Workshop on Algorithmic Behavior Change Support Nele Albers1,∗ , Amal Abdulrahman1,2 , Deborah Richards2 , Caroline Figueroa1 , Bibhas Chakraborty3,4 , Ananya Bhattacharjee5 , Linwei He6 , Mark A. Neerincx1 , Joseph Jay Williams5 , Nezih Younsi7 , Tibor Bosse8 , Annemiek Linn9 , Crystal Smit10 and Willem-Paul Brinkman1 1 Delft University of Technology, Delft, Netherlands 2 Macquarie University, Sydney, Australia 3 National University of Singapore, Singapore, Singapore 4 Duke University, Durham, NC, United States 5 University of Toronto, Toronto, ON, Canada 6 Tilburg University, Tilburg, Netherlands 7 Sorbonne University, Paris, France 8 Radboud University, Nijmegen, Netherlands 9 University of Amsterdam, Amsterdam, Netherlands 10 Erasmus University Rotterdam, Rotterdam, Netherlands Abstract To increase the effectiveness of behavior change applications, a large variety of algorithms has been developed to adapt what the applications offer, when, how, and with whom. Given the multitude of challenges related to the concept of algorithmic behavior change support, its development, evaluation, and impact on behavior change, this workshop aims to strengthen the community of people with diverse backgrounds (e.g., computer science, psychology, human-computer interaction) and roles in behavior change support (e.g., researcher, designer, practitioner). Combining keynotes of leading researchers with sessions in which individual workshop participants present their work and discuss problems with the audience, the workshop encouraged a lively exchange of ideas that benefits current and future research on algorithmic behavior change support. Keywords Behavior change support systems, Persuasion, Persuasive technology, Algorithmic acceptance 1. Introduction Health, sustainability, education - being potentially easy to use, available at all times, scalable, and cost-effective, behavior change applications have a large potential and have thus been developed for diverse domains (e.g., [1, 2, 3]). However, despite their potential, users commonly In: Kiemute Oyibo, Wenzhen Xu, Elena Vlahu-Gjorgievska (eds.): The Adjunct Proceedings of the 19𝑡ℎ International Conference on Persuasive Technology, April 10, 2024, Wollongong, Australia ∗ Corresponding author. Envelope-Open N.Albers@tudelft.nl (N. Albers); W.P.Brinkman@tudelft.nl (W. Brinkman) Orcid 0000-0002-0502-6176 (N. Albers); 0000-0001-5360-0833 (A. Abdulrahman); 0000-0002-7363-1511 (D. Richards); 0000-0003-0692-2244 (C. Figueroa); 0000-0002-7366-0478 (B. Chakraborty); 0000-0002-9116-3766 (A. Bhattacharjee); 0000-0002-6593-1661 (L. He); 0000-0002-8161-5722 (M. A. Neerincx); 0000-0002-9122-5242 (J. J. Williams); 0000-0003-4233-0406 (T. Bosse); 0000-0003-0912-3712 (A. Linn); 0000-0001-8485-7092 (W. Brinkman) © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR ceur-ws.org Workshop ISSN 1613-0073 Proceedings do not adhere to these applications or abandon them entirely [4, 5, 6]. Thus, there appears to be a mismatch between what the applications offer and what users need. To create behavior change applications that do meet users’ needs, a variety of algorithms has been developed to adapt what the applications offer, when, how, and with whom. For example, Zhou et al. [7], Adams et al. [8], and Costa et al. [9] have adapted daily step goals and the choice of activities for elderly people, Wang et al. [10] and Trella et al. [11] have developed algorithms for adapting the timing of notifications for physical activity and oral self-care, Kaptein et al. [12] and Bertolotti et al. [13] have adapted the use of Cialdini’s persuasion principles and different message frames, and Piette et al. [14] and Forman et al. [15] have optimized the addition of human support to eHealth applications for pain management and weight loss. These algorithms thereby consider various elements from dynamic factors (e.g., user states derived from the COM-B model [16], self-efficacy [13]) to more stable user characteristics (e.g., personality and gender [17]) and employ various algorithmic techniques from reinforcement learning (e.g., [10]) to recommender systems (e.g., [18]). Placing the human at the center and combining the strengths of humans and technology (i.e., augmented or hybrid intelligence) is often an important design guideline, as is accounting for ethical and societal values. This workshop brings together researchers, designers, developers, practitioners, and edu- cators who are interested in the concept, development, evaluation, and impact of algorithmic behavior change support. We explicitly invite participants from various backgrounds such as artificial intelligence, human-computer interaction, psychology, medical practice, and ethics of technology to contribute their perspectives and experiences. The broader objective of this workshop is to strengthen the community of people working on adaptive support in behavior change applications. To this end, the workshop aims to create a lively exchange of ideas that benefits the individual workshop participants’ current and future research. Specifically, the workshop’s aim is to a) learn about each other’s work, b) jointly work on problems of the workshop participants, and c) establish a vision for future work on algorithmic behavior change support. 