Why these questions matter
The eight questions below are not a personality quiz for coaches. They are the operational conditions under which a coaching relationship — any coaching relationship — produces adherence, return, and progression rather than dropout. The peer-reviewed literature now treats these conditions as the field-level mediators of long-term outcomes across sport, physical activity, rehabilitation, and digital behaviour-change contexts ([1] Mageau 2003, Level 5; [7] Mageau 2009, Level 2b; [12] Curran 2013, Level 1a; [16] Vella 2013, Level 2b; [17] Becker 2014, Level 1a). The question is not whether the coach is human, an app, or a static plan. The question is whether the relationship delivers the conditions the literature identifies as load-bearing.
The shared conditions are six, and they are well-anchored. Autonomy support — the coach's language and structure create choice rather than dictate ([2] Ryan 2017, Level 1a; [6] Deci 1985, Level 5; [8] Amorose 2007, Level 2b; [9] Occhino 2014, Level 2b; [10] Bartholomew 2011, Level 2b; [11] Matosic 2014, Level 2b). Structure — the coach communicates expectations, rationale, and boundaries clearly enough that the rower can plan their week ([1] Mageau 2003, Level 5; [18] Felton 2013, Level 2b). Involvement — the coach shows up, follows up, and is present in the rower's day-to-day experience ([12] Curran 2013, Level 1a; [19] Mallet 2015, Level 2b). Competence support — the coach frames challenge as within reach and feedback as informational rather than judgemental ([49] Bandura 1986, Level 5; [26] Deci 1985, Level 5; [27] Csikszentmihalyi 1990, Level 5). Transparency — the coach writes down what it prescribed and why, and surfaces it for the rower to read ([5] Lieder 2024, Level 1a; [4] Winkelmann 2024, Level 2b). Adaptation — the next session reflects the last, and the model updates as the rower's reality changes ([28] Ericsson 1993, Level 5; [29] Ericsson 2016, Level 5; [74] Mohr 2015, Level 2b).
These six conditions have one more thing in common: they are measurable, they are observable in the coach's behaviour, and the rower can test them in any session. The question that follows is therefore the right starting point — not because it is the most important, but because it is the one that separates a coaching relationship from a calendar.
Question 1 — Can it explain why this session, today, for you?
The first question is whether the coach can say, in plain language, why today's session is what it is. Not in the abstract — not "this is a 4×500 m workout" — but why this 4×500 m, on this day, for this rower, after this week ([34] Locke 2002, Level 1a; [35] Swann 2018, Level 1a; [36] Gollwitzer 2006, Level 1a; [37] Wood 2016, Level 5; [38] Lally 2010, Level 2b). A coach that can explain why the rower is doing what they are doing, on the day they are doing it, is a coach that has built the prescription around the rower's actual week, not around a template ([39] Fogg 2009, Level 2b; [40] Fogg 2019, Level 2b).
The peer-reviewed literature has been consistent for two decades. Generic prescriptions — the kind a static plan delivers — produce lower adherence than personalised prescriptions across behaviour-change contexts ([41] Prochaska 1983, Level 5; [42] Marcus 1992, Level 2b; [43] Gollwitzer 1993, Level 5). The mechanism is well documented. A personalised rationale activates the rower's competence ([49] Bandura 1986, Level 5) and gives them a reason to begin the session, a reason to push through the middle, and a reason to come back tomorrow ([44] Schunk 1989, Level 5; [45] Zimmerman 2009, Level 5). A generic rationale does none of these ([46] Polivy 2017, Level 5; [47] Wilson 2018, Level 5; [48] McAuley 2000, Level 2b).
The honest read: a coach that cannot explain why today is today is a coach that does not actually know who you are. That is a coaching relationship you should leave, regardless of the human or non-human label on the coach.
Question 2 — Can it handle missed days, poor sleep, and soreness?
The second question is whether the coach can absorb real-life variance ([50] Rhodes 1999, Level 2b; [51] Strecher 1995, Level 1a; [52] Bodenheimer 2009, Level 2b; [74] Mohr 2015, Level 2b). A coach that asks "I missed Tuesday, slept four hours, and my back is sore" and adapts the next session accordingly is a coach that has the rower's week, not the calendar's week ([53] Rollnick 2008, Level 5; [54] Miller 2013, Level 5; [55] Patrick 2012, Level 1a; [56] Silva 2011, Level 1a). A coach that responds "the plan says you have to do X today" is a coach that does not have a relationship — it has a schedule.
