2012, p. 77). Having employees apply their capabilities in the organizational context can help organizations advance and gives employees the opportunity to show their skills and secure their place in the company as valuable human capital. In the late 1950s and early 1960s, Kirkpatrick developed a model which describes training evaluation criteria in four steps: reaction, learning, behavioral, and results (Kirkpatrick, 1959a, 1959b, 1960). The popularity of the model has endured since then, and it has been used numerous times (e.g., Alliger & Janak, 1989; Bates, 2004; Kraiger & Ford, 2007). The reaction criteria refer to a trainee’s satisfaction and enjoyment. The learning criteria refer to how much a trainee has learned in terms of the gained knowledge. The behavioral criteria describe measures to assess the actual application in the trainee’s context. And the results criteria refer to how well a training is reflected concerning an impact on organizational outcomes. Following Kirkpatrick’s four training evaluation levels, we can position the DT transfer gap between the learning and behavioral criteria, as we refer to a gap between DT knowledge and DT application. This is in line with previous literature, stating that “the distinction between learning (level two) and behavior (level three) has drawn increased attention to the importance of the learning transfer process in making training truly effective” (Alliger & Janak, 1989, p. 342). Understanding the bridge from knowledge to application is therefore a crucial factor in investigating training effectiveness. In addition to understanding evaluation processes of training, researchers have investigated numerous factors that predict training success (e.g., Burke & Hutchins, 2007; Holton III et al., 2000). Commonly named factors are, for example, cognitive ability, opportunity to perform skills on the job, and supervisor support (see metaanalysis by Colquitt et al., 2000). In the following section, we derive and propose this study’s research questions building on the outlined theory from the educational and training literature and quests from DT researchers to investigate DT application in the organizational context.

364

1.2

L. Mayer et al.

DT Knowledge Vs. DT Application

In this study, we introduce the constructs DT knowledge and DT application with DT elements adapted from Carlgren et al. (2016) and Royalty and Roth (2016). Employees’ rating of how familiar they are with all eight DT elements comprises the construct DT knowledge. Employees’ rating of the extent to which they use all eight DT elements depicts the construct DT application. In line with the previously described training evaluation model by Kirkpatrick, one critical element is a gap between what people know and what they apply after a training. We therefore assume that employees have more knowledge about DT elements than what they apply in their work. The ensuing research question reads as follows: RQ1a: Do employees know (DT knowledge) more about DT elements than what they apply in their company (DT Application)? The aim of Carlgren et al.’s (2016) study is to introduce a framework of DT for theoretical and practical research avenues. Thirty-six employees across six organizations took part in their interview study. Mapping the commonalities and discrepancies across all interviews, Carlgren and colleagues introduce mutual characteristics of design thinking. The framework consists of five themes: user focus, problem framing, visualization, experimentation, and diversity. For each theme, they name a set of DT principles, practices, and techniques. With user focus, the authors refer to “empathy building, deep user understanding and user involvement” (p. 46). Problem framing explains the process of looking beyond the problem scope, challenging the problem at hand, and reframing an initial problem. Visualization describes visual representations of ideas, like mock-ups. Experimentation entails “testing and trying things out in an iterative way” (p. 47). Carlgren et al. also include divergent and convergent thinking as part of the experimentation theme. Diversity refers to the team mode of collaborating in multidisciplinary teams for a broader perspective on the topic. Royalty and Roth (2016) report on design thinking metrics for the organizational context. The authors divide them into measures for empathy, reframing, iteration, and team collaboration. Based on and adapted from Carlgren et al.’s (2016) five DT themes, and Royalty and Roth’s (2016) measure categories, we assess eight DT elements in this study: user focus, problem framing/reframing problems, field research, experimentation, visualization or prototyping, divergent thinking, convergent thinking, and diverse team collaboration.

1.3

The Expertise Factors

Additionally, we want to know if employees’ expertise influences DT application, on a global and on an item level for single DT elements. Therefore, we assess three

Design Thinking Transfer Gap: Differences Between Knowledge. . .