2. Organization This was a one-day workshop that was primarily designed as an in-person event. It combined sessions where participants presented their work and worked together on their individual problems (e.g., provide feedback on an idea for an algorithm, serve as a focus group, or brainstorm on a research question) with keynotes by leading researchers. 2.1. Accepted Papers Researchers interested in attending the workshop were asked to submit a short position paper of their work as well as a problem that they would like to work on together with the other workshop participants. We received ten submissions by authors from diverse countries such as Canada, India, and Australia. Each submission was reviewed by at least two reviewers and the authors of four accepted papers presented their work and problem at the workshop. The accepted papers broadly covered the personalization of eHealth applications for different types of behavior change (e.g., sleep adherence, stress management, diabetes lifestyle change) and algorithm design guidelines. 2.2. Keynotes The workshop featured two keynotes, one online and one in-person: • Nina Deliu is an Assistant Professor in Statistics at Sapienza University of Rome, Italy, and a long-term visitor of the MRC–Biostatistics Unit of the University of Cambridge, UK. Her research focuses on exploring and developing reinforcement learning and multi- armed bandit algorithms for applications in behavioral sciences (e.g., education, mobile health, clinical trials). Moreover, she investigates how we can perform valid inference based on data that has been adaptively collected in such settings. • Deborah Richards is a Professor with the School of Computing in the Faculty of Science and Engineering at Macquarie University in Sydney, Australia. She is an expert on intelligent virtual agents, which she has developed for various domains including health and education. 2.3. Organizing Committee The workshop was organized by a multidisciplinary team of researchers from several institutes, with expertise in human-computer interaction, artificial intelligence, health communication, statistics, and behavioral science: • Nele Albers is a PhD student in Computer Science at Delft University of Technology. She studies how adaptive algorithms can be used to tailor behavior change applications to individuals and their state in time, especially using reinforcement learning. • Amal Abdulrahman is a research fellow at Macquarie University and Delft University of Technology. Her main research interest lies in exploring how technology can support humans in achieving a better quality of life. For this, she has developed embodied and text-based virtual agents that use techniques ranging from reinforcement learning to argumentation. • Deborah Richards did, besides giving a keynote, also help with organizing the workshop. Deborah is a Professor at Macquarie University. Her research focuses on intelligent systems, agent technologies, and virtual worlds to support human learning and well- being. • Caroline Figueroa’s research focuses on developing, testing, and disseminating person- alized digital health tools for behavior change, and tailoring these tools to the needs of underserved populations such as people from ethnic and racial minority backgrounds and low-income individuals. • Bibhas Chakraborty is an Associate Professor at the Duke-NUS Medical School. His main research focus revolves around the development of novel statistical methods and associated study designs aimed at advancing data-driven precision health, particularly in settings with varying temporal factors. • Ananya Bhattacharjee is a Computer Science PhD student at the University of Toronto, Canada. His main research interest lies in developing and understanding technology that can help people manage their psychological well-being. To this end, he has developed several text message services, mobile applications, and websites. • Linwei He is a PhD student at Tilburg University in the Netherlands with a background in communication science and especially persuasive communication. Her