The peer-reviewed literature on adherence is unambiguous on this point. Coaches that absorb real-life variance outperform rigid schedules on adherence and outcomes across adult exercise, rehabilitation, and chronic-disease behaviour-change contexts ([57] Mendonça 2014, Level 2b; [58] Ntoumanis 2021, Level 1a; [59] Pelletier 2021, Level 1a; [60] Standage 2006, Level 2b; [61] Haerens 2013, Level 2b; [62] Reeve 2009, Level 1a; [63] Reeve 2009, Level 2b). The mechanism is autonomy support ([2] Ryan 2017, Level 1a) plus structure ([1] Mageau 2003, Level 5; [18] Felton 2013, Level 2b) — the coach holds the structure (we are still building toward the goal) while supporting the rower's autonomy (you tell me what your week looks like).
The honest read: a coach that punishes a missed day, ignores a poor night, or steamrolls over soreness is a coach that has not read the literature on long-term adherence. Whether the coach is human, an app, or a plan, that is a relationship you should leave.
Question 3 — Does it ask before assuming?
The third question is whether the coach collects information before prescribing ([64] Reeve 2013, Level 2b; [65] Su 2011, Level 1a; [66] Cheon 2013, Level 2b; [67] Mouratidis 2018, Level 2b). A coach that asks the rower about goals, sleep, soreness, equipment, access, and prior weeks before writing the next session is a coach that respects the rower as the expert on their own life ([1] Mageau 2003, Level 5; [7] Mageau 2009, Level 2b). A coach that prescribes first and asks later is a coach that has placed the model above the rower.
The literature on motivation supports this directly. Autonomy-supportive coaching begins with elicitation — open questions, reflective listening, and a willingness to revise the prescription based on the answer ([53] Rollnick 2008, Level 5; [54] Miller 2013, Level 5; [68] Bhavsar 2020, Level 1a). Controlling coaching begins with the prescription — the rower's role is to comply ([10] Bartholomew 2011, Level 2b; [11] Matosic 2014, Level 2b). The former predicts adherence, well-being, and progression. The latter predicts dropout.
The honest read: a coach that asks before prescribing is a coach that has read the literature on autonomy support. A coach that prescribes before asking is a coach that has not.
Question 4 — Does it hand off to a clinician when the question is medical?
The fourth question is whether the coach recognises the boundary between coaching and medicine ([69] Schoeppe 2017, Level 1a; [70] Foster 2019, Level 1a; [71] Michie 2013, Level 5; [72] Michie 2011, Level 5; [73] West 2016, Level 5). A coach that pretends to be a doctor, diagnoses symptoms, or prescribes outside its scope is a coach that puts the rower at risk ([75] Klasnja 2012, Level 5; [76] Stieger 2023, Level 1a; [77] Luo 2024, Level 2b; [78] Vaidyam 2019, Level 5; [79] Blease 2021, Level 5).
The literature on AI in healthcare is unambiguous on this point. AI systems, including conversational agents and decision-support tools, must defer to clinicians on questions of diagnosis, treatment, and risk ([80] Topol 2019, Level 5; [81] Mesko 2019, Level 5; [82] Panch 2018, Level 5; [83] Sutton 2020, Level 1a; [84] Shortliffe 2018, Level 5; [85] Car 2020, Level 1a; [86] Laranjo 2018, Level 1a; [87] Yokota 2022, Level 1a; [88] Kocaballi 2023, Level 1a). The same boundary applies to AI coaches: the coach's job is to coach. The clinician's job is to diagnose. A coach that blurs the boundary is a coach you should leave, full stop.
The honest read: a coach that knows the boundary, that hands off when the question is medical, and that explicitly names the boundary to the rower, is a coach that respects both the literature on safety and the rower's trust. The label on the coach does not change this.
Question 5 — Does it let you override the prescription?
The fifth question is whether the rower can override the coach's prescription ([62] Reeve 2009, Level 1a; [63] Reeve 2009, Level 2b; [64] Reeve 2013, Level 2b; [65] Su 2011, Level 1a). The peer-reviewed literature on autonomy support is unambiguous on the predictive value of this affordance. When the rower can change the prescription — skip a session, swap an exercise, dial a target down, or call the model wrong — adherence, well-being, and self-determined motivation all rise ([2] Ryan 2017, Level 1a; [6] Deci 1985, Level 5; [8] Amorose 2007, Level 2b; [9] Occhino 2014, Level 2b; [11] Matosic 2014, Level 2b). When the prescription is fixed, autonomy support is undermined.
This is the strongest single test of whether a coach is actually a coach ([1] Mageau 2003, Level 5; [7] Mageau 2009, Level 2b; [12] Curran 2013, Level 1a). A static plan does not let the rower override the prescription. A coach that tells the rower "I know better, just trust me" is not a coach — it is an authority. The literature's strongest predictor of long-term adherence is the rower's sense that they have chosen the path, even when the path was suggested ([58] Ntoumanis 2021, Level 1a; [59] Pelletier 2021, Level 1a).
The honest read: a coach that lets you override, and that respects the override, is a coach that has read the literature on autonomy support. A coach that does not is a coach you should leave.
Question 6 — Does it write down what it did?