365

types of expertise: DT expertise, innovation expertise, and tenure. Tenure describes the number of years employees have already worked in the company, DT expertise refers to the years of employees’ DT expertise, and innovation expertise captures how many years subjects have been working on innovation topics. Razzouk and Shute (2012) introduce the factor expertise in their review on DT characteristics, processes, and differences between novice and expert design thinkers. The authors highlight that accumulated experience is instrumental in moving from a novice to an expert level. Cross (2004, p. 428) defines expertise as “the result of a dedicated application to a chosen field.” In reference to the training literature, Baldwin and Ford (1988) explain that learning and transfer is characterized by trainee characteristics (individual), work environment (organizational), and training characteristics (also see Grossman & Salas, 2011). High correlations between expertise and DT application might be explained by characteristics such as motivation (e.g., meta-analysis by Blume et al., 2010), learning confidence (longitudinal study by Warr et al., 1999), and self-efficacy (e.g., Colquitt et al., 2000; Burke & Hutchins, 2007). Considerations from DT scholars and training literature lead us to examine expertise as a possible explanatory factor. Conclusively, we pose the following question: RQ1b: What is the impact of expertise (DT expertise, innovation expertise, and tenure) on DT knowledge and DT application?

1.4

DT Practice as a Method, Process, and Mindset

As mentioned above, there are numerous studies investigating DT (Beckman & Barry, 2007; Carlgren et al., 2016; Liedtka, 2017; Micheli et al., 2019; Roth et al., 2020). Yet, they do not necessarily create an easier understanding of DT. Instead, they offer a high variation of what DT actually is—meaning it often ends up being a fuzzy concept. Studies differ from showing 10 underlying attributes and 8 essential tools and methods (Micheli et al., 2019), over 5 themes with 17 underlying principles/mindsets (Carlgren et al., 2016) to identifying 19 mindset elements (Dosi et al., 2018). Researchers have therefore started to analyze commonalities between these conceptualizations. Carlgren et al. (2016, p.40) postulate that all DT descriptions include certain process stages, a set of tools, and a way of thinking (mindset). Brenner et al. (2016) differentiate between mindset, process, and toolbox to define DT in their book chapter. For this study, we focus on DT as a concept composed of three categories: method, process, and mindset. Schmiedgen et al. (2015) propose a similar categorization for the spectrum on which practitioners perceived DT in their mixed-methods study. They introduce four categories along a continuum: from tool (box), method/process/protocol, and methodology to mindset. Based on these conceptualizations of DT, we adapt Schmiedgen et al.’s (2015) terminology and distinguish between practicing DT as a method, process, and mindset. Our definitions are as follows:

366

L. Mayer et al.

(a) Method: DT as a toolbox of independent creativity methods (e.g., use of warmups, tools like semantic analysis, POV). (b) Process: entertaining and fostering team innovation projects working along the DT process, e.g., divergent-convergent phases with extensive problem space exploration with interviews, etc. (c) Mindset: Work along the DT principles of experimentation/rapid prototyping, testing, and user focus/empathy, and encourage, facilitate, and develop teamwork. The research question resulting from this is as follows: RQ2: How do employees’ perceptions of their own vs. company DT practice vary, for (a) method, (b) process, or (c) mindset (employee DT practice vs. company DT practice)?