research focus lies on using conversational agents to accomplish long-term health behavior change, for example in the context of quitting smoking. • Mark Neerincx is Full Professor in Human-Centered Computing at the Delft University of Technology, and Senior Research Scientist at TNO Perceptual and Cognitive Systems. He is an expert on fundamental and applied research on human-computer interaction in domains such as health, security, and defense. • Joseph Jay Williams is an Assistant Professor in Computer Science at the University of Toronto. His vision is to develop intelligent, adaptive systems that are continuously enhancing and personalizing interventions to help people in contexts such as education and mental health. • Nezih Younsi is a PhD student at Sorbonne University in France. He is interested in how algorithms such as reinforcement learning can be used to help people change their behavior, specifically in the context of healthy eating. • Tibor Bosse is Full Professor of Artificial Intelligence and Communication Science at Radboud University. His main research interest is the social interaction between humans and socially intelligent systems, applied to various domains such as social skills training and behavior change. • Annemiek Linn is an Associate Professor in Health Communication at the University of Amsterdam. Her research lies at the intersection of technology and health communication. She is specifically interested in developing healthcare technologies that place the patient at the center. • Crystal Smit is an Assistant Professor in Clinical Child and Family Studies at Eras- mus University Rotterdam. In her research, she focuses on encouraging positive health behaviors among young individuals, especially considering the impact of their social networks. • Willem-Paul Brinkman is an Associate Professor at Delft University of Technology. His research focuses on human-computer interaction, human-centered artificial intelligence, and behavior change support systems, specifically in eHealth. 3. Outcome The workshop’s goal was to enable a fruitful exchange of ideas that benefits both the current and future research of the workshop participants. More specifically, the workshop outcomes were that a) participants gained insights into current research projects that their peers from different disciplines are working on, b) participants contributed to addressing problems of other workshop participants, and c) participants established a vision for future work on algorithmic behavior change support. Acknowledgments This work is part of the multidisciplinary research project Perfect Fit, which is supported by several funders organized by the Netherlands Organization for Scientific Research (NWO), program Commit2Data - Big Data & Health (project number 628.011.211). Besides NWO, the funders include the Netherlands Organisation for Health Research and Development (ZonMw), Hartstichting, the Ministry of Health, Welfare and Sport (VWS), Health Holland, and the Netherlands eScience Center. References [1] G. Chan, A. Alslaity, J. K. Reen, S. Anukem, R. Orji, Gardenquest: Using hexad player types to design a step-based multiplayer persuasive game for motivating physical activity, in: A. Meschtscherjakov, C. J. H. Midden, J. Ham (Eds.), Persuasive Technology - 18th International Conference, PERSUASIVE 2023, Eindhoven, The Netherlands, April 19-21, 2023, Proceedings, volume 13832 of Lecture Notes in Computer Science, Springer, 2023, pp. 337–356. doi:10.1007/978- 3- 031- 30933- 5_22 . [2] A. Ciocarlan, J. Masthoff, N. Oren, Kindness makes you happy and happiness makes you healthy: Actual persuasiveness and personalisation of persuasive messages in a behaviour change intervention for wellbeing, in: A. Meschtscherjakov, C. J. H. Midden, J. Ham (Eds.), Persuasive Technology - 18th International Conference, PERSUASIVE 2023, Eindhoven, The Netherlands, April 19-21, 2023, Proceedings, volume 13832 of Lecture Notes in Computer Science, Springer, 2023, pp. 198–214. doi:10.1007/978- 3- 031- 30933- 5_13 . [3] E. Meijer, J. S. Korst, K. G. Oosting, E. Heemskerk, S. Hermsen, M. C. Willemsen, B. van den Putte, N. H. Chavannes, J. Brown, “at least someone thinks i’m doing well”: a real-world evaluation of the quit-smoking app stopcoach for lower socio-economic status smokers, Addiction Science & Clinical Practice 16 (2021). doi:10.1186/s13722- 021- 00255- 5 . [4] R. Beun, W. Brinkman, S. Fitrianie, F. Griffioen-Both, C. H. G. Horsch, J. Lancee, S. Spruit, Improving adherence in automated e-coaching - A case from insomnia therapy, in: A. Meschtscherjakov, B. E. R. de Ruyter, V. Fuchsberger, M. Murer, M. Tscheligi (Eds.), Persuasive Technology - 