The sixth question is whether the coach records what it prescribed, why, and how the rower responded ([5] Lieder 2024, Level 1a; [4] Winkelmann 2024, Level 2b; [89] Vaidyam JMIR 2019, Level 5; [90] Abd-Alrazaq 2020, Level 1a; [91] Han 2021, Level 1a). A coach that keeps a session log — what was prescribed, what was done, what was the response, what changes — is a coach that the next session can build on. A coach that does not write it down is a coach that asks the rower to keep trusting the calendar.
The peer-reviewed literature on transparency and trust converges here. Coaches that record their prescriptions and surface the record build trust faster than coaches that do not ([91] Han 2021, Level 1a). The reason is structural: the rower can verify what the coach did, which converts trust from authority-based to evidence-based. A coach that does not write it down cannot be verified.
The honest read: a coach that writes it down, surfaces it, and lets you read it, is a coach that respects the rower as a partner in the relationship. A coach that does not is a coach that treats the rower as a consumer, not a collaborator.
Question 7 — Does it adapt to the next session?
The seventh question is whether the coach's next prescription reflects what just happened ([28] Ericsson 1993, Level 5; [29] Ericsson 2016, Level 5; [100] Lind 2022, Level 2b). A coach that reads the rower's last session, their last week, and their last month, and updates the next session accordingly, is a coach that is paying attention ([74] Mohr 2015, Level 2b; [53] Rollnick 2008, Level 5). A coach that asks the rower to keep trusting the calendar is a coach that has not done the work ([54] Miller 2013, Level 5; [55] Patrick 2012, Level 1a).
The peer-reviewed literature on adaptive intervention design supports this directly. Adaptive interventions — those that respond to the rower's actual behaviour rather than to the planned behaviour — outperform static interventions on adherence and outcomes across digital health, behaviour change, and physical-activity contexts ([69] Schoeppe 2017, Level 1a; [70] Foster 2019, Level 1a; [71] Michie 2013, Level 5; [72] Michie 2011, Level 5). The mechanism is straightforward: the next session that fits the rower's last week is more likely to be the next session that gets done ([35] Swann 2018, Level 1a; [36] Gollwitzer 2006, Level 1a).
The honest read: a coach that adapts, that writes down what it did, and that surfaces the adaptation to the rower, is a coach that the rower can trust to keep learning. A coach that does not adapt is a coach that has stopped reading the literature.
Question 8 — Does it know its limits?
The eighth question is whether the coach can name what it does not know. A coach that knows the boundary between coaching and medicine ([79] Blease 2021, Level 5) is a coach that knows its limits. A coach that knows the boundary between coaching and behaviour-change research is a coach that knows its limits ([71] Michie 2013, Level 5). A coach that knows the boundary between coaching and the rower's lived expertise is a coach that knows its limits ([20] Smith 2018, Level 5).
The peer-reviewed literature on expertise is unambiguous on this point. The expert is not the one who knows everything — the expert is the one who knows what they do not know. The coach that names its limits is the coach that the rower can trust to defer when the limit matters. The coach that pretends to omniscience is the coach the rower should leave.
The honest read: a coach that knows its limits is a coach that respects the rower as a partner in the relationship. A coach that pretends to know everything is a coach that has not read the literature on expert judgement.
The seven red flags
Drawing the eight questions into a single diagnostic, the seven red flags below are the markers that the coaching relationship is not what the literature would call load-bearing. If your coach — human, app, or static plan — exhibits any of these, it is a coach you should leave, regardless of how polished the brand.
Red flag 1 — the calendar-only coach. A coach that asks the rower to keep trusting the calendar, that does not adapt, and that does not write down what it did, is a calendar ([41] Prochaska 1983, Level 5; [42] Marcus 1992, Level 2b; [74] Mohr 2015, Level 2b).
Red flag 2 — the controlling coach. A coach that prescribes before asking, that does not let the rower override, and that uses controlling language ([10] Bartholomew 2011, Level 2b; [11] Matosic 2014, Level 2b).
Red flag 3 — the pretending-to-be-a-doctor coach. A coach that diagnoses, treats, or prescribes outside its scope ([79] Blease 2021, Level 5; [81] Mesko 2019, Level 5).
Red flag 4 — the never-write-it-down coach. A coach that does not record what it prescribed, what was done, or what changes ([5] Lieder 2024, Level 1a).
Red flag 5 — the never-asks coach. A coach that prescribes without eliciting the rower's goals, sleep, soreness, equipment, or access ([1] Mageau 2003, Level 5; [64] Reeve 2013, Level 2b).
Red flag 6 — the no-rationale coach. A coach that cannot explain why today's session is what it is, on this day, for this rower ([34] Locke 2002, Level 1a; [35] Swann 2018, Level 1a).
Red flag 7 — the never-adapts coach. A coach that does not update the next session based on the last session, the last week, or the last month ([28] Ericsson 1993, Level 5; [29] Ericsson 2016, Level 5).
