Skip to content

Meal, nap and sleep prompts: what the literature says

Research for Sioul meal reminders (N meals/day, eating time + duration + preparation time; no work planned in those slots; notice 15 min before preparation; reminder at preparation time; same mechanism for naps and night sleep). Compiled 2026-10-05. Sections 1–6 follow the research brief; "Design criteria" and "Open questions / conflicts" are at the end. Nothing here is about any particular person.

Strength scale used below: strong = guideline, sound meta-analysis, or RCT with n > 100; moderate = RCT < 100, large cross-sectional or cohort study, review or meta-analysis of weak or heterogeneous studies; weak = single small study, qualitative study, case series; opinion = expert/clinical manual, theory, no trial behind the specific claim. Ratings are downgraded where a study's own limits warrant; the reason is given in brackets. Every reference was checked against PubMed, Crossref, the ACM DL or the publisher page before being written here unless marked "unverified".

Summary

  • Eating at planned times instead of waiting for hunger is the shared first step of CBT-E, CBT-AR (ARFID) and NICE-recommended treatment for BN, BED and OSFED, and it is an active ingredient (regular eating mediates the drop in binge eating). Hunger is also an unreliable cue in AN, ARFID, ADHD and depression.
  • Meal reminders split users: structure for some; for others a reminder of illness in the middle of other things, a nuisance in company, or condescension ("It's not like my problem is that I forget to eat, but that I sometimes don't want to"). User control over every parameter, and the option to silence, decided which.
  • Food numbers, logs, streaks, red/green colours, praise, warnings and "missed" messages are linked to harm in people at risk; a time-keeping feature needs none of them.
  • Interruption research supports a heads-up before the transition and a "where I stopped" note; autonomy over the break decides whether it restores.
  • Sleep: fixed times plus a wind-down notice is a component of tested treatments; sleep scores can feed perfectionism. Naps need a recovery buffer after them.
  • The strong evidence supports regular eating itself; evidence on reminder design is mostly small qualitative studies (see "Open questions").

1. Regular eating as treatment

  • Murphy R, Straebler S, Cooper Z, Fairburn CG (2010). Cognitive behavioral therapy for eating disorders. Psychiatr Clin North Am 33(3):611–627. doi:10.1016/j.psc.2010.04.004 (PMC2928448)

    • Narrative review by the CBT-E authors; describes the manual (Fairburn 2008, Cognitive Behavior Therapy and Eating Disorders, Guilford). Opinion/manual for the procedure; CBT-E itself is supported by RCTs.
    • Regular eating is "fundamental to successful treatment whatever the form of the eating disorder. It addresses an important type of dieting ('delayed eating'); it displaces most episodes of binge eating; it structures people's days".
    • Spec: "3 planned meals each day plus 2 or 3 planned snacks so that there is rarely more than a 4-hour interval between them"; the person chooses what to eat; "The new eating pattern should take precedence over other activities but should not be so inflexible as to preclude the possibility of adjusting timings to suit the patients' commitments each day."
    • Self-monitoring (real-time food record) is central in CBT-E but is reviewed jointly with the therapist every session and is discontinued at the end of treatment (stage four). It is a time-limited clinical tool, not a lifelong habit.
    • Clinical perfectionism leads to "overly detailed recording" — recording itself can feed the disorder.

  • Fairburn CG, Cooper Z, Shafran R (2003). Cognitive behaviour therapy for eating disorders: a "transdiagnostic" theory and treatment. Behav Res Ther 41(5):509–528. doi:10.1016/S0005-7967(02)00088-8

    • Theory paper. Opinion (foundational; widely tested since).
    • Same maintaining mechanisms (over-evaluation of eating/shape/weight, dietary restraint, plus perfectionism, low self-esteem, mood intolerance) across AN, BN and atypical EDs (now OSFED): one transdiagnostic treatment, regular eating included for all.

  • NICE (2017). Eating disorders: recognition and treatment. NICE guideline NG69, published 23 May 2017, last updated 16 December 2020. https://www.nice.org.uk/guidance/ng69 Strong (national guideline, evidence-graded).

    • 1.4.7 (BED, individual CBT-ED): "advise people to eat regular meals and snacks to avoid feeling hungry"; also "advise the person not to try to lose weight (for example, by dieting) during treatment, because this is likely to trigger binge eating".
    • 1.5.5 (BN, individual CBT-ED): first phase focuses on "establishing a pattern of regular eating, and providing encouragement, advice and support while people do this".
    • 1.5.7 (BN, young people): "a collaborative approach ... to establish regular eating patterns".
    • 1.3.5 (AN, CBT-ED) and 1.4.5/1.4.7 (BED): self-monitoring of dietary intake is part of therapy, inside sessions with a practitioner.
    • 1.6.1: OSFED → use the treatment of the disorder it most resembles. First-line for adult BED and BN is guided self-help (1.4.3, 1.5.3) using CBT self-help materials (in practice Fairburn's Overcoming Binge Eating, whose step 2 is regular eating).

  • Zendegui EA, West JA, Zandberg LJ (2014). Binge eating frequency and regular eating adherence: the role of eating pattern in cognitive behavioral guided self-help. Eating Behaviors 15(2):241–243. doi:10.1016/j.eatbeh.2014.03.002

    • Session-by-session data, n = 38, 10 sessions of CBT guided self-help. Weak (small, correlational).
    • Weeks with higher adherence to "3 meals and 2–3 planned snacks daily" had fewer binges, within persons.

  • Sivyer K, Allen E, Cooper Z, Bailey-Straebler S, O'Connor ME, Fairburn CG, Murphy R (2020). Mediators of change in cognitive behavior therapy and interpersonal psychotherapy for eating disorders: a secondary analysis of a transdiagnostic randomized controlled trial. Int J Eat Disord 53(12):1928–1940. doi:10.1002/eat.23390

    • Mediation analysis inside an RCT, n = 130, transdiagnostic non-underweight. Moderate.
    • Regular eating mediated CBT-E's effect on binge frequency; weighing frequency and shape checking did not show clear mediation. The best direct evidence that regular eating is an active ingredient.

  • Ellison JM, Simonich HK, Wonderlich SA, et al. (2016). Meal patterning in the treatment of bulimia nervosa. Eating Behaviors 20:39–42. doi:10.1016/j.eatbeh.2015.11.008

    • Secondary analysis of an RCT, n = 80 adults with BN. Weak/moderate.
    • More evening meals over treatment → fewer binges and purges at 4-month follow-up, independent of depression change. The evening meal mattered most.

  • Zunker C, Peterson CB, Crosby RD, Cao L, Engel SG, Mitchell JE, Wonderlich SA (2011). Ecological momentary assessment of bulimia nervosa: does dietary restriction predict binge eating? Behav Res Ther 49(10):714–717. doi:10.1016/j.brat.2011.06.006

    • EMA, n = 133 women with BN, 2 weeks. Moderate.
    • Odds of a binge rose on the day of restriction and the next day. Long gaps/restriction precede binges in daily life, not just in theory.

