Home / Research / Week 12–13 Report

Week 12–13 Progress Report

Navigable outline of the INFS 603 research proposal draft: introduction, three-chapter literature review, research question, method (CSO), analysis, limitations, and references.

Introduction

  • AI embedded in health systems; older adults as priority users but underrepresented in safety and risk-mitigation design.
  • Barriers and trust tensions (Wildenbos et al.; Sakaguchi-Tang et al.; Ellis et al.); conversational AI enables adaptive, behavior-responsive influence.
  • Credibility veneer risk: institutional cues and clinical tone can relax risk checking before substantive evidence (Yeung; Faraoni; Duane).
  • Governance principles (WHO; NIST; Shneiderman) vs. gap at interaction level: how interfaces surface uncertainty, contestability, and trust reasoning from proxy audit evidence.
  • Definitions: proxy audit evidence as what a trusted assistant saves when reviewing an AI recommendation; replay formats as alternative presentations of that material.
  • Study aim: qualitative CSO-style comparison of two replay formats; reflexive thematic analysis; no “winning” format or presupposed correct trust (Lee & See lens).
  • Literature structure: four argumentative lines (institutional shortcuts → hypernudge; UI audit limits → intervention layer; cognitive forcing → seamful evidence; mechanisms → capture-and-replay proxy workflow).

Literature review

Chapter 1: Institutional trust shortcuts and calibration limits (older adults)

  • Multi-path institutional trust: Guerrero; Ganguli; Wyman — not reducible to halo-only narrative.
  • Ellis et al. (2025) pivot: institutional embedding → lower privacy concerns; caveats (literacy, familiarity, sampling); Lee & See framing of calibration drift.
  • Wong et al.: endorsement cues suspend scrutiny; Yu & Chen: social influence strongest predictor; synthesis with Sakaguchi-Tang on trust–usability coupling.
  • Bounded halo: Pang; Peng; KangJie et al.; Isaksen et al. — conditional, context-sensitive reasoning.
  • Transition to Ch2: family-mediated support; adult children as informal proxy auditors; Wang et al. on mental-model gaps; escalation to adaptive manipulation.

Chapter 2: Static dark patterns to adaptive hypernudge — credibility veneer

  • Baseline: Thaler & Sunstein; Gray et al.; Brignull — static dark patterns, externally auditable.
  • Yeung: hypernudge as dynamic, personalized feedback loop; Duane: digital nudge taxonomy, dark nudges, conversational dampening of evaluation.
  • Two-axis map: governance strength × personalization/visibility; hypernudge in high-personalization, weak-governance quadrant vs. Shneiderman HCAI ideal.
  • Faraoni: dual invisibility (users + designers); UI audit structurally insufficient — demand for different intervention logic.
  • Paired conclusions: endorsement + hypernudge upgrades Ch1 shortcuts; insufficiency of UI-only audit. Bridges: Faraoni → Gullí → Ehsan; Duane → HelpCall; Lee & See → HelpCall; Shneiderman normative ceiling.

Chapter 3: HITL proxy auditor — cognitive forcing to capture-and-replay

  • Normative anchors: Lee & See (appropriate reliance); Shneiderman (high automation + high human control).
  • Gullí: cognitive forcing functions vs. Faraoni’s dynamic manipulation; vocabulary for Ehsan and HelpCall.
  • Ehsan et al.: seamful XAI; Inan et al.: positive friction.
  • Older-adult grounding: Zubatiy et al.; Brewer et al.; Baghestani et al. (proxy collaboration fragility).
  • HelpCall (Tanprasert et al.): capture-and-replay precedent; Duane contrast (replay vs. real-time hypernudge pressure).
  • Deployment boundary: Hirsch (contestability); Deng et al. (user participation in auditing/contesting).
  • Closing qual question: replay format → what participants notice, refer to, and draw on when reasoning about trust.

Conclusion of the literature review

  • Three-layer chain: Ch1 shortcuts → Ch2 hypernudge + UI-audit insufficiency → Ch3 structured human judgment + capture-and-replay.
  • Two openings for qualitative work: (1) credibility veneer in dynamic conversational deployment (limited direct HCI observation); (2) how replay format shapes noticing / referring / drawing on proxy audit evidence — underexplored.

Research question

Research question How do different replay formats shape how participants notice, refer to, and draw on proxy audit evidence when they reason about trust in an AI health recommendation during medical decision-making?
  • Qualitative; no presupposed “better” format; “notice, refer to, and draw on” as behavioral salience; “reason about trust” (not presuppose correct calibration); thick description over inferential statistics.

Method

  • Framework: Comparative Structured Observation (CSO; Mackay & McGrenere, 2025); HelpCall as practice instance; Lee & See as organizing lens without judging correct trust.
  • Design: One-factor within-participants; Format A vs. B; CSO Criteria 1–12 mapping (counterbalancing, vignettes, interviews, think-aloud, recordings, RTA Braun & Clarke).
  • Participants: N 10–14; purposive + snowball; rolling analysis from 6–8; demographics per HelpCall; Baghestani scale precedent.
  • Materials: Two replay formats; 2–3 scenario vignettes; medium-fidelity prototypes; interview guide; think-aloud during replay.
  • Procedure (60–75 min): consent → both formats + vignettes + think-aloud → comparative interview → member-check.
  • Positionality / ethics: per HelpCall; McGill REB; TCPS2 CORE (Certificate #0001446233); vignette sensitivity.