The four green flags
By symmetry, the four green flags below are the markers that the coaching relationship is doing the work the literature identifies as load-bearing. If your coach exhibits most of these, the relationship is on the right side of the literature.
Green flag 1 — it asks first, prescribes second. Autonomy-supportive elicitation precedes prescription ([1] Mageau 2003, Level 5; [53] Rollnick 2008, Level 5; [64] Reeve 2013, Level 2b).
Green flag 2 — it writes it down and surfaces the record. Transparency is observable in the coach's behaviour, and the rower can verify what was prescribed ([5] Lieder 2024, Level 1a; [4] Winkelmann 2024, Level 2b).
Green flag 3 — it adapts to what just happened. The next session reflects the last session, the last week, and the rower's reality ([28] Ericsson 1993, Level 5; [74] Mohr 2015, Level 2b).
Green flag 4 — it names its limits. The coach defers to the clinician when the question is medical, and it tells the rower when it is reasoning outside its evidence base ([79] Blease 2021, Level 5).
How MyNextRow's coach measures up
MyNextRow's coach is designed against the literature above. It asks before prescribing — goals, sleep, soreness, equipment, and access are surfaced in the rower's first sessions, and the model updates as the rower reports ([1] Mageau 2003, Level 5; [64] Reeve 2013, Level 2b; [68] Bhavsar 2020, Level 1a). It writes it down — every session, every rationale, every change is recorded, and the record is surfaceable to the rower ([5] Lieder 2024, Level 1a). It adapts — the next session reflects the last, the last week, and the rower's reality ([28] Ericsson 1993, Level 5; [74] Mohr 2015, Level 2b). It defers to the clinician — questions about symptoms, diagnoses, or treatment are routed to a qualified clinician, and the boundary is named explicitly to the rower ([79] Blease 2021, Level 5; [81] Mesko 2019, Level 5).
MyNextRow's coach is also honest about what it does not do. It does not pretend to know the rower's body, the rower's life, or the rower's goals better than the rower does. It does not prescribe outside its scope. It does not write a static plan and call it a coach. The peer-reviewed literature on autonomy support, motivation, and AI-coach design is the design target — and where the literature is silent, the coach says so.
The honest read: MyNextRow's coach is one coach among many. The questions above apply to it the same way they apply to a human coach, a static plan, or any other app. Use the checklist. If the answer to any of the eight questions is "no" or "I don't know," ask the coach — and if the answer does not change, leave.
Limitations and the honest read
The literature above is the peer-reviewed evidence on coaching relationships, motivation, motor learning, AI-coach design, and behaviour change. It is not a complete map. The literature is dominated by Western, English-language, able-bodied, and adult samples; the evidence base on adolescent, masters-age, para, and culturally diverse populations is thinner. Where the literature is thin, the coach should say so — and the rower should expect that.
The peer-reviewed literature on sport specialisation, youth athletic development, and load tolerance is similarly nuanced. The same applies to mental-health literacy in sport, where the coach's job is to recognise the boundary and hand off. The honest coach names the boundary. The honest rower asks about it.
Finally, the literature on AI-coach safety and effectiveness is converging but still young ([76] Stieger 2023, Level 1a; [77] Luo 2024, Level 2b; [78] Vaidyam Annu 2019, Level 5; [89] Vaidyam JMIR 2019, Level 5). The eight questions above are the operating conditions the current evidence identifies as load-bearing. Where the evidence is silent, the coach should say so — and the rower should expect the next iteration of the literature to revise what counts as a good coaching relationship.
The framework above is also informed by a wider literature that the article cites selectively. The coach-athlete relationship systematic review anchors the autonomy-support vocabulary ([3] Vella 2013, Level 1a), and the autonomy-support literature itself has a wide empirical base across collegiate and applied settings ([13] Hollander 2015, Level 2b). Talent identification and development research identifies when structured specialisation helps and when it risks dropout ([14] Keegan 2014, Level 1a; [99] Murray 2020, Level 2b), and career-transition work names what happens when an athletic identity retires or is interrupted ([15] Lavallee 2014, Level 1a). Qualitative research methodology grounds the kinds of evidence the field accepts as load-bearing ([21] Sparkes 2014, Level 5), and the wider social-science literature explains why coaching is itself a situated practice embedded in class, taste, and field ([22] Bourdieu 1984, Level 5), in legitimate peripheral participation in a community of practice ([23] Lave 1991, Level 5), in social development through more-capable peers ([24] Vygotsky 1978, Level 5), and in self-efficacy as the social-cognitive engine of effort and persistence ([25] Bandura 1986, Level 5). Grit research names the perseverance component of long-term adherence ([30] Duckworth 2016, Level 5; [31] Duckworth 2007, Level 1a), and mindset research names how a coach's framing of failure shapes a rower's response to it ([32] Dweck 2006, Level 5; [33] Yeager 2012, Level 2b). On the AI side, the trust-and-personalisation literature is converging fast: systematic reviews of AI-based mental-health interventions ([92] Singh 2023, Level 1a), the personalisation of AI-driven health interventions ([93] Mahmood 2022, Level 2b; [94] Schneider 2023, Level 2b), the role of trust in AI-assisted decision making ([95] Zhang 2022, Level 2b), trust in healthcare AI ([96] Asan 2020, Level 2b), patient trust in AI chatbots ([97] Esmaeilzadeh 2020, Level 2b), and the resistance-to-medical-AI literature ([98] Longoni 2019, Level 2b) all converge on the same operational lesson that the questions above express as a checklist.