  • Masheb RM, Grilo CM, White MA (2011). An examination of eating patterns in community women with bulimia nervosa and binge eating disorder. Int J Eat Disord 44(7):618–624. doi:10.1002/eat.20853

    • Online survey, n = 311 (BN 39, BED 69, controls 203). Weak/moderate.
    • BN group ate fewer meals, especially lunch; nibbling, double meals and night eating more common in ED groups; more meals ↔ less binge eating in BED.

1b. Eating by the clock ("mechanical eating"), ARFID, unreliable hunger

  • MacDonald DE, Trottier K, Cao L, Crosby RD, Wonderlich SA, Engel SG, Olmsted MP (2024). Momentary skills use predicts decreased binge eating and purging early in day treatment: an ecological momentary assessment study. Int J Eat Disord 57(3):548–557. doi:10.1002/eat.24128

    • EMA several times a day for 28 days, n = 58 with BN or purging disorder in CBT day treatment. Moderate (real-time data, small n, observational).
    • Using "planning ahead", "distraction", "social support" and "mechanical eating" (lay summary: "eating meals and snacks regardless of how one is feeling") was linked to lower odds of bingeing/purging in the same period. The only real-time evidence found that eating by the clock helps in the moment.

  • Bell E, Rieger E, Murray K, Monaghan C (2026). Clarifying the relationship between intuitive eating, mindful eating, and maladaptive eating: a factor analytic study. J Health Psychol 31(12):4958–4973. doi:10.1177/13591053261418428 (PMC13612868)

    • Two community samples of women, n = 400 + 423, factor analysis. Moderate (cross-sectional).
    • "Mechanical Eating" (routine-based eating, e.g. "I eat the same thing for lunch each day") came out as an ambiguous factor, moderately correlated with clinical impairment (r = 0.21, 0.37). Authors: it may mix the adaptive CBT meal plan with rigid control ("unable to relinquish control over their diet plans"). Routine around times is the therapeutic part; routine around content can be rigidity.

  • McMaster CM, Wade T, Basten C, Franklin J, Ross J, Hart S (2021). Rationale and development of a manualised dietetic intervention for adults undergoing psychological treatment for an eating disorder. Eat Weight Disord 26(5):1467–1481. doi:10.1007/s40519-020-00955-y

    • Intervention description + feasibility. Opinion/weak.
    • Five dietitian sessions; session 2 is "mechanical eating and dietary rules", session 4 "maximising your meal plan and meal preparation". Mechanical eating and meal preparation are treated as skills to teach, next to each other.
  • McMaster CM, Wade T, Franklin J, Hart S (2021). A review of treatment manuals for adults with an eating disorder: nutrition content and consistency with current dietetic evidence. Eat Weight Disord 26(1):47–60. doi:10.1007/s40519-020-00850-6
    • Review of 22 manuals. Weak (narrative review). 13/22 contained food advice not backed by evidence; only 2 had dietitian input. Caution against an app inventing food rules.

  • Grey literature, for the definition only: clinical dietitian sources define mechanical eating as eating "by the clock" at planned times instead of waiting for hunger/fullness cues, until cues become reliable again (e.g. Equip Health, https://equip.health/articles/food-and-fitness/mechanical-eating-eating-disorder-recovery). Opinion, not peer-reviewed.

  • Thomas JJ, Becker KR, Kuhnle MC, et al. (2020). Cognitive-behavioral therapy for avoidant/restrictive food intake disorder: feasibility, acceptability, and proof-of-concept for children and adolescents. Int J Eat Disord 53(10):1636–1646. doi:10.1002/eat.23355 (PMC7719612)

  • Thomas JJ, Becker KR, Breithaupt L, et al. (2021). Cognitive-behavioral therapy for adults with avoidant/restrictive food intake disorder. J Behav Cogn Ther 31(1):47–55. doi:10.1016/j.jbct.2020.10.004 (PMC8375627)
    • Open trials, n = 20 (ages 10–17) and n = 15 adults. Weak (no control group). Manual: Thomas JJ, Eddy KT (2019), Cognitive-Behavioral Therapy for Avoidant/Restrictive Food Intake Disorder, Cambridge University Press.
    • Stage 1 of CBT-AR: psychoeducation, self-monitoring, and "establishing (or reinforcing) a regular schedule of eating", same for children and adults. The "lack of interest in eating / low appetite" presentation improved least, and the authors suggest low appetite there may be neurobiological (low ghrelin), so carers "need to continue to support ... eating sufficient amounts even upon treatment completion". For an adult living alone, an external time cue is the nearest stand-in (my inference, not the paper's).
  • Thomas JJ, Wons OB, Eddy KT (2018). Cognitive-behavioral treatment of avoidant/restrictive food intake disorder. Curr Opin Psychiatry 31(6):425–430. doi:10.1097/YCO.0000000000000454 (PMC6235623)
    • Review by the CBT-AR authors. Opinion (evidence base was case reports and series). Stage 1 (Table 1): "Establishing a pattern of regular eating to normalize hunger cues". Regular eating is common to all CBT variants for ARFID found in case reports.

  • Thomas JJ, Lawson EA, Micali N, Misra M, Deckersbach T, Eddy KT (2017). Avoidant/restrictive food intake disorder: a three-dimensional model of neurobiology with implications for etiology and treatment. Curr Psychiatry Rep 19(8):54. doi:10.1007/s11920-017-0795-5

    • Theory/review. Opinion. Three ARFID presentations: sensory sensitivity, lack of interest in eating (homeostatic appetite), fear of aversive consequences (choking, vomiting). Only the second is helped by time cues; the third makes food-related wording in a notification potentially aversive.

  • Khalsa SS, Craske MG, Li W, Vangala S, Strober M, Feusner JD (2015). Altered interoceptive awareness in anorexia nervosa: effects of meal anticipation, consumption and bodily arousal. Int J Eat Disord 48(7):889–897. doi:10.1002/eat.22387

    • Lab experiment, 15 AN vs 15 controls. Weak.
    • People with AN over-reported body sensations at low arousal (interoceptive "prediction errors"); in the AN group cardiorespiratory sensations, especially breathlessness, were stronger while waiting for the meal, "particularly enhanced prior to, but not following, meal receipt". Body signals around meals are distorted, and the wait before a meal is itself aversive — an argument for short, predictable lead times rather than long countdowns (my inference).

  • Kutscheidt K, Dresler T, Hudak J, et al. (2019). Interoceptive awareness in patients with attention-deficit/hyperactivity disorder (ADHD). Atten Defic Hyperact Disord 11(4):395–401. doi:10.1007/s12402-019-00299-3

    • Heartbeat-counting task, 14 adults with ADHD vs 16 controls. Weak.
    • ADHD group was worse at detecting their own heartbeat → less aware of internal body signals. Consistent with "forgetting to eat"; does not test hunger directly.