Analysis

  • Reflexive thematic analysis (Braun & Clarke, 2006, 2019); CSO-aligned.
  • Dual streams: Criterion 8 (comparative interview reflections); Criterion 9 (think-aloud + screen observation).
  • Rolling analysis; inductive + deductive coding; cross-format thick comparison without ranking trust.
  • Rigor: single-researcher RTA, journaling, member checks; numbers descriptive only (Criterion 11); design implications explicit (Criterion 12).

Limitations

  • Ecological validity: vignettes + prototypes vs. live deployment; hard to fully simulate Duane-style dynamic personalization.
  • Sample: purposive skew; proxy-auditor stance not exhaustive; focus on proxy review, not patient self-care only.
  • Claims: descriptive format effects on trust reasoning — not causal generalization or format “winner.”

References

Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa

Braun, V., & Clarke, V. (2019). Reflecting on reflexive thematic analysis. Qualitative Research in Sport, Exercise and Health, 11(4), 589–597. https://doi.org/10.1080/2159676X.2019.1628806

Brewer, R., Pierce, C., Upadhyay, P., & Park, L. (2022). An empirical study of older adults’ voice assistant use for health information seeking. ACM Transactions on Interactive Intelligent Systems, 12(2), 1–32. https://doi.org/10.1145/3484507

Brignull, H. (2023). Deceptive Patterns: Exposing the tricks tech companies use to control you (First edition: 2 January 2023). Testimonium Ltd. https://www.deceptive.design/book

Baghestani, A., Latulipe, C., & Bunt, A. (2024). Older adults’ collaborative learning dynamics when exploring feature-rich software. Proceedings of the ACM on Human-Computer Interaction, 8(CSCW1), 1–27. https://doi.org/10.1145/3637378

Duane, J.-N., Ericson, J., & McHugh, P. (2025). Digital nudges: A systematic narrative review and taxonomy. Behaviour & Information Technology, 44(13), 3250–3270. https://doi.org/10.1080/0144929X.2024.2440116

Deng, W. H., Lam, M. S., Cabrera, Á. A., Metaxa, D., Eslami, M., & Holstein, K. (2023). Supporting user engagement in testing, Auditing, and contesting AI. Computer Supported Cooperative Work and Social Computing, 556–559. https://doi.org/10.1145/3584931.3611279

Ehsan, U., Liao, Q. V., Passi, S., Riedl, M. O., & Daumé, H. (2024). Seamful XAI: Operationalizing seamful design in explainable AI. Proceedings of the ACM on Human-Computer Interaction, 8(CSCW1), 1–29. https://doi.org/10.1145/3637396

Ellis, J. R., Dellavalle, N. S., Hamer, M. K., Akerson, M., Andazola, M., Moore, A. A., Campbell, E. G., & DeCamp, M. (2025). The Halo Effect: Perceptions of information privacy among healthcare chatbot users. Journal of the American Geriatrics Society, 73(5), 1472–1483. https://doi.org/10.1111/jgs.19393

Faraoni, S. (2023). Persuasive technology and computational manipulation: Hypernudging out of mental self-determination. Frontiers in Artificial Intelligence, 6, 1216340. https://doi.org/10.3389/frai.2023.1216340

Ganguli, I., Chant, E. D., Orav, E. J., Mehrotra, A., & Ritchie, C. S. (2024). Health care contact days among older adults in traditional medicare: A cross-sectional study. Annals of Internal Medicine, 177(2), 125–133. https://doi.org/10.7326/M23-2331

Gray, C. M., Kou, Y., Battles, B., Hoggatt, J., & Toombs, A. L. (2018). The Dark (Patterns) Side of UX Design. Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems, 1–14. https://doi.org/10.1145/3173574.3174108

Guerrero, N., Mendes De Leon, C. F., Evans, D. A., & Jacobs, E. A. (2015). Determinants of trust in health care in an older population. Journal of the American Geriatrics Society, 63(3), 553–557. https://doi.org/10.1111/jgs.13316

Gullí, A. (2025). Agentic design patterns: A hands-on guide to building intelligent systems. Springer.