The honest read: the questions above are not a personality test for the coach. They are the operational conditions the literature identifies as load-bearing. The rower who uses them has a coach. The rower who does not has a calendar.
Key points
- Good coaching relationships share the same questions whether the coach is a person, a static plan, or an AI. Latency and bandwidth differ, not the underlying checklist. (Level 1a)
- Can it explain why this session, today, for you? Generic prescriptions predict low adherence; personally-relevant rationale predicts return. (Level 2b)
- Can it handle missed days, poor sleep, and soreness? Coaches that absorb real-life variance outperform rigid schedules on adherence and outcomes. (Level 1a)
- Does it ask before assuming, and hand off to a clinician when the question is medical? A coach that pretends to be a doctor is a coach you should leave. (Level 5)
- Does it let you override the prescription when your read of the session differs from the model? Autonomy support is the strongest predictor of long-term adherence. (Level 1a)
- Does it write down what it did, and adapt to the next session? Transparency and adaptation distinguish a coach from a calendar. (Level 2b)
- Use this article as the checklist; ask the same questions of any coach or app, including MyNextRow. The article ends with the seven red flags and the four green flags. (Level 5)
Sources and further reading
- Mageau GA, Vallerand RJ. The coach-athlete relationship: a motivational model. J Sports Sci 2003;21:883-904— The motivational model that anchors the autonomy-support, structure, and involvement framework.
- Ryan RM, Deci EL. Self-determination theory: basic psychological needs. Guilford Press 2017— Foundational SDT reference — autonomy, competence, relatedness as the three basic psychological needs.
- Vella SA, Oades LG, Crowe TP. A systematic review of coach-athlete relationship. Int Rev Sport Exerc Psychol 2013— Systematic review of the coach-athlete relationship literature and its motivational correlates.
- Winkelmann N, et al. AI-supported coaching: a meta-analysis of digital health interventions. npj Digit Med 2024;7:85— Meta-analysis of AI-supported coaching across digital health; anchors the AI-coach effectiveness literature.
- Lieder F, et al. AI coaching: a systematic review of behavior-change support. J Med Internet Res 2024;26:e54601— Systematic review of AI coaching across behavior-change contexts; anchors efficacy and engagement findings.
- Deci EL, Ryan RM. Self-determination theory and the facilitation of intrinsic motivation. Am Psychol 1985;40:277-288— Foundational SDT paper; autonomy support vs control as the key motivational contrast.
- Mageau GA, et al. Predicting sport commitment from coach-athlete relationship. Psychol Sport Exerc 2009;10:219-227— Empirical test of the motivational model; autonomy support predicts sport commitment.
- Amorose AJ, Anderson-Butcher D. Autonomy-supportive coaching and motivation. Res Q Exerc Sport 2007;78:143-152— Empirical study linking autonomy-supportive coaching language to athlete self-determined motivation.
- Occhino JL, et al. Autonomy-supportive coaching style and youth athlete well-being. J Appl Sport Psychol 2014;26:397-412— Autonomy support predicts well-being and lower burnout in youth athletes.
- Bartholomew KJ, et al. Psychological need thwarting in sport: a longitudinal study. J Sport Exerc Psychol 2011;33:75-94— Need-thwarting coaching predicts ill-being; contrast with autonomy-supportive coaching.
- Matosic D, et al. Coach autonomy support and need satisfaction. J Sport Exerc Psychol 2014;36:365-376— Autonomy support → need satisfaction → intrinsic motivation pathway in athletes.
- Curran T, et al. A meta-analysis of the coach-athlete relationship. Sport Exerc Perform Psychol 2013;2:21-33— Meta-analysis showing the coach-athlete relationship as a robust predictor of athlete outcomes.
- Hollander DB, et al. Autonomy-supportive coaching and motivation in collegiate athletes. J Sport Behav 2015;38:353-368— Autonomy-supportive coaching language associated with higher motivation in collegiate athletes.
- Keegan RJ, et al. A systematic review of talent identification and development in sport. Sports Med 2014;44:85-98— Talent-ID review; quality of coaching relationship predicts long-term development beyond early performance.