  • Turner H, Bryant-Waugh R, Marshall E (2015). The impact of early symptom change and therapeutic alliance on treatment outcome in cognitive-behavioural therapy for eating disorders. Behav Res Ther 73:165–169. doi:10.1016/j.brat.2015.08.006

    • Service cohort, n = 94 adults. Moderate.
    • Early drop in dietary restraint by session 6 (the regular-eating phase) predicted end-of-treatment outcome better than early alliance. Behavioural change in eating pattern first, then the rest.

2. Eating-disorder apps and meal reminders

  • Lindgreen P, Lomborg K, Clausen L (2018). Patient experiences using a self-monitoring app in eating disorder treatment: qualitative study. JMIR mHealth uHealth 6(6):e10253. doi:10.2196/10253 (PMC6035344)

    • Field observation of 41 outpatients (AN, BN; aged 15+, Denmark) using Recovery Record in treatment; 26 interviewed. Weak/moderate (qualitative, but the only in-depth study of real meal-reminder use found).
    • The single most relevant study for Sioul. Each feature sat on a "continuum from supportive to obstructive", and the same person moved along it with context and over time.
    • Meal reminders, supportive: "I need that structure in my life in order to eat what I'm supposed to."
    • Meal reminders, obstructive, for four distinct reasons: (1) inconvenient times, e.g. with friends — "supportive when alone but obstructive when with friends"; (2) they "pinpoint their illness to them when otherwise engaged, for instance, in school work" — "you just wanna move on"; (3) they misread the problem: "It's not like my problem is that I forget to eat, but that I sometimes don't want to."; (4) condescending — being treated as "incapable of structuring meals and remembering to eat". Those patients turned reminders off or dropped the app.
    • Affirmations after logging: helpful for some; for others "childish and condescending": "here's a treat for you, since you've been good ... That's reinforced by the app treating you like a child." Gamification (puzzle piece per logged meal) felt "too much fun and games" for something serious.
    • Logging: some patients transferred obsessive calorie-logging habits from fitness apps into the recovery app; preset questions about exercise or purging triggered urges. Logging meals that caused no distress felt pointless.
    • Notifications that arrived at unexpected times (clinician reviews on an unexpected weekday) caused rumination and a feeling of surveillance.
    • Authors: "build flexible apps that may easily be adjusted to fit individual patient's preferences".

  • Lindgreen P, Clausen L, Lomborg K (2018). Clinicians' perspective on an app for patient self-monitoring in eating disorder treatment. Int J Eat Disord 51(4):314–321. doi:10.1002/eat.22833

    • Field studies with 31 clinicians, 23 interviewed. Weak (qualitative).
    • Clinicians worried about "pacified patients" (the app taking over initiative) and about surveillance/workload. Relevant: a reminder tool should leave the decision with the person.

  • Tregarthen JP, Lock J, Darcy AM (2015). Development of a smartphone application for eating disorder self-monitoring. Int J Eat Disord 48(7):972–982. doi:10.1002/eat.22386

    • Case report + usage data (Recovery Record), >100,000 users over two years, EDE-Q from 57,940. Weak (descriptive).
    • App logs meals, emotions, behaviours, thoughts, with meal reminders, positive reinforcement and coping suggestions. ~50% of users were not in treatment and 33% had told nobody about their ED. Many people who will meet a meal reminder in a general app have an undisclosed ED; the reminder may be the only "clinical" contact they have.

  • Juarascio AS, Goldstein SP, Manasse SM, Forman EM, Butryn ML (2015). Perceptions of the feasibility and acceptability of a smartphone application for the treatment of binge eating disorders: qualitative feedback from a user population and clinicians. Int J Med Inform 84(10):808–816. doi:10.1016/j.ijmedinf.2015.06.004 (PMC4860812)

    • Focus groups with 11 people with binge eating + phone interviews with 10 clinicians, about a mock-up app. Weak (qualitative, hypothetical app).
    • Users wanted reminders but brief ("if the app could have reminders with short text then that would be helpful"); frequency and type must be user-set ("desired frequency ... was highly varied"); a snooze "if they were not able to utilize the intervention in that specific moment, but wanted a reminder later".
    • Wording: avoid clinical terms ("binge", "binge episodes") and acronyms; prefer "personal and caring" phrasing over "You may be at risk for a binge episode".
    • A requested reminder type was preparatory: "to bring a snack to work".
    • Some users wanted the option to keep data from their clinician: "I wouldn't be as honest in the app if it went straight to someone else."
    • Conflict: this group also liked badges and levels for engagement, which Eikey 2021 and Lindgreen 2018 found harmful or condescending in practice.

  • Fairburn CG, Rothwell ER (2015). Apps and eating disorders: a systematic clinical appraisal. Int J Eat Disord 48(7):1038–1046. doi:10.1002/eat.22398 (PMC4737215)

    • Systematic app-store review: 805 apps screened, 39 for people with EDs. Moderate (systematic, but apps change).
    • Advice quality ranged "from sound to potentially harmful"; "The enthusiasm for apps outstrips the evidence supporting their use."

2b. App trials, text messages, and general findings on reminders and wording

  • Hildebrandt T, Michaelides A, Mackinnon D, Greif R, DeBar L, Sysko R (2017). Randomized controlled trial comparing smartphone assisted versus traditional guided self-help for adults with binge eating. Int J Eat Disord 50(11):1313–1322. doi:10.1002/eat.22781 (PMC5755703)
    • RCT, n = 66 (BED or BN), CBT guided self-help + Noom Monitor app vs paper. Moderate.
    • App group ate the prescribed "three meals and two snacks" on more days, and regular-meal adherence mediated the larger drop in binge episodes. Self-monitoring is framed as a skill used during treatment, "not required post-CBT-GSH". No group difference at 6 months.

  • Hildebrandt T, Michaeledes A, Mayhew M, et al. (2020). Randomized controlled trial comparing health coach-delivered smartphone-guided self-help with standard care for adults with binge eating. Am J Psychiatry 177(2):134–142. doi:10.1176/appi.ajp.2019.19020184

    • RCT, n = 225 in a health system, 52-week outcomes. Strong. Coach-led CBT guided self-help + Noom Monitor cut objective binge days more than standard care. Evidence that app-supported CBT self-help (whose early core is regular eating) works at scale with a human coach.

