Pith. sign in

REVIEW 4 major objections 4 minor 29 references

When motivation can be more than a message: designing agents to boost physical activity

T0 review · 4 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A virtual exercise agent motivates users not through its message but through whether its displayed activity looks like genuine effort.

desk verdict A plausible reframing of virtual agents as co-participants, but the evidence is thin and the methods are unverifiable from the supplied text; still worth referee time as a pilot study. read the letter →

arxiv 2508.12388 v1 pith:VBLROUI3 submitted 2025-08-17 cs.HC

Alessandro Silacci , Maurizio Caon , Mauro Cherubini This is my paper · ORCID
classification cs.HC
keywords virtualagentsphysicalactivityinterventionsco-participantperceivedeffortauthenticitytrustsocialcomparisonbehaviorchange
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper asks whether a virtual agent can motivate physical activity by acting as a co-participant who visibly works alongside the user, rather than only as an instructor who delivers encouragement. Based on interviews with 12 participants from an earlier activity intervention, it reports that users treat the agent's activity levels as evidence of whether it is genuinely trying. When the agent's performance is plausible, effortful, and tied to relatable human benchmarks, users trust it and respond to it; when activity levels are ambiguous, implausible, or unexplained, trust and motivation fall. The authors conclude that perceived authenticity of effort may matter more than message content, and they offer early design directions for making an agent's exertion visible and believable.

What carries the argument

Perceived genuine effort in a social comparison context is the mechanism doing the work. The paper treats the believability of the agent's reported activity, and the visible signs of exertion attached to it, as the bridge between the agent's output and the user's trust and motivation. Its proposed design levers—behavioral cues, narrative grounding, and personalized performance—are all ways of making that perceived effort credible.

What would settle it

Run a controlled trial with two versions of the same coach agent giving identical messages, one with plausible, effortful activity (realistic step counts, pauses, fatigue cues) and one with perfect, unexplained high activity, and compare user trust and motivation over several weeks. If the two conditions produce the same outcomes, the claim that visible genuine effort drives motivation fails.

Watch

Extended reading notes

Core claim

The study's central claim is that people evaluate a virtual exercise agent by asking whether its reported performance reflects real effort, and this judgment determines whether the agent can motivate them. In social comparison, users do not just compare numbers; they assess whether the other party is really trying. Participants valued social features when they believed others, including the agent, were genuinely making an effort, and they became skeptical when activity levels seemed ambiguous or too good to be true. The paper names the resulting design problem a tension between perceived performance and authenticity, and argues that agents should be designed to be seen as exerting themselves

Load-bearing premise

The conclusion rests on twelve people's retrospective interview accounts being a faithful and unbiased record of how they actually responded to the agent, and on the coding of those interviews not imposing the reported performance–authenticity tension.

Editorial extensions

If this is right

  • Physical activity agents should surface effort and struggle, not only achievements, because visible exertion is what makes performance believable.
  • Agents that display ambiguous or implausible activity levels are likely to reduce trust and motivation, so performance numbers need to be explicable to users.
  • Grounding agent activity in relatable human benchmarks could make social comparison features more motivating.
  • Personalizing the agent's performance to the user's own level may preserve believability and prevent the agent from seeming unrelatable.
  • The co-participant role, rather than the coach role, is a viable design stance for virtual agents in behavior-change interventions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • [Editorial inference] The design directions are not yet tested; a direct experiment varying only the agent's visible effort cues would show whether perceived effort is truly the active ingredient.
  • [Editorial inference] If perceived authenticity is what drives motivation, then evaluations of such agents should measure trust and perceived genuineness, not just step counts or message recall.
  • [Editorial inference] The logic likely extends beyond exercise to any persuasive agent that reports performance—diet, sleep, or learning—where implausible reports may trigger the same skepticism.
  • [Editorial inference] Engineering visible effort raises an ethical question the paper leaves open: making an agent look like it is struggling is manufactured authenticity, and designers would need to decide whether that is honest representation or persuasion by illusion.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. This paper reports a pilot qualitative study on virtual agents as co-participants in physical activity interventions. Based on thematic analysis of semi-structured interviews with 12 participants from a prior intervention, it claims a recurring tension between perceived performance and authenticity: users value social features when agents or others appear to be genuinely trying, while ambiguous or implausible activity levels reduce trust and motivation. The paper then proposes early design directions for fostering co-experienced exertion, including behavioral cues, narrative grounding, and personalized performance. Only the abstract is readable in the provided source; the full text is corrupt/unreadable and even contains a fragment of an unrelated arXiv preprint (arXiv:2508.12380v1 [math.ST]).