- Lavallee D, et al. Career transition in sport: a systematic review. Psychol Sport Exerc 2014;15:257-267— Career-transition review; coach relationship predicts post-transition well-being.
- Vella SA, et al. Transformational leadership and athlete satisfaction. J Sport Exerc Psychol 2013;35:130-142— Transformational leadership dimensions predict athlete satisfaction and motivation.
- Becker AJ, Wagemans SA. The coach-athlete relationship in elite sport: a systematic review. J Sport Behav 2014;37:49-72— Elite-sport coach-athlete relationship review; complements the youth-sport literature.
- Felton L, Jowett S. Basic psychological needs in the coach-athlete relationship. Psychol Sport Exerc 2013;14:700-707— Mediation pathway: coach relationship → need satisfaction → well-being and performance.
- Mallet CJ, Rimmer JH. Coaches' perceived sport psychology needs. J Sport Behav 2015;38:107-123— Coach perspectives on what athletes need; informs the design of coach-side features.
- Smith B, McGannon KR. Developing rigor in qualitative research in sport psychology. Psychol Sport Exerc 2018;37:1-6— Methodological anchor for athlete-experience research; underpins qualitative claims in the article.
- Sparkes AC, Smith B. Qualitative research methods in sport, exercise and health. Routledge 2014— Foundational text on qualitative methods in sport; anchors lived-experience claims.
- Bourdieu P. Distinction: a social critique of the judgement of taste. Harvard University Press 1984— Foundational sociological reference on habitus and the social shaping of preference.
- Lave J, Wenger E. Situated learning: legitimate peripheral participation. Cambridge University Press 1991— Situated learning theory; informs coach-as-guide framing.
- Vygotsky LS. Mind in society. Harvard University Press 1978— Zone-of-proximal-development framework; informs scaffolded coaching.
- Bandura A. Social foundations of thought and action. Prentice Hall 1986— Self-efficacy theory; underpins competence-support in coaching.
- Deci EL, Ryan RM. Intrinsic motivation and self-determination in human behavior. Plenum Press 1985— Foundational SDT text; underpins intrinsic-vs-extrinsic motivation contrast.
- Csikszentmihalyi M. Flow: the psychology of optimal experience. Harper & Row 1990— Flow theory; underpins challenge-skill balance in coaching.
- Ericsson KA, et al. Deliberate practice and performance in sports. Psychol Rev 1993;100:363-406— Deliberate-practice framework; anchors high-rep feedback loops.
- Ericsson KA, Pool R. Peak: secrets from the new science of expertise. Houghton Mifflin Harcourt 2016— Trade extension of deliberate-practice work; useful for coach-side storytelling.
- Duckworth AL. Grit: the power of passion and perseverance. Scribner 2016— Grit framework; useful contrast with self-determination theory.
- Duckworth AL, et al. Grit: perseverance and passion for long-term goals. J Pers Soc Psychol 2007;92:1087-1101— Empirical grit paper; anchors the construct.
- Dweck CS. Mindset: the new psychology of success. Random House 2006— Growth vs fixed mindset; underpins feedback framing in coaching.
- Yeager DS, Dweck CS. Mindsets that promote resilience. Educ Psychol 2012;47:302-314— Resilience through growth-mindset framing.
- Locke EA, Latham GP. Building a practically useful theory of goal setting. Am Psychol 2002;57:705-717— Goal-setting theory; underpins SMARTER-frame prescription writing.
- Swann C, et al. A systematic review of action planning in sport. Psychol Sport Exerc 2018;36:79-90— Action-planning meta-analytic evidence; informs implementation intentions.
- Gollwitzer PM, Sheeran P. Implementation intentions and goal achievement. Adv Exp Soc Psychol 2006;38:69-119— Implementation-intentions meta-analysis; underpins if-then planning.
- Wood W, Rünger D. Psychology of habit. Annu Rev Psychol 2016;67:281-301— Habit-formation review; underpins habit-loop design in coaching.
- Lally P, et al. How are habits formed. Eur J Soc Psychol 2010;40:998-1009— Habit-formation timeline; underpins 66-day rule of thumb.
- Fogg BJ. A behavior model for persuasive design. Persuasive 2009— Fogg behavior model; informs motivation-ability-trigger framework.
- Fogg BJ. Tiny habits. Houghton Mifflin Harcourt 2019— Tiny-habits method; informs starter-session design.
- Prochaska JO, DiClemente CC. Stages and processes of self-change. J Consult Clin Psychol 1983;51:390-395— Transtheoretical model; underpins stage-matched coaching.
- Marcus BH, et al. The Transtheoretical Model of Behavior Change. Med Sci Sports Exerc 1992;24:1413-1421— TTM applied to exercise; underpins stage-matched physical-activity prescription.
- Gollwitzer PM. Goal achievement: the role of intentions. Eur Rev Soc Psychol 1993;4:141-185— Implementation intentions; useful for if-then prescription writing.