  • Linardon J, Shatte A, Rosato J, Fuller-Tyszkiewicz M (2022). Efficacy of a transdiagnostic cognitive-behavioral intervention for eating disorder psychopathology delivered through a smartphone app: a randomized controlled trial. Psychol Med 52(9):1679–1690. doi:10.1017/S0033291720003426

    • RCT, n = 392 people with binge eating, app (CBT-E principles) vs waitlist. Strong for short-term efficacy (8 weeks); 35% attrition.
    • Global ED psychopathology fell more with the app (d = −0.80). Shows a standalone CBT app can help without a clinician (the abstract does not list modules); says nothing specific about reminder design.
  • Linardon J, Messer M, Shatte A, Skvarc D, Rosato J, Rathgen A, Fuller-Tyszkiewicz M (2023, online 2021). Targeting dietary restraint to reduce binge eating: a randomised controlled trial of a blended internet- and smartphone app-based intervention. Psychol Med 53(4):1277–1287. doi:10.1017/S0033291721002786
    • RCT, n = 403 (90% bingeing weekly), intervention vs information control. Strong for short term (8-week follow-up).
    • An intervention made "solely of cognitive-behavioural techniques designed to modify dietary restraint" (in CBT terms that means regular eating and loosening food rules; the abstract does not list the techniques) reduced objective binge episodes (small effect) and restraint, shape/weight/eating concerns (small to large). Targeting restraint alone was enough to move core symptoms.

  • Linardon J, King T, Shatte A, Fuller-Tyszkiewicz M (2022). Usability evaluation of a cognitive-behavioral app-based intervention for binge eating and related psychopathology: a qualitative study. Behav Modif 46(5):1002–1020. doi:10.1177/01454455211021764

    • Interviews, n = 14. Weak. Liked: flexible content formats, interactivity, plain information, tracking progress; disliked: visual aesthetics, no professional feedback.

  • Tregarthen J, Paik Kim J, Sadeh-Sharvit S, Neri E, Welch H, Lock J (2019). Comparing a tailored self-help mobile app with a standard self-monitoring app for the treatment of eating disorder symptoms: randomized controlled trial. JMIR Ment Health 6(11):e14972. doi:10.2196/14972

    • RCT inside Recovery Record, 3294 recruited, 959 analysed (high loss). Moderate. Both versions improved; recruitment shows how many app users with ED symptoms have no treatment access.

  • Neumayr C, Voderholzer U, Tregarthen J, Schlegl S (2019). Improving aftercare with technology for anorexia nervosa after intensive inpatient treatment: a pilot randomized controlled trial with a therapist-guided smartphone app. Int J Eat Disord 52(10):1191–1201. doi:10.1002/eat.23152

    • Pilot RCT, n = 40 women with AN after discharge, 8 weeks of Recovery Record + therapist feedback. Weak. High adherence and acceptance; small non-significant benefits that wore off by 6 months.

  • Shapiro JR, Bauer S, Andrews E, et al. (2010). Mobile therapy: use of text-messaging in the treatment of bulimia nervosa. Int J Eat Disord 43(6):513–519. doi:10.1002/eat.20744

    • Uncontrolled, n = 31 women in group CBT; nightly SMS with symptom counts, automatic tailored feedback. Weak. 87% adherence, good acceptability.
  • Bauer S, Okon E, Meermann R, Kordy H (2012). Technology-enhanced maintenance of treatment gains in eating disorders: efficacy of an intervention delivered via text messaging. J Consult Clin Psychol 80(4):700–706. doi:10.1037/a0028030
    • RCT, n = 165 women after inpatient treatment for BN/EDNOS; weekly SMS symptom report + tailored feedback for 16 weeks. Strong/moderate. Remission at 8 months 51% vs 36% (ITT, p = .05); the effect was in those without outpatient care. A low-frequency, weekly contact was enough.

  • Juarascio AS, Presseller EK, Srivastava P, Manasse SM, Forman EM (2023). A randomized controlled trial of CBT+: a clinician-controlled, just-in-time, adjunctive intervention for bulimia-spectrum disorders. Behav Modif 47(3):551–572. doi:10.1177/01454455221109434

    • RCT, n = 56. Weak/moderate. Just-in-time prompts were feasible and acceptable but added nothing over self-monitoring alone. Smarter, more frequent prompting is not automatically better.

  • Bakker D, Kazantzis N, Rickwood D, Rickard N (2016). Mental health smartphone apps: review and evidence-based recommendations for future developments. JMIR Ment Health 3(1):e7. doi:10.2196/mental.4984

    • Narrative review → 16 recommendations. Opinion backed by cited studies.
    • "too many annoying or interruptive reminders can lead to disengagement ... anything that quashes a sense of autonomy, such as a series of insistent reminders, can reduce intrinsic motivation". Reminders on by default "at a frequency that is not overwhelming", settings customisable. Language "nonclinical, nonpsychopathological, and nondiagnostic"; labelling users with a diagnosis can be stigmatising.
  • Carpenter CJ (2013). A meta-analysis of the effectiveness of the "but you are free" compliance-gaining technique. Communication Studies 64(1):6–17. doi:10.1080/10510974.2012.727941
    • Meta-analysis, 42 studies (as cited by Bakker 2016). Moderate. Saying the person is free to refuse raises compliance (r ≈ 0.13). Small effect, but in the right direction for autonomy-respecting wording.
  • Miller CH, Lane LT, Deatrick LM, Young AM, Potts KA (2007). Psychological reactance and promotional health messages: the effects of controlling language, lexical concreteness, and the restoration of freedom. Human Communication Research 33(2):219–240. doi:10.1111/j.1468-2958.2007.00297.x
    • Experiment with young adults. Moderate. Controlling language ("must", "should") had "a number of negative outcomes"; a postscript restoring freedom helped; concrete wording got more attention and was rated more important than abstract wording. Supports "Lunch at 12:30 — kitchen time" over "Remember to take care of yourself".

  • Rains SA (2013). The nature of psychological reactance revisited: a meta-analytic review. Human Communication Research 39(1):47–73. doi:10.1111/j.1468-2958.2012.01443.x — meta-analytic support for the reactance model (freedom threat → anger and counter-arguing → rejection). Moderate/strong.

  • Dennison L, Morrison L, Conway G, Yardley L (2013). Opportunities and challenges for smartphone applications in supporting health behavior change: qualitative study. J Med Internet Res 15(4):e86. doi:10.2196/jmir.2583 (PMC3636318)

    • 4 focus groups, n = 19 young adults (general population). Weak.
    • People wanted choice of frequency and timing, positive tone, and the ability to "silence or disable the entire app at certain times". Prompts felt like "nagging" or "harassing" and led to abandonment. "I would get really annoyed if I was trying to do my work or something and it was telling me I hadn't moved, I'd be like 'I'm busy!'" A user whose app sent progress to her boyfriend was embarrassed that "he continues to receive notifications of what she is not doing". Punitive or didactic feedback and seeing failed goals upset and demotivated.

3. Harms of food and fitness tracking

  • Levinson CA, Fewell L, Brosof LC (2017). My Fitness Pal calorie tracker usage in the eating disorders. Eating Behaviors 27:14–16. doi:10.1016/j.eatbeh.2017.08.003

    • Cross-sectional survey, n = 105 people diagnosed with an eating disorder. Moderate (clinical sample, self-report, no causality).
    • ~75% had used MyFitnessPal; 73% of those users said the app contributed to their eating disorder; that perception correlated with symptom severity.