Significance. If the qualitative finding is substantiated, it offers a useful and falsifiable design hypothesis for socially resonant virtual agents: perceived genuine effort, anchored in believable and relatable activity, may be a key determinant of trust and motivation. The paper is explicitly exploratory, and the proposed design directions are plausible extensions of the reported participant accounts. However, the current evidentiary base is thin—retrospective self-reports from 12 participants, with no visible behavioral measures, comparison conditions, coding reliability metrics, or member checks. As it stands, the contribution is best characterized as hypothesis generation rather than validated mechanism. The paper's strength is that it articulates a concrete design-relevant tension that can be tested in future intervention studies; its weakness is that the present evidence is not sufficient to establish that tension as a stable psychological mechanism.

major comments (4)
  1. [Full text (provided source)] The full manuscript text is unreadable in the provided source: it appears as encoding corruption and contains an unrelated fragment of arXiv:2508.12380v1 [math.ST]. Consequently, the interview protocol, coding scheme, analysis steps, participant demographics, and supporting participant quotes cannot be inspected. The central claim rests entirely on this thematic analysis. Please supply a clean, readable manuscript and verify file integrity.
  2. [Abstract] The central claim (a performance–authenticity tension) is supported only by retrospective self-reports from 12 participants in a prior intervention. The abstract reports no inter-rater reliability, member checks, negative-case analysis, or triangulation with behavioral logs or system data. Without such evidence, 'ambiguous or implausible activity levels undermined trust and motivation' could reflect recall bias or demand characteristics rather than a stable mechanism. Provide coding reliability metrics and evidence that the thematic coding did not impose the reported tension.
  3. [Abstract] The sample and the prior intervention are not described: What was the prior intervention? What were the eligibility criteria for the 12 participants? How was the interview guide developed? Was thematic saturation reached? Additionally, the phrase 'ambiguous or implausible activity levels' is not operationalized. Please specify how plausibility was assessed and report how many participants expressed each theme, so readers can judge prevalence versus idiosyncratic views.
  4. [Abstract / Design directions] The paper moves from a qualitative tension to design directions (behavioral cues, narrative grounding, personalized performance) without evaluating those directions. This is an inductive leap, not a validated implication. The proposed directions should be explicitly labeled as hypotheses requiring user evaluation, and any preliminary design rationale should be separated from the empirical findings.
minor comments (4)
  1. [Abstract / Introduction] The terms 'co-participant' and 'co-experienced exertion' are used without operational definitions; define them early and clarify how they differ from existing constructs such as social presence or companionship.
  2. [Abstract / Findings] The notion of 'relatable human benchmarks' is vague. Provide concrete examples from participant accounts and explain how benchmarks were elicited or inferred.
  3. [Full text] The embedded fragment of arXiv:2508.12380v1 [math.ST] is a serious file-integrity issue. Remove it and confirm the submitted PDF matches the intended manuscript.
  4. [Limitations (if present)] The abstract and visible fragments do not mention limitations such as small sample size, retrospective nature of the data, or lack of behavioral validation. Add an explicit limitations paragraph if not already present.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is an inductive qualitative analysis whose central claim is grounded in participant interviews, not derived from fitted parameters or self-cited equations.

full rationale

The paper's central claim—that users trust and are motivated by virtual agents when they perceive genuine effort, grounded in believable activity—is presented as a thematic finding from semi-structured interviews with 12 participants. There is no derivation chain, no fitted parameter, and no mathematical prediction that reduces to its inputs by construction. The analysis is explicitly exploratory and inductive. The only potential circularity concern would be if the authors' prior intervention shaped the interview responses or if the coding imposed the performance–authenticity tension, but the abstract does not provide enough detail to exhibit any specific reduction, and per the review rules speculation about demand characteristics or researcher influence is an evidentiary limitation, not demonstrable circularity. No self-citations are invoked as load-bearing; no uniqueness theorem or imported ansatz appears; no known result is renamed. The provided full text is heavily corrupted and even contains a fragment of an unrelated arXiv preprint (arXiv:2508.12380v1 [math.ST]), making methods verification impossible, but corruption is not circularity. Given the lack of quotable evidence of any step where a prediction is equivalent to its input, the appropriate score is 0.