İnan, M., Sicilia, A., Dey, S., Dongre, V., Srinivasan, T., Thomason, J., Tür, G., Hakkani-Tür, D., & Alikhani, M. (2025). Better low than Sorry: Introducing Positive Friction for Reliable Dialogue Systems (arXiv:2501.17348). arXiv. https://doi.org/10.48550/arXiv.2501.17348

Hirsch, T., Merced, K., Narayanan, S., Imel, Z. E., & Atkins, D. C. (2017). Designing contestability: Interaction Design, machine learning, and mental health. Proceedings of the 2017 Conference on Designing Interactive Systems, 95–99. https://doi.org/10.1145/3064663.3064703

Isaksen, A. A., Schaarup, J. R., Bjerg, L., & Hulman, A. (2025). Changes in public perception of artificial intelligence in healthcare after exposure to ChatGPT. Npj Digital Medicine, 8(1), 795. https://doi.org/10.1038/s41746-025-02169-x

KangJie, E. T., Song, T., Zhu, Z., Li, J., & Lee, Y.-C. (2025). AI literacy education for older adults: Motivations, challenges and preferences (Version 1). arXiv. https://doi.org/10.48550/ARXIV.2504.14649

Lee, J. D., & See, K. A. (2004). Trust in automation: Designing for appropriate reliance. Human Factors: The Journal of the Human Factors and Ergonomics Society, 46(1), 50–80. https://doi.org/10.1518/hfes.46.1.50_30392

Ma, B., Yang, J., Wong, F. K. Y., Wong, A. K. C., Ma, T., Meng, J., Zhao, Y., Wang, Y., & Lu, Q. (2023). Artificial intelligence in elderly healthcare: A scoping review. Ageing Research Reviews, 83, 101808. https://doi.org/10.1016/j.arr.2022.101808

Mackay, W. E., & McGrenere, J. (2025). Comparative structured observation. ACM Transactions on Computer-Human Interaction, 32(2), 1–27. https://doi.org/10.1145/3711838

Tabassi, E. (2023). NIST_2023_Artificial Intelligence Risk Management Framework (AI RMF 1.0) (NIST AI 100-1; p. NIST AI 100-1). National Institute of Standards and Technology (U.S.). https://doi.org/10.6028/NIST.AI.100-1

Pang, C., Collin Wang, Z., McGrenere, J., Leung, R., Dai, J., & Moffatt, K. (2021). Technology adoption and learning preferences for older adults: Evolving perceptions, ongoing challenges, and emerging design opportunities. Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, 1–13. https://doi.org/10.1145/3411764.3445702

Peng, W., Lee, H. R., & Lim, S. (2024). Leveraging chatbots to combat health misinformation for older adults: Participatory design study. JMIR Formative Research, 8, e60712. https://doi.org/10.2196/60712

Sakaguchi-Tang, D. K., Bosold, A. L., Choi, Y. K., & Turner, A. M. (2017). Patient portal use and experience among older adults: Systematic review. JMIR Medical Informatics, 5(4), e38. https://doi.org/10.2196/medinform.8092

Shneiderman, B. (2020). Human-Centered Artificial Intelligence: reliable, safe & trustworthy. International Journal of Human–Computer Interaction, 36(6), 495–504. https://doi.org/10.1080/10447318.2020.1741118

Tanprasert, T., Dai, J., & McGrenere, J. (2024). HelpCall: Designing informal technology assistance for older adults via videoconferencing. Proceedings of the CHI Conference on Human Factors in Computing Systems, 1–23. https://doi.org/10.1145/3613904.3642938

Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving decisions about health, wealth, and happiness. Yale University Press.

Wang, X., Wang, X., Park, S., & Yao, Y. (2025). Users’ Mental models of generative AI chatbot ecosystems. Proceedings of the 30th International Conference on Intelligent User Interfaces, 1016–1031. https://doi.org/10.1145/3708359.3712125

World Health Organization. (2021). Ethics and governance of artificial intelligence for health: WHO guidance (1st ed.).

Wildenbos, G. A., Peute, L., & Jaspers, M. (2018). Aging barriers influencing mobile health usability for older adults: A literature based framework (MOLD-US). International Journal of Medical Informatics, 114, 66–75. https://doi.org/10.1016/j.ijmedinf.2018.03.012

Wong, A. K. C., Lee, J. H. T., Zhao, Y., Lu, Q., Yang, S., & Hui, V. C. C. (2025). Exploring older adults’ perspectives and acceptance of AI-driven health technologies: qualitative study. JMIR Aging, 8, e66778–e66778. https://doi.org/10.2196/66778

Wyman, M. F., Shiovitz-Ezra, S., & Bengel, J. (2018). Ageism in the health care system: providers, patients, and systems. In L. Ayalon & C. Tesch-Römer (Eds.), Contemporary Perspectives on Ageism (Vol. 19, pp. 193–212). Springer International Publishing. https://doi.org/10.1007/978-3-319-73820-8_13

Yeung, K. (2017). ‘Hypernudge’: Big Data as a mode of regulation by design. Information, Communication & Society, 20(1), 118–136. https://doi.org/10.1080/1369118X.2016.1186713

Yu, S., & Chen, T. (2024). Understanding older adults’ acceptance of chatbots in healthcare delivery: An extended UTAUT model. Frontiers in Public Health, 12, 1435329. https://doi.org/10.3389/fpubh.2024.1435329

Zubatiy, T., Mathur, N., Heck, L., Vickers, K. L., Rozga, A., & Mynatt, E. D. (2023). “I don’t know how to help with that”—Learning from limitations of modern conversational agent systems in caregiving networks. Proceedings of the ACM on Human-Computer Interaction, 7(CSCW2), 1–28. https://doi.org/10.1145/3610170