  • Simpson CC, Mazzeo SE (2017). Calorie counting and fitness tracking technology: associations with eating disorder symptomatology. Eating Behaviors 26:89–92. doi:10.1016/j.eatbeh.2017.02.002

    • Cross-sectional, n = 493 college students. Moderate (association only).
    • Calorie-tracker users had higher eating concern and dietary restraint (controlling for BMI); fitness tracking was uniquely associated with ED symptoms. Authors: "for some individuals, these devices might do more harm than good."

  • Linardon J, Messer M (2019). My fitness pal usage in men: associations with eating disorder symptoms and psychosocial impairment. Eating Behaviors 33:13–17. doi:10.1016/j.eatbeh.2019.02.003

    • Cross-sectional, n = 122 men (fitness social media). Moderate/weak.
    • 56% had used MFP; ~40% of users said it contributed to disordered eating "to some extent"; users had higher dichotomous thinking, binge eating, restraint, impairment (large effects).

  • Messer M, McClure Z, Norton B, Smart M, Linardon J (2021). Using an app to count calories: motives, perceptions, and connections to thinness- and muscularity-oriented disordered eating. Eating Behaviors 43:101568. doi:10.1016/j.eatbeh.2021.101568

    • Cross-sectional, n = 1357 adults. Moderate (large, but self-selected, cross-sectional).
    • 71% had used a calorie app. Those counting for weight/shape reasons more often said the app fed food preoccupation, all-or-none thinking about food, food anxiety, purging. Motive matters: the same feature harms some users more than others.

  • Plateau CR, Bone S, Lanning E, Meyer C (2018). Monitoring eating and activity: links with disordered eating, compulsive exercise, and general wellbeing among young adults. Int J Eat Disord 51(11):1270–1276. doi:10.1002/eat.22966

    • Cross-sectional, n = 352 young adults. Moderate/weak.
    • Users of food/activity monitoring tools scored higher on EDE-Q and compulsive exercise; frequency of use correlated with scores; worst when tools were used for weight/shape.

  • Romano KA, Swanbrow Becker MA, Colgary CD, Magnuson A (2018). Helpful or harmful? The comparative value of self-weighing and calorie counting versus intuitive eating on the eating disorder symptomology of college students. Eat Weight Disord 23(6):841–848. doi:10.1007/s40519-018-0562-6

    • Cross-sectional, n = 902 students. Moderate (association).
    • More frequent self-weighing and calorie counting, and lower intuitive-eating scores, predicted higher ED severity.

  • Hahn SL, Kaciroti N, Eisenberg D, Weeks HM, Bauer KW, Sonneville KR (2021). Introducing dietary self-monitoring to undergraduate women via a calorie counting app has no effect on mental health or health behaviors: results from a randomized controlled trial. J Acad Nutr Diet 121(12):2377–2388. doi:10.1016/j.jand.2021.06.311

    • RCT, n = 200 women screened as low ED risk, 1 month of MyFitnessPal vs nothing. Strong for low-risk people, says nothing about at-risk people.
    • No change in ED risk, anxiety, depression, body satisfaction. The only experimental test so far; it deliberately excluded the population Sioul worries about.

  • Hahn SL, Linxwiler AN, Huynh T, Rose KL, Bauer KW, Sonneville KR (2021). Impacts of dietary self-monitoring via MyFitnessPal to undergraduate women: a qualitative study. Body Image 39:221–226. doi:10.1016/j.bodyim.2021.08.010

    • Qualitative interviews, n = 20 from the RCT above. Weak.
    • Even in low-risk women: 9/20 reported more negative feelings, 10/20 more weight/shape concern after one month of logging. "Harmful if the behavior becomes obsessive."

  • Eikey EV, Reddy MC, Booth KM, Kvasny L, Blair JL, Li V, Poole ES (2017). Desire to be underweight: exploratory study on a weight loss app community and user perceptions of the impact on disordered eating behaviors. JMIR mHealth uHealth 5(10):e150. doi:10.2196/mhealth.6683

    • Profile data (n = 18,601) + forum posts. Weak/moderate (observational).
    • 6.78% of users set underweight goals; no user started with a weight-gain goal. Users with underweight goals saw the app as helpful (less bingeing) but some described it worsening restriction.

  • Eikey EV, Reddy MC (2017). "It's definitely been a journey": a qualitative study on how women with eating disorders use weight loss apps. Proc. CHI 2017, 642–654. doi:10.1145/3025453.3025591

    • Interviews, n = 16 women with a history of eating disorders who used weight-loss apps. Weak (small qualitative).
    • Apps contributed to and worsened ED behaviours, but use and perception shifted as women moved between stages of change: the same app helped at one stage and harmed at another. Argues against a fixed "good tool / bad tool" view; the tool must let the person change how they use it over time.

  • Eikey EV (2021). Effects of diet and fitness apps on eating disorder behaviours: qualitative study. BJPsych Open 7(5):e176. doi:10.1192/bjo.2021.1011 (PMC8485346)

    • Think-aloud + interviews, n = 24 university women with self-identified ED behaviours (17 current, 7 former app users). Weak (qualitative), but the most detailed feature-level account.
    • Eight harms: fixation on numbers, rigid diet, obsession, app dependency, high sense of achievement, extreme negative emotions, motivation from "negative" messages, excess competition — also present when the user's goal was to eat more or to recover.
    • Feature-level: green/red budget colours rewarded under-eating and shamed over-eating; MyFitnessPal "Complete Diary" warnings ("you are likely not eating enough") were read as motivation to restrict "regardless of the content"; daily logging reminders and streaks fed obsession and dependency; users logged meals in advance to control intake.
    • Recommendations (Table 3): "Reduce reminders to log daily", "Encourage breaks", treat abandonment as possibly healthy, find non-numeric ways to record, warnings must not take away agency or label behaviour as "bad".

  • Honary M, Bell BT, Clinch S, Wild SE, McNaney R (2019). Understanding the role of healthy eating and fitness mobile apps in the formation of maladaptive eating and exercise behaviors in young people. JMIR mHealth uHealth 7(6):e14239. doi:10.2196/14239 (PMC6604512)

    • Survey n = 106, 2 workshops n = 8, 3 expert interviews, review of top 100 Google Play apps. Weak/moderate (mixed methods, non-clinical).
    • 41/106 reported a negative experience (most often obsessive logging: "I got obsessive over calorie counting"; "constant guilt"). Themes: guilt from persuasive design, fear of the app's response when targets are missed, feeling controlled. Meal skipping after hitting a budget early: "you could reach your calories at like 3pm and then you don't feel like you could eat anything else for that day".
    • Participants wanted encouragement "regardless of whether the daily goal was achieved". App review: 26/100 apps highlighted what the user had not done; 60/100 rewarded continuous use (streaks, trophies); some praised under-logging ("You made that look easy").