Assumptions & free parameters 0 free parameters · 2 assumptions · 0 invented entities

No free parameters or invented entities. The analysis rests on two qualitative assumptions: the validity of retrospective self-report and the generalizability of a 12-participant thematic analysis.

assumptions (2)
  • domain assumption Participants' retrospective self-reports accurately reflect their perceptions of agent behavior during the prior intervention
    The central findings depend on interview data; this premise is not validated by any behavioral measure.
  • domain assumption Thematic analysis of 12 interviews yields results that generalize beyond the sample
    The paper draws design implications from a small pilot sample without saturation or statistical support.

how reviews work

0 comments
Cite this review

Pith. "Pith review of When motivation can be more than a message: designing agents to boost physical activity." pith.science (2026). https://pith.science/paper/VBLROUI3

@misc{pith2026250812388,
  author       = {Pith},
  title        = {Pith review of: When motivation can be more than a message: designing agents to boost physical activity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VBLROUI3}},
  note         = {Machine review of arXiv:2508.12388}
}
read the original abstract

Virtual agents are commonly used in physical activity interventions to support behavior change, often taking the role of coaches that deliver encouragement and feedback. While effective for compliance, this role typically lacks relational depth. This pilot study explores how such agents might be perceived not just as instructors, but as co-participants: entities that appear to exert effort alongside users. Drawing on thematic analysis of semi-structured interviews with 12 participants from a prior physical activity intervention, we examine how users interpret and evaluate agent effort in social comparison contexts. Our findings reveal a recurring tension between perceived performance and authenticity. Participants valued social features when they believed others were genuinely trying. In contrast, ambiguous or implausible activity levels undermined trust and motivation. Many participants expressed skepticism toward virtual agents unless their actions reflected visible effort or were grounded in relatable human benchmarks. Based on these insights, we propose early design directions for fostering co-experienced exertion in agents, including behavioral cues, narrative grounding, and personalized performance. These insights contribute to the design of more engaging, socially resonant agents capable of supporting co-experienced physical activity.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 22 canonical work pages

  1. [1]

    Psychology & Health 13(4), 623--649 (Jul 1998)

    Bandura, A.: Health promotion from the perspective of social cognitive theory. Psychology & Health 13(4), 623--649 (Jul 1998). doi:10.1080/08870449808407422

    work page doi:10.1080/08870449808407422 1998

  2. [2]

    Patient Education and Counseling 85(1), 53--59 (Oct 2011)

    Bennett, J.K., Fuertes, J.N., Keitel, M., Phillips, R.: The role of patient attachment and working alliance on patient adherence, satisfaction, and health-related quality of life in lupus treatment. Patient Education and Counseling 85(1), 53--59 (Oct 2011). doi:10.1016/j.pec.2010.08.005

    work page doi:10.1016/j.pec.2010.08.005 2011

  3. [3]

    doi:10.1080/00048402.2024.2351208

    Berm \'u dez, J.P.: What is the Feeling of Effort About ? Australasian Journal of Philosophy 103(1), 88--105 (Jan 2025). doi:10.1080/00048402.2024.2351208

    work page arXiv 2025

  4. [4]

    Patient Education and Counseling 59(1), 21--30 (Oct 2005)

    Bickmore, T., Gruber, A., Picard, R.: Establishing the computer--patient working alliance in automated health behavior change interventions. Patient Education and Counseling 59(1), 21--30 (Oct 2005). doi:10.1016/j.pec.2004.09.008

    work page doi:10.1016/j.pec.2004.09.008 2005

  5. [5]

    ACM Transactions on Computer-Human Interaction 12(2), 293--327 (Jun 2005)

    Bickmore, T.W., Picard, R.W.: Establishing and maintaining long-term human-computer relationships. ACM Transactions on Computer-Human Interaction 12(2), 293--327 (Jun 2005). doi:10.1145/1067860.1067867

    arXiv 2005

  6. [6]

    Qualitative Research in Psychology 3(2), 77--101 (Jan 2006)