2 Empirical Method 2.1

Sample and Procedure

Our study was conducted at the German headquarters of a global chemical company. All study participants represent a group of employees who are highly interested in innovation topics and are on an interest group list for DT topics. For this study, we invited participants via email to a workshop titled “Learn How to Kickstart Your Design Thinking Workshops.” At the beginning, we had an internal interest list of employees who were keen to be informed about DT activities in the organization. We sent workshop invitation emails to this employee list. From those, several forwarded the email to more colleagues, creating a snowball and selfselection effect. At the end, we had a sample of around 40 interested participants. We decided to run the workshop on 2 separate dates to limit the group to max. 22 participants, due to the capacity of the workshop room. The company contact suggested setting up and collecting data in a physical workshop setting to create a win-win situation for employees and to ensure survey responses. Data collection took place at the end of each workshop. Additional data was gathered from employees who communicated their interest in the workshop but could not attend due to a scheduling conflict, illness, or other reasons. Several of these employees had asked us to run the workshop a third time or expressed a desire to participate in any future workshop activities. Hence, we also asked them to participate in the survey. The workshop consisted of 3 h of training that included a presentation of recent DT research findings, a hands-on DT team activity, and recommendations for free DT resources and tools. At the end of the workshop, employees filled in the paperpencil survey individually and submitted it to us. In total, we received 55 completed surveys. For data analysis, four people were excluded from the dataset, as they were interns and/or graduate students, who only

Design Thinking Transfer Gap: Differences Between Knowledge. . .

367

had limited access to and knowledge of the company. Thus, the study sample consists of N = 51 employees, nF = 19 female and nM = 32 male. N = 36 survey respondents had attended the workshop. The remaining respondents (n = 15) communicated their interest in the format and upcoming similar events.

2.2

Measures

The survey consisted of (a) demographic and organizational data, (b) questions assessing employees’ DT knowledge and DT application, and (c) questions comparing employees’ perception of their own vs. the company’s practice of DT as a (A) method, (B) process, or (C) mindset. All questions and definitions were discussed and finalized by four DT experts (with over 5 years of experience in DT research and practice). All questions were answered via self-report.

2.2.1

Demographic and Organizational Data

We collected the following demographic and organizational data: gender and type of employment. We asked about type of employment in a forced choice format: permanent position, fixed-term contract, or other (providing space to comment). Type of employment served to eliminate subjects on fixed-term contracts. For analysis, we only included employees with permanent positions. We also assessed employees’ tenure (how many years had they already worked in the company), DT expertise (how many years their DT expertise spanned), and innovation expertise (how many years they have been working on innovation topics). These questions served to give an overview of the employees’ experience level and how closely they are related to the DT and innovation field.

2.2.2

DT Knowledge and DT Application

The focus of this study was on the question-sets to assess employees’ DT knowledge and DT application. The questions are based on the suggested categories by Carlgren et al. (2016) and Royalty and Roth (2016): user focus, problem framing/reframing problems, field research, experimentation, visualization or prototyping, divergent thinking, convergent thinking, and diverse team collaboration. Both constructs are measured with an eight item scale. Example items are “To what extent are you familiar with the following DT elements?” (DT knowledge) and “To what extent do you use the following DT elements [your company]?” (DT application). Internal consistency was excellent for DT knowledge (α = 0.92) and good for DT application (α = 0.89). Answers were provided on a seven-point Likert scale (1: very poor or very rarely; 7: very good or all the time). We added 0: not at all as a response option for those

368

L. Mayer et al.

new to the topic or in case employees were only familiar or used certain elements and others were not known or applied.

2.2.3

Employee vs. Company DT Practice as Methods, Process, and Mindset

Subjects rated the extent to which they practiced DT as a (a) method, (b) process, and (c) mindset and how they thought the company practiced them. The intention was to analyze perceived self-rating versus perceived company rating. The DT categorizations method, process, and mindset were based on previous research by Schmiedgen et al. (2015), Carlgren et al. (2016), and Brenner et al. (2016). Internal consistencies were acceptable for the perceived employee and the perceived company items of these two three-item scales (employee: α = 0.76; company: α = 0.73).

2.3

Preliminary Analysis

We used independent t-tests to compare the mean values of the participants who finished the DT workshop with those who participated in the survey without having attended the workshop. No significant differences between the groups could be detected. This was true for all variables included in the analysis ( ps > 0.16). Nonetheless, there was a general trend that workshop participants had lower values in every DT related construct (see correlations in Table 1). Effect sizes were calculated with the Cohen’s d statistic (Cohen, 1988). All significance testing was performed at the 0.05 level. We conducted statistical analyses in the statistical software R (version 4.0.3).