3b. HCI studies: reminders, logging, quantification, privacy

  • Stawarz K, Cox AL, Blandford A (2014). Don't forget your pill! Designing effective medication reminder apps that support users' daily routines. Proc. CHI 2014, 2269–2278. doi:10.1145/2556288.2557079
    • Review of 229 medication-reminder apps + thematic analysis of 1,012 user reviews. Weak/moderate (descriptive; abstract-level reading only, full text not retrieved).
    • Taking medication is mostly routine-driven, but apps rely on "timer-based reminders"; authors call for reminders that support the routine "and its individual nature", moving "from passive alerts to a smarter memory and routine assistant". The closest analogue to a meal reminder: a regular daily act that people forget, not avoid.

  • Stawarz K, Cox AL, Blandford A (2015). Beyond self-tracking and reminders: designing smartphone apps that support habit formation. Proc. CHI 2015, 2653–2662. doi:10.1145/2702123.2702230

    • 4-week field study of cue types + review of 115 habit apps. Weak/moderate.
    • "relying on reminders supported repetition but hindered habit development, while the use of event-based cues led to increased automaticity; positive reinforcement was ineffective." Time-based alarms make people dependent on the alarm; anchoring a meal to an event ("when I close the laptop at noon") builds the habit. Praise did not help.

  • Cordeiro F, Epstein DA, Thomaz E, Bales E, Jagannathan AK, Abowd GD, Fogarty J (2015). Barriers and negative nudges: exploring challenges in food journaling. Proc. CHI 2015, 1159–1162. doi:10.1145/2702123.2702155 (PMC4755274)

    • Survey of 141 current/lapsed food journalers + 5,526 forum posts (MyFitnessPal, FatSecret, CalorieCount). Weak/moderate (non-clinical).
    • Logging nudged people toward packaged food (easy to scan) and away from home-cooked and shared meals: "I just avoided eating things that were hard to log"; "I could make life easier by eating the same things regularly"; "It discourages you from eating out or at a friend's". Guilt and obsession: "I did not want to obsess about food"; one respondent: "it was more of an on the way to an eating disorder thing". Logging is "not always that discreet"; people avoided logging in front of friends.
  • Etkin J (2016). The hidden cost of personal quantification. J Consumer Research 42(6):967–984. doi:10.1093/jcr/ucv095
    • Six experiments. Moderate.
    • Measuring an activity increased how much people did but reduced enjoyment: "measurement can make enjoyable activities feel more like work", lowering continued engagement and well-being. Counting meals risks turning eating into a task to complete.
  • Kerner C, Goodyear VA (2017). The motivational impact of wearable healthy lifestyle technologies: a self-determination perspective on Fitbits with adolescents. Am J Health Educ 48(5):287–297. doi:10.1080/19325037.2017.1343161
    • Pre/post + focus groups, n = 84 adolescents, 8 weeks. Weak/moderate.
    • Need satisfaction and autonomous motivation fell, amotivation rose; short-term motivation came "through feelings of competition, guilt, and internal pressure".

  • Epstein DA, Ping A, Fogarty J, Munson SA (2015). A lived informatics model of personal informatics. Proc. UbiComp 2015, 731–742. doi:10.1145/2750858.2804250

    • Surveys (n = 105, 99, 83) + 22 interviews. Weak/moderate. Lapsing and resuming are normal phases of self-tracking, not failures. Basis for designing "no penalty for gaps".

  • Eikey EV (2016). Privacy and weight loss apps: a first look at how women with eating disorders use social features. Proc. GROUP 2016, 413–415. doi:10.1145/2957276.2996282

    • Preliminary qualitative results. Weak. Women with EDs rarely used social features; trade-off between support and privacy.
  • Eiband M, Khamis M, von Zezschwitz E, Hussmann H, Alt F (2017). Understanding shoulder surfing in the wild: stories from users and observers. Proc. CHI 2017, 4254–4265. doi:10.1145/3025453.3025636
    • Survey, n = 174. Weak/moderate. Shoulder surfing on phones is common, opportunistic, mostly personal content, and leaves both sides uneasy. A notification text on a lock screen or a shared desktop is read by others; its wording must be safe to be seen.

4. Lived experience, preferences, prompt design

  • Barakat S, Maguire S, Smith KE, Mason TB, Crosby RD, Touyz S (2019). Evaluating the role of digital intervention design in treatment outcomes and adherence to eTherapy programs for eating disorders: a systematic review and meta-analysis. Int J Eat Disord 52(10):1077–1094. doi:10.1002/eat.23131
    • Meta-analysis/meta-regression of 23 web-based ED self-help studies. Moderate (coarse coding scale; authors flag its limits).
    • More multimedia → better outcomes. More automated feedback (reminders, personalised feedback) was associated with results favouring the control group; no feature predicted dropout. The only quantitative synthesis found on reminder "dose" in ED e-therapy, and it points the wrong way for heavy prompting.
  • Kruger M, Liu J, Graver H, Manasse S, Velkoff EA (2025). Ecological momentary interventions for eating disorders: a systematic review of the nascent science and recommendations for future research. J Eat Disord 13(1):122. doi:10.1186/s40337-025-01329-4 (PMC12217200)
    • Systematic review, 8 studies. Moderate (few, mixed-quality studies).
    • Both RCTs found momentary prompts added nothing beyond CBT; participants raised "intrusiveness of frequent notifications" and data-entry burden. Recommends user-set notification frequency, content and channel, notifications "only ... during the patient's waking hours", and concise messages for stressed users.
  • Presseller EK, Wilkinson ML, Trainor C, Lampe EW, Juarascio AS (2022). Self-regulation deficits moderate treatment outcomes in a clinical trial evaluating just-in-time adaptive interventions as an augmentation to cognitive-behavioral therapy for bulimia-spectrum eating disorders. Int J Eat Disord 55(5):709–716. doi:10.1002/eat.23695
    • Secondary analysis of RCT, n = 55. Weak/moderate.
    • Prompts helped people with emotion-regulation difficulties or high impulsivity (fewer binges and compensatory behaviours), not the others. External cues matter most where self-regulation is weakest — plausibly ADHD, burnout, depression (extrapolation; not tested in those groups).
  • Liu J, Giannone A, Wang H, Wetherall L, Juarascio A (2024). Understanding patients' preferences for a digital intervention to prevent posttreatment deterioration for bulimia-spectrum eating disorders: user-centered design study. JMIR Form Res 8:e60865. doi:10.2196/60865 (PMC11612586)
    • Interviews, n = 12 after CBT-E. Weak.
    • 8/12 wanted push-notification reminders (mostly to use skills); "The barrier would be knowing when to send the notification." Example weekly goal they found useful: "I will keep eating regularly with 3 meals and 2 snacks every day", with "planning for snacks ahead of time" as the strategy. After therapy ends, people lose the check-ins that kept them practising; a light external structure replaces them.
  • Linardon J, Shatte A, Tepper H, Fuller-Tyszkiewicz M (2020). A survey study of attitudes toward, and preferences for, e-therapy interventions for eating disorder psychopathology. Int J Eat Disord 53(6):907–916. doi:10.1002/eat.23268
    • Survey, n = 713 across the ED-risk spectrum. Moderate. ~75% preferred face-to-face therapy, but ~50% intended to use e-therapy (70% with probable BN); e-therapy preference was linked to stigma about professional help-seeking. Discretion is part of why people choose digital tools.
  • Kim JP, Sadeh-Sharvit S, Darcy AM, et al. (2018). The utility and acceptability of a self-help smartphone application for eating disorder behaviors. J Technol Behav Sci 3(3):161–164. doi:10.1007/s41347-018-0048-4 (PMC6203337)
    • Pilot, n = 189 (64 exit surveys), 8 days. Weak. High rated acceptability of a Recovery Record self-help program with meal logs.