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

    doi:10.1191/1478088706qp063oa 2006

  7. [7]

    Scientific Reports 8(1), 15692 (Oct 2018)

    Chennells, M., Michael, J.: Effort and performance in a cooperative activity are boosted by perception of a partner's effort. Scientific Reports 8(1), 15692 (Oct 2018). doi:10.1038/s41598-018-34096-1

    work page doi:10.1038/s41598-018-34096-1 2018

  8. [8]

    In: Traum, D., Swartout, W., Khooshabeh, P., Kopp, S., Scherer, S., Leuski, A

    Chowanda, A., Flintham, M., Blanchfield, P., Valstar, M.: Playing with Social and Emotional Game Companions . In: Traum, D., Swartout, W., Khooshabeh, P., Kopp, S., Scherer, S., Leuski, A. (eds.) Intelligent Virtual Agents , vol. 10011, pp. 85--95. Springer International Publishing, Cham (2016). doi:10.1007/978-3-319-47665-0_8

    work page doi:10.1007/978-3-319-47665-0_8 2016

Show all 29 references
  1. [9]

    What " and

    Deci, E.L., Ryan, R.M.: The " What " and " Why " of Goal Pursuits : Human Needs and the Self-Determination of Behavior . Psychological Inquiry 11(4), 227--268 (Oct 2000). doi:10.1207/S15327965PLI1104_01

    doi

  2. [10]

    Philosophies 5(4), 37 (Nov 2020)

    Egri-Nagy , A., T \"o rm \"a nen, A.: The Game Is Not over Yet --- Go in the Post-AlphaGo Era . Philosophies 5(4), 37 (Nov 2020). doi:10.3390/philosophies5040037

    2020 doi

  3. [11]

    In: Proceedings of the 2018 Annual Symposium on Computer-Human Interaction in Play

    Emmerich, K., Ring, P., Masuch, M.: I'm Glad You Are on My Side : How to Design Compelling Game Companions . In: Proceedings of the 2018 Annual Symposium on Computer-Human Interaction in Play . pp. 141--152. CHI PLAY '18, Association for Computing Machinery, New York, NY, USA ...

    2018

  4. [12]

    doi:10.1111/j.1751-9004.2011.00409.x

    Estabrooks, P.A., Harden, S.M., Burke, S.M.: Group Dynamics in Physical Activity Promotion : What works? Social and Personality Psychology Compass 6(1), 18--40 (2012). doi:10.1111/j.1751-9004.2011.00409.x

    2012

  5. [13]

    Methods of Information in Medicine 58(01), 009--023 (Jun 2019)

    Fadhil, A., Wang, Y., Reiterer, H.: Assistive Conversational Agent for Health Coaching : A Validation Study . Methods of Information in Medicine 58(01), 009--023 (Jun 2019). doi:10.1055/s-0039-1688757

    2019 doi

  6. [14]

    Griffiths, S., Alpay, T., Sutherland, A., Kerzel, M., Eppe, M., Strahl, E., Wermter, S.: Exercise with Social Robots : Companion or Coach ? (2018)

    2018

  7. [15]

    In: Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems

    J \"o rke, M., Sapkota, S., Warkenthien, L., Vainio, N., Schmiedmayer, P., Brunskill, E., Landay, J.A.: GPTCoach : Towards LLM-Based Physical Activity Coaching . In: Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems . pp. 1--46. ACM, Yokohama Japan (...

    2025

  8. [16]

    In: Wallner, G., Kriglstein, S., Hlavacs, H., Malaka, R., Lugmayr, A., Yang, H.S

    Kniestedt, I., G \'o mez Maureira, M.A.: Little Fitness Dragon : A Gamified Activity Tracker . In: Wallner, G., Kriglstein, S., Hlavacs, H., Malaka, R., Lugmayr, A., Yang, H.S. (eds.) Entertainment Computing - ICEC 2016. pp. 205--210. Lecture Notes in Computer Science , Spring...

    2016 doi

  9. [17]

    Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 2(2), 70:1--70:26 (Jul 2018)

    Kocielnik, R., Xiao, L., Avrahami, D., Hsieh, G.: Reflection Companion : A Conversational System for Engaging Users in Reflection on Physical Activity . Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 2(2), 70:1--70:26 (Jul 2018). doi:10.114...