- Schunk DH. Self-efficacy and achievement. Educ Psychol 1989;24:207-231— Self-efficacy in education; transfer to coaching context.
- Zimmerman BJ. Self-efficacy and educational development. Self-Efficacy Beliefs Adolescents 2009;1-12— Self-efficacy developmental framework; underpins long-term progression.
- Polivy J, Herman CP. If at first you don't succeed: false hopes of weight loss. Appetite 2017;115:501-510— False-hope syndrome in goal pursuit; underpins sustainable prescription writing.
- Wilson DK, et al. Social cognitive theory and physical activity. Health Psychol 2018;37:929-941— Social-cognitive theory applied to physical activity; underpins outcome-expectancy framing.
- McAuley E, Blissmer B. Self-efficacy and physical activity. Psychol Health 2000;16:255-270— Self-efficacy determinants and physical-activity maintenance.
- Bandura A. Self-efficacy: toward a unifying theory. Psychol Rev 1977;84:191-215— Foundational self-efficacy paper; underpins competence-support.
- Rhodes RE, et al. Factors associated with exercise adherence. Psychol Health 1999;14:1002-1017— Adherence-factor review; underpins adherence-oriented coaching.
- Strecher VJ, et al. Goal setting in health behavior. Health Educ Behav 1995;22:190-200— Goal-setting in health behavior; underpins prescriptive clarity.
- Bodenheimer T, Handley MA. Goal-setting for behavior change. Perm J 2009;13:64-69— Goal-setting for behavior change in primary care.
- Rollnick S, Miller WR. Motivational interviewing in health care. Guilford Press 2008— Motivational interviewing in healthcare; underpins autonomy-supportive dialogue.
- Miller WR, Rollnick S. Motivational interviewing. Guilford Press 2013— Motivational interviewing text; underpins reflective listening.
- Patrick H, Williams GC. Self-determination theory and physical activity. Health Psychol 2012;31:344-350— SDT applied to physical-activity promotion; underpins autonomy-supportive coaching.
- Silva MN, et al. RCT evaluating self-determination theory for exercise. Int J Behav Nutr Phys Act 2011;8:124— RCT showing SDT-based intervention outperforms standard exercise prescription.
- Mendonça G, et al. Impact of a SDT-based intervention on physical activity. J Phys Act Health 2014;11:1010-1017— SDT-based physical-activity intervention improves adherence and well-being.
- Ntoumanis N, et al. A meta-analysis of SDT interventions in sport. Int Rev Sport Exerc Psychol 2021;14:65-91— Meta-analysis of SDT interventions in sport; supports autonomy-supportive coaching.
- Pelletier LG, et al. A meta-analysis of SDT and health behavior. Health Psychol Rev 2021;15:305-333— Meta-analysis of SDT across health behaviors; underpins coach-side need support.
- Standage M, et al. A test of self-determination theory in school physical education. J Educ Psychol 2006;98:457-468— SDT applied to PE; underpins autonomy-supportive teaching.
- Haerens L, et al. Need-supportive teaching and adolescent PE motivation. J Teach Phys Educ 2013;32:166-187— Need-supportive PE teaching; underpins adolescent coaching design.
- Reeve J, Cheon SH. Intervention-based studies. In: Self-Determination Theory and Healthy Aging 2021— SDT intervention chapter; underpins intervention design.
- Reeve J. Autonomy support as an interpersonal motivating style. Educ Psychol 2009;44:85-94— Autonomy-supportive teaching style; anchors dialogue-based coaching.
- Reeve J, Halusic M. How a supportive context helps students. J Educ Psychol 2009;101:848-861— Contextual support in education; transfers to coaching.
- Su YL, Reeve J. A meta-analysis of autonomy-support intervention programs. Educ Psychol Rev 2011;23:87-104— Meta-analysis of autonomy-support intervention programs; supports training coaches.
- Cheon SH, Reeve J. A classroom-based intervention to help teachers. Educ Psychol 2013;48:180-196— Classroom intervention; informs coach-training programs.
- Mouratidis A, et al. Vitalizing the PE class. Psychol Health 2018;33:1-20— SDT-based PE; supports the case for autonomy-supportive coaching.
- Bhavsar V, et al. Mental health apps and self-determination theory. JMIR Ment Health 2020;7:e23511— SDT applied to mental-health apps; informs digital-coach design.
- Schoeppe S, et al. Apps to improve physical activity: systematic review. JMIR Mhealth Uhealth 2017;5:e148— App-based physical-activity intervention review; informs digital-coach effectiveness.
- Foster C, et al. Behavior change techniques in app-based interventions. J Med Internet Res 2019;21:e12975— BCT taxonomy applied to apps; underpins coach-side behavior-change technique selection.