4b. Shame, all-or-nothing thinking, automated messages

  • Blythin SPM, Nicholson HL, Macintyre VG, Dickson JM, Fox JRE, Taylor PJ (2020). Experiences of shame and guilt in anorexia and bulimia nervosa: a systematic review. Psychol Psychother 93(1):134–159. doi:10.1111/papt.12198
    • Systematic review, 24 studies. Moderate. Shame is more common in AN and BN than in controls, rises with symptom severity, and is associated with binge/purge onset (moderate–large effects). Any wording that implies failure ("missed", "you skipped") feeds a known maintaining factor.
  • Lethbridge J, Watson HJ, Egan SJ, Street H, Nathan PR (2011). The role of perfectionism, dichotomous thinking, shape and weight overvaluation, and conditional goal setting in eating disorders. Eating Behaviors 12(3):200–206. doi:10.1016/j.eatbeh.2011.04.003
    • Cross-sectional, 238 women with EDs vs 248 community women. Moderate. Dichotomous (all-or-nothing) thinking and conditional goal setting predicted ED psychopathology beyond perfectionism. Scores, completion rates and "x of 3 meals" invite exactly this thinking.
  • Kelly AC, Carter JC (2015). Self-compassion training for binge eating disorder: a pilot randomized controlled trial. Psychol Psychother 88(3):285–303. doi:10.1111/papt.12044
    • Pilot RCT, n = 41, 3 weeks. Weak. Food planning plus self-compassion exercises cut binge days and reduced ED pathology more than food planning plus behavioural strategies; people high in fear of self-compassion benefited less. Planned eating is the common base; the tone around it matters.
  • Sharp G, Torous J, West ML (2023). Ethical challenges in AI approaches to eating disorders. J Med Internet Res 25:e50696. doi:10.2196/50696
    • Viewpoint. Opinion. Documents the 2023 case of a publicly released ED prevention chatbot ("Tessa") giving dietary advice and encouraging weight loss, and canned replies missing distress ("Keep on recognising your great qualities!" in answer to "I hate my appearance ... I don't have any friends"). Automated text about food or self-worth can harm; development needs people with lived experience.
  • Chan WW, Fitzsimmons-Craft EE, Smith AC, et al. (2022). The challenges in designing a prevention chatbot for eating disorders: observational study. JMIR Form Res 6(1):e28003. doi:10.2196/28003
    • 2,409 users, 52,129 comments reviewed. Weak/moderate. Most common problem: inappropriate responses to unanticipated user input. Fixed, short, pre-reviewed text is safer than generated text.

4c. Skipped meals, streaks, nagging, if-then cues

  • Grilo CM, Shiffman S (1994). Longitudinal investigation of the abstinence violation effect in binge eaters. J Consult Clin Psychol 62(3):611–619. doi:10.1037/0022-006X.62.3.611
    • Prospective phone interviews, n = 50 women who binge. Moderate (small, but within-person and prospective).
    • When a lapse was attributed to internal, global, uncontrollable causes ("I always fail"), the next binge came "significantly sooner"; guilt-like states after an event predicted repetition better than stable traits. A message that makes a missed meal feel like a personal failure is the mechanism to avoid.
  • Silverman J, Barasch A (2022). On or off track: how (broken) streaks affect consumer decisions. J Consumer Research 49(6):1095–1117. doi:10.1093/jcr/ucac029
    • Seven experiments. Moderate. Showing an intact streak raised later engagement relative to showing a broken one, "independent of actual past behavior" — only the representation changed. The drop was larger when people blamed themselves for the break. For meals: a streak display turns one skipped lunch into "broken", which lowers the chance of eating the next meal on plan.
  • Gray CM, Kou Y, Battles B, Hoggatt J, Toombs AL (2018). The dark (patterns) side of UX design. Proc. CHI 2018, paper 534. doi:10.1145/3173574.3174108
    • Qualitative analysis of practitioner-collected examples. Opinion/weak (taxonomy). "Nagging" = "redirection of expected functionality that persists beyond one or more interactions" (e.g. a prompt with no "no" option). Repeating a reminder after the person has dismissed it is nagging by this definition.
  • Mathur A, Acar G, Friedman MJ, Lucherini E, Mayer J, Chetty M, Narayanan A (2019). Dark patterns at scale: findings from a crawl of 11K shopping websites. Proc. ACM Hum-Comput Interact 3(CSCW):1–32. doi:10.1145/3359183 — taxonomy includes "confirmshaming" (guilting the user for declining). Moderate (large crawl; shopping, not health). "Skip lunch? Your body needs fuel!" on a dismiss button would be confirmshaming.
  • Adriaanse MA, Vinkers CDW, De Ridder DTD, Hox JJ, De Wit JBF (2011). Do implementation intentions help to eat a healthy diet? A systematic review and meta-analysis of the empirical evidence. Appetite 56(1):183–193. doi:10.1016/j.appet.2010.10.012
    • Meta-analysis, 23 studies. Moderate. If-then plans increased healthy eating (d = .51); weaker for reducing unhealthy eating (d = .29).
  • Gollwitzer PM, Sheeran P (2006). Implementation intentions and goal achievement: a meta-analysis of effects and processes. Adv Exp Soc Psychol 38:69–119. doi:10.1016/S0065-2601(06)38002-1 — meta-analysis of 94 independent tests from 63 reports: if-then plans improved goal attainment, d = 0.65, including "getting started" and protecting an ongoing activity from interference. Strong for the general technique. Together with Stawarz 2015: tie each meal to a situational cue chosen by the person ("when the 12:30 notice comes, I save and stand up"), not only to a clock alarm.