3 Results 3.1

Descriptive Statistics

On average, employees worked in the company for approximately 13 years, ranging from 1 month to 35 years. The sample’s DT expertise is close to 2 years, ranging from none to 10 years of DT experience. However, subjects have worked in the innovation field for approximately 7 years on average, ranging from 0 to 24 years. Descriptive statistics and intercorrelations of all variables are displayed in Table 1. RQ1a: DT Knowledge-Application Link We conducted a paired t-test to compare the mean item scores of DT knowledge and DT application. The mean of all eight items differed significantly depending on whether participants were asked about DT knowledge or DT application

Mean 12.89 1.98 6.47 4.16 3.67 4.18 2.58

SD 8.46 2.19 6.39 1.66 1.64 1.74 1.05

2

0.53** 0.45** 0.31* 0.34* 0.07

1 0.24 0.33* 0.00 -0.02 -0.04 0.17

Note: N = 51. DT = design thinking. *p < 0.05; **p < 0.01, two-tailed tests

Variable 1. Tenure 2. DT expertise 3. Innovation expertise 4. DT knowledge 5. DT application 6. DT employee 7. DT company 0.35* 0.25 0.34* 0.08

3

Table 1 Means, standard deviations, and intercorrelations among variables in the present study

-0.13 -0.16 -0.03 0.02

4

-0.18 -0.22 -0.15 -0.07

5

0.84** 0.70** 0.35*

6

0.70** 0.41**

7

0.33*

8

Design Thinking Transfer Gap: Differences Between Knowledge. . . 369

370

L. Mayer et al.

Fig. 1 Mean of employees’ knowledge vs. application of all DT elements. Note: N = 51. Error bars represent standard errors

(t(50) = 3.71, p < 0.001, d = 0.52). This means that on average, participants rated their DT knowledge significantly higher than their DT application, DT knowledge: M = 4.16, SD = 1.66; DT application: M = 3.67, SD = 1.64 (see Fig. 1). Results further show that the employees’ knowledge of single DT elements is higher than their actual application of these aspects in the workplace (see Fig. 2). RQ1b: Impact of Expertise on DT Knowledge and DT Application The difference between DT knowledge and DT application is also present within the correlation of each construct with employees’ DT expertise. Simple linear regression analysis reveals a significant correlation between DT knowledge and DT expertise (t(47) = 3.46, p < 0.01). All eight DT knowledge items significantly depend on DT expertise (see Table 4 in Appendix). Table 2 shows that only single elements of DT application are significantly related to years of DT expertise, in particular field research, experimentation, and diverse team collaboration. This suffices for a significant correlation between mean item scores of DT application and DT expertise overall (t(47) = 2.26, p < 0.05).

Design Thinking Transfer Gap: Differences Between Knowledge. . .

371

Fig. 2 DT knowledge vs. DT application of single DT elements across all employees. Note: N = 51. Error bars represent standard errors. For a clearer overview, some DT elements as outlined in the Introduction are shortened in the visualization as follows: collaboration = diverse team collaboration, convergence = convergent thinking, divergence = divergent thinking, framing = problem framing/reframing problems, prototyping = visualization or prototyping

Table 2 Results of the simple linear regression analysis to predict the score of DT application items based on years of DT expertise (N = 49) Variable User focus Framing Field research Experimentation Prototyping Divergence Convergence Collaboration

Intercept a 4.60 3.98 1.97 2.10 3.44 2.71 2.94 3.50

Slope b -0.01 0.02 0.41 0.47 0.18 0.21 0.23 0.30

SE b 0.14 0.15 0.12 0.12 0.14 0.14 0.14 0.14

t -0.04 0.15 3.32 3.84 1.27 1.46 1.66 2.08

p 0.97 0.88 0.002 **