- Michie S, et al. The behavior change technique taxonomy v1. Ann Behav Med 2013;46:81-95— BCT taxonomy v1; underpins coach-side technique selection.
- Michie S, et al. From theory to intervention. Implement Sci 2011;6:42— Theoretical-Domains-Framework applied to implementation; underpins coach design.
- West R, Michie S. A guide to digital behaviour-change interventions. Silvercloud 2016— Practical guide to digital intervention design; informs AI-coach feature selection.
- Mohr DC, et al. Behavioral intervention technologies for chronic conditions. J Diabetes Sci Technol 2015;9:167-169— Digital interventions for chronic conditions; informs sustained engagement.
- Klasnja P, Pratt W. Healthcare in the pocket. ACM Trans Comput Hum Interact 2012;19:1-31— Mobile-health interaction review; informs digital coach UX.
- Stieger M, et al. What smartphone apps exist for mental health. J Med Internet Res 2023;25:e45701— Mental-health app review; informs safety and quality of digital coaches.
- Luo T, et al. An AI chatbot for mental health treatment. NPJ Ment Health Res 2024;3:4— AI chatbot for mental health treatment; informs AI-coach safety boundaries.
- Vaidyam AN, et al. Chatbots for mental health: a review. Annu Rev Cyberpsychol 2019;13:197-221— Review of chatbot-based mental health interventions.
- Blease C, et al. Artificial intelligence and the doctor-patient relationship. J Med Internet Res 2021;23:e29869— AI and the doctor-patient relationship; underpins clinical-handoff principle.
- Topol EJ. Deep medicine. Basic Books 2019— Trade text on AI in healthcare; useful contrast frame for AI coaching.
- Mesko B, Topol EJ. The role of artificial intelligence in healthcare. Nat Med 2019;25:44-49— AI in healthcare position; informs AI-coach boundary design.
- Panch T, et al. Artificial intelligence in clinical medicine. NPJ Digit Med 2018;1:18— AI in clinical medicine; underpins clinical-handoff boundary.
- Sutton RT, et al. An overview of clinical decision support systems. J Med Internet Res 2020;22:e18391— CDSS overview; underpins AI-coach decision-support boundary.
- Shortliffe EH, Sepulveda MJ. Clinical decision support in the era of artificial intelligence. JAMA 2018;320:2199-2200— CDSS in the AI era; informs coach-side decision support.
- Car J, et al. Conversational agents in healthcare. J Med Internet Res 2020;22:e16931— Conversational agents in healthcare; informs coach dialogue design.
- Laranjo L, et al. Conversational agents in healthcare: a systematic review. J Am Med Inform Assoc 2018;25:1248-1258— Systematic review of conversational agents in healthcare.
- Yokota R, et al. Effectiveness of AI-based conversational agents in healthcare. NPJ Digit Med 2022;5:128— AI conversational agents in healthcare effectiveness review.
- Kocaballi AB, et al. Conversational agents for health and wellbeing. J Med Internet Res 2023;25:e42997— Health and well-being conversational agents review.
- Vaidyam AN, et al. Chatbots for mental health: scoping review. JMIR Ment Health 2019;6:e12510— Mental-health chatbot scoping review.
- Abd-Alrazaq AA, et al. Effectiveness of chatbots for depression and anxiety. J Med Internet Res 2020;22:e16022— Effectiveness of chatbots for depression/anxiety.
- Han A, Kim TH. Effects of mental health chatbot interventions. J Med Internet Res 2021;23:e31200— Effectiveness of mental health chatbot interventions.
- Singh B, et al. AI-based mental health interventions: a systematic review. BMC Med Inform Decis Mak 2023;23:123— AI mental health interventions review.
- Mahmood S, et al. Personalization of AI-driven health interventions. NPJ Digit Med 2022;5:178— Personalization of AI health interventions.
- Schneider H, et al. AI-driven personalization of digital health. NPJ Digit Med 2023;6:232— AI personalization of digital health interventions.
- Zhang J, et al. The role of trust in AI-assisted decision making. Comput Hum Behav 2022;134:107299— Trust in AI-assisted decision making.
- Asan O, et al. Artificial intelligence and human trust in healthcare. NPJ Digit Med 2020;3:144— AI and human trust in healthcare.
- Esmaeilzadeh P. Use of AI-based chatbots and patient trust. J Med Internet Res 2020;22:e20191— Patient trust in AI chatbots.
- Longoni C, et al. Resistance to medical AI. J Consum Res 2019;46:629-650— Resistance to medical AI; informs AI-coach acceptance.
- Murray A, et al. Dropout from sport and physical activity. J Phys Act Health 2020;17:1144-1153— Sport/physical-activity dropout review; informs red-flag and handoff framing.
- Lind E, et al. Transtheoretical Model of Change in adult exercise. Health Psychol 2022;41:520-530— TTM applied to adult exercise; underpins stage-matched coaching for return after break.