11th International Conference, PERSUASIVE 2016, Salzburg, Austria, April 5-7, 2016, Proceedings, volume 9638 of Lecture Notes in Computer Science, Springer, 2016, pp. 276–287. doi:10.1007/978- 3- 319- 31510- 2_24 . [5] T. Greenhalgh, J. Wherton, C. Papoutsi, J. Lynch, G. Hughes, S. Hinder, N. Fahy, R. Proc- ter, S. Shaw, et al., Beyond adoption: a new framework for theorizing and evaluating nonadoption, abandonment, and challenges to the scale-up, spread, and sustainability of health and care technologies, Journal of Medical Internet Research 19 (2017) e367. doi:10.2196/jmir.8775 . [6] S. M. Kelders, L. E. Van Zyl, G. D. Ludden, The concept and components of engagement in different domains applied to ehealth: a systematic scoping review, Frontiers in Psychology 11 (2020) 926. doi:10.3389/fpsyg.2020.00926 . [7] M. Zhou, Y. Mintz, Y. Fukuoka, K. Goldberg, E. Flowers, P. Kaminsky, A. Castillejo, A. Aswani, Personalizing mobile fitness apps using reinforcement learning, in: CEUR Workshop Proceedings, volume 2068, NIH Public Access, 2018. [8] M. A. Adams, J. F. Sallis, G. J. Norman, M. F. Hovell, E. B. Hekler, E. Perata, An adaptive physical activity intervention for overweight adults: a randomized controlled trial, PLoS ONE 8 (2013) e82901. doi:10.1371/journal.pone.0082901 . [9] Â. Costa, S. Heras, J. Palanca, J. Jordán, P. Novais, V. Julian, Using argumentation schemes for a persuasive cognitive assistant system, in: N. C. Pacheco, C. Carrascosa, N. Osman, V. J. Inglada (Eds.), Multi-Agent Systems and Agreement Technologies - 14th European Conference, EUMAS 2016, and 4th International Conference, AT 2016, Valencia, Spain, December 15-16, 2016, Revised Selected Papers, volume 10207 of Lecture Notes in Computer Science, Springer, 2016, pp. 538–546. doi:10.1007/978- 3- 319- 59294- 7_43 . [10] S. Wang, C. Zhang, B. Kröse, H. van Hoof, Optimizing adaptive notifications in mo- bile health interventions systems: reinforcement learning from a data-driven behavioral simulator, Journal of Medical Systems 45 (2021). doi:10.1007/s10916- 021- 01773- 0 . [11] A. L. Trella, K. W. Zhang, I. Nahum-Shani, V. Shetty, F. Doshi-Velez, S. A. Murphy, Reward design for an online reinforcement learning algorithm supporting oral self-care, in: B. Williams, Y. Chen, J. Neville (Eds.), Thirty-Seventh AAAI Conference on Artificial Intelligence, AAAI 2023, Thirty-Fifth Conference on Innovative Applications of Artificial Intelligence, IAAI 2023, Thirteenth Symposium on Educational Advances in Artificial Intelligence, EAAI 2023, Washington, DC, USA, February 7-14, 2023, AAAI Press, 2023, pp. 15724–15730. doi:10.1609/AAAI.V37I13.26866 . [12] M. Kaptein, P. Markopoulos, B. De Ruyter, E. Aarts, Personalizing persuasive technologies: Explicit and implicit personalization using persuasion profiles, International Journal of Human-Computer Studies 77 (2015) 38–51. doi:https://doi.org/10.1016/j.ijhcs. 2015.01.004 . [13] M. Bertolotti, V. Carfora, P. Catellani, Different frames to reduce red meat intake: The moderating role of self-efficacy, Health Communication 35 (2020) 475–482. doi:10.1080/ 10410236.2019.1567444 . [14] J. D. Piette, S. Newman, S. L. Krein, N. Marinec, J. Chen, D. A. Williams, S. N. Edmond, M. Driscoll, K. M. LaChappelle, R. D. Kerns, et al., Patient-centered pain care using artificial intelligence and mobile health tools: a randomized comparative effectiveness trial, JAMA Internal Medicine 182 (2022) 975–983. doi:10.1001/jamainternmed.2022.3178 . [15] E. M. Forman, S. G. Kerrigan, M. L. Butryn, A. S. Juarascio, S. M. Manasse, S. Ontañón, D. H. Dallal, R. J. Crochiere, D. Moskow, Can the artificial intelligence technique of reinforcement learning use continuously-monitored digital data to optimize treatment for weight loss?, Journal of Behavioral Medicine 42 (2019) 276–290. doi:10.1007/s10865- 018- 9964- 1 . [16] N. Albers, M. A. Neerincx, W.-P. Brinkman, Addressing people’s current and future states in a reinforcement learning algorithm for persuading to quit smoking and to be physically active, PLoS ONE 17 (2022) e0277295. doi:10.1371/journal.pone.0277295 . [17] R. A. de Vries, K. P. Truong, C. Zaga, J. Li, V. Evers, A word of advice: how to tailor motiva- tional text messages based on behavior change theory to personality and gender, Personal and Ubiquitous Computing 21 (2017) 675–687. doi:10.1007/S00779- 017- 1025- 1 . [18] S. Hors-Fraile, S. Malwade, F. Luna-Perejon, C. Amaya, A. Civit, F. Schneider, P. Bamidis, S. Syed-Abdul, Y.-C. Li, H. De Vries, Opening the black box: Explaining the process of basing a health recommender system on the i-change behavioral change model, IEEE Access 7 (2019) 176525–176540. doi:10.1109/ACCESS.2019.2957696 .