    2018 doi

  10. [18]

    In: Dourish, P., Friday, A

    Lin, J.J., Mamykina, L., Lindtner, S., Delajoux, G., Strub, H.B.: Fish'n' Steps : Encouraging Physical Activity with an Interactive Computer Game . In: Dourish, P., Friday, A. (eds.) UbiComp 2006: Ubiquitous Computing . pp. 261--278. Lecture Notes in Computer Science , Springe...

    2006 doi

  11. [19]

    JMIR mHealth and uHealth 8(7), e17558 (Jul 2020)

    Maher, C.A., Davis, C.R., Curtis, R.G., Short, C.E., Murphy, K.J.: A Physical Activity and Diet Program Delivered by Artificially Intelligent Virtual Health Coach : Proof-of-Concept Study . JMIR mHealth and uHealth 8(7), e17558 (Jul 2020). doi:10.2196/17558

    2020 doi

  12. [20]

    ACM Trans

    Mohan, S., Venkatakrishnan, A., Hartzler, A.L.: Designing an AI Health Coach and Studying Its Utility in Promoting Regular Aerobic Exercise . ACM Trans. Interact. Intell. Syst. 10(2) (May 2020). doi:10.1145/3366501

    2020 doi

  13. [21]

    In: 2016 15th International Conference on Ubiquitous Computing and Communications and 2016 International Symposium on Cyberspace and Security ( IUCC-CSS )

    Mollee, J.S., Klein, M.C.A.: The Effectiveness of Upward and Downward Social Comparison of Physical Activity in an Online Intervention . In: 2016 15th International Conference on Ubiquitous Computing and Communications and 2016 International Symposium on Cyberspace and Securit...

    2016 doi

  14. [22]

    Proceedings of the ACM on Human-Computer Interaction 2(CSCW), 146:1--146:20 (Nov 2018)

    Ren, X., Yu, B., Lu, Y., Brombacher, A.: Exploring Cooperative Fitness Tracking to Encourage Physical Activity among Office Workers . Proceedings of the ACM on Human-Computer Interaction 2(CSCW), 146:1--146:20 (Nov 2018). doi:10.1145/3274415

    2018 doi

  15. [23]

    Multimodal Technologies and Interaction 7(10), 93 (Oct 2023)

    Salman, S., Richards, D., Dras, M.: Identifying Which Relational Cues Users Find Helpful to Allow Tailoring of e- Coach Dialogues . Multimodal Technologies and Interaction 7(10), 93 (Oct 2023). doi:10.3390/mti7100093

    2023 doi

  16. [24]

    Frontiers in Digital Health 7--2025 (2025)

    Silacci, A., Cherubini, M., Caon, M.: Navigating the design of simulated exercising peers: Insights from a participatory design study. Frontiers in Digital Health 7--2025 (2025)

    2025

  17. [25]

    Health Education & Behavior 45(1), 124--131 (Feb 2018)

    Stragier, J., Mechant, P., De Marez, L., Cardon, G.: Computer- Mediated Social Support for Physical Activity : A Content Analysis . Health Education & Behavior 45(1), 124--131 (Feb 2018). doi:10.1177/1090198117703055

    2018 doi

  18. [26]

    Interaction Studies 20(2), 234--255 (Oct 2019)

    Sz \'e kely, M., Powell, H., Vannucci, F., Rea, F., Sciutti, A., Michael, J.: The perception of a robot partner's effort elicits a sense of commitment to human-robot interaction. Interaction Studies 20(2), 234--255 (Oct 2019). doi:10.1075/is.18001.sze

    2019 doi

  19. [27]

    Scientific Reports 13(1), 2995 (Feb 2023)

    Zhou, C., Bian, Y., Zhang, S., Zhang, Z., Wang, Y., Liu, Y.J.: Exploring user experience and performance of a tedious task through human--agent relationship. Scientific Reports 13(1), 2995 (Feb 2023). doi:10.1038/s41598-023-29874-5

    2023 doi

  20. [28]

    , " * write output.state after.block = add.period write

    ENTRY address author booktitle chapter doi edition editor eid howpublished institution journal key month note number organization pages publisher school series title type url volume year label INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION in...

  21. [29]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.