5. Preparation, transitions, and protecting meal time

5a. Interruptions, task switching, breaks

  • Adamczyk PD, Bailey BP (2004). If not now, when? The effects of interruption at different moments within task execution. Proc. CHI 2004, 271–278. doi:10.1145/985692.985727
    • Lab experiment. Moderate (well replicated line of work). Interrupting at different moments changed emotional state and how users judged the system; interruptions at subtask boundaries caused less frustration, annoyance and anxiety than mid-task ones.
  • Iqbal ST, Bailey BP (2008). Effects of intelligent notification management on users and their tasks. Proc. CHI 2008, 93–102. doi:10.1145/1357054.1357070
    • Lab studies. Moderate. Deferring notifications to breakpoints "reduces frustration and reaction time relative to delivering them immediately"; how relevant the notice is decides which breakpoint level suits it.
  • Trafton JG, Altmann EM, Brock DP, Mintz FE (2003). Preparing to resume an interrupted task: effects of prospective goal encoding and retrospective rehearsal. Int J Hum-Comput Stud 58(5):583–603. doi:10.1016/S1071-5819(03)00023-5
    • Lab experiments. Moderate. A brief warning before an interruption (the "interruption lag") let people encode where they were; they resumed the main task faster. The mechanism behind a heads-up notice before preparation time.
  • Leroy S (2009). Why is it so hard to do my work? The challenge of attention residue when switching between work tasks. Organ Behav Hum Decis Process 109(2):168–181. doi:10.1016/j.obhdp.2009.04.002
    • Experiments. Moderate. Leaving a task unfinished leaves "attention residue" on it; the next activity gets only part of one's attention. A meal started straight out of a half-finished task is eaten with the mind still at work.
  • Leroy S, Glomb TM (2018). Tasks interrupted: how anticipating time pressure on resumption of an interrupted task causes attention residue and low performance on interrupting tasks and how a "ready-to-resume" plan mitigates the effects. Organization Science 29(3):380–397. doi:10.1287/orsc.2017.1184
    • Four studies. Moderate. Expecting time pressure on return increases residue; "A ready-to-resume intervention, in which one briefly reflects on and plans one's return to the interrupted task, mitigates this effect". Directly usable: offer a one-line "where I stopped" note at the preparation notice.
  • Trougakos JP, Hideg I, Cheng BH, Beal DJ (2014). Lunch breaks unpacked: the role of autonomy as a moderator of recovery during lunch. Academy of Management Journal 57(2):405–421. doi:10.5465/amj.2011.1072
    • Multi-source experience sampling. Moderate. Effects of lunch activities on end-of-day fatigue depended on lunch-break autonomy: with more autonomy, social and even work activities switched from increasing to reducing fatigue; relaxation helped regardless. A break that feels imposed recovers less. The person must own the meal block.
  • Sianoja M, Syrek CJ, de Bloom J, Korpela K, Kinnunen U (2018). Enhancing daily well-being at work through lunchtime park walks and relaxation exercises: recovery experiences as mediators. J Occup Health Psychol 23(3):428–442. doi:10.1037/ocp0000083
    • Intervention study, n = 51 + 46, 15 minutes at lunch for 10 days. Moderate. Park walks → better afternoon concentration and less fatigue via enjoyment; relaxation → better concentration via detachment from work. Detachment during the break is the active part.

5b. ADHD

  • Nazar BP, Bernardes C, Peachey G, Sergeant J, Mattos P, Treasure J (2016). The risk of eating disorders comorbid with attention-deficit/hyperactivity disorder: a systematic review and meta-analysis. Int J Eat Disord 49(12):1045–1057. doi:10.1002/eat.22643
    • Meta-analysis, 12 + 5 studies. Strong for the association. Odds of any ED in ADHD 3.82 (AN 4.28, BN 5.71, BED 4.13); ADHD in ED 2.57. Meal prompts for "ADHD users" reach many people with an ED too.
  • Kaisari P, Dourish CT, Higgs S (2017). Attention deficit hyperactivity disorder (ADHD) and disordered eating behaviour: a systematic review and a framework for future research. Clin Psychol Rev 53:109–121. doi:10.1016/j.cpr.2017.03.002
    • Systematic review, 75 studies. Moderate. Moderate evidence for ADHD–disordered eating association, strongest for overeating; impulsivity linked to overeating and BN.
  • Ptacek R, Kuzelova H, Stefano GB, et al. (2014). Disruptive patterns of eating behaviors and associated lifestyles in males with ADHD. Med Sci Monit 20:608–613. doi:10.12659/MSM.890495
    • Case-control, 100 boys with ADHD vs 100 controls (parent interviews). Weak (children only). Markedly less adherence to a breakfast–lunch–dinner schedule; more irregular eating times (>5/day).
  • Hupfeld KE, Abagis TR, Shah P (2019). Living "in the zone": hyperfocus in adult ADHD. Atten Defic Hyperact Disord 11(2):191–208. doi:10.1007/s12402-018-0272-y
    • Two survey samples, n = 251 and 372. Moderate. Higher ADHD symptoms → more frequent hyperfocus across school/work, hobbies and screen time. Hyperfocus is the mechanism behind "forgot to eat until 4 pm"; a single alert at the moment of the meal arrives mid-hyperfocus.
  • Canela C, Buadze A, Dube A, Eich D, Liebrenz M (2017). Skills and compensation strategies in adult ADHD – a qualitative study. PLoS One 12(9):e0184964. doi:10.1371/journal.pone.0184964
    • Interviews, n = 32 adult outpatients. Weak. The most common self-made strategies were organisational: phones, synced electronic calendars, reminder apps, routines. Some deliberately avoided rigid commitments because they felt restricting — rigid schedules can backfire.
  • Kutscheidt et al. (2019) (section 1b): poorer interoceptive awareness in adult ADHD.
  • Cortese S, Holtmann M, Banaschewski T, et al. (2013). Practitioner review: current best practice in the management of adverse events during treatment with ADHD medications in children and adolescents. J Child Psychol Psychiatry 54(3):227–246. doi:10.1111/jcpp.12036 — loss of appetite is among the common adverse events of ADHD medication and is covered with management strategies; full text not read, so no specific timing advice is claimed here. Unverified detail.

5c. Burnout and depression

  • Nevanperä NJ, Hopsu L, Kuosma E, Ukkola O, Uitti J, Laitinen JH (2012). Occupational burnout, eating behavior, and weight among working women. Am J Clin Nutr 95(4):934–943. doi:10.3945/ajcn.111.014191
    • Cohort inside an RCT, n = 230 working women, 12 months. Moderate. Burnout → more emotional and uncontrolled eating, and a "hindered ability to make changes in their eating behavior". Authors: treat burnout first. Supports low-effort structure over behaviour-change demands.
  • Simmons WK, Burrows K, Avery JA, et al. (2016). Depression-related increases and decreases in appetite: dissociable patterns of aberrant activity in reward and interoceptive neurocircuitry. Am J Psychiatry 173(4):418–428. doi:10.1176/appi.ajp.2015.15020162
    • fMRI, unmedicated MDD with increased vs decreased appetite vs controls. Moderate (mechanistic). Appetite loss in depression goes with hypoactivation of the mid-insula (interoception); appetite gain with reward-region hyperactivity. In depression, hunger can be a missing or misleading cue, as in AN and ADHD.
  • Ehlers CL, Frank E, Kupfer DJ (1988). Social zeitgebers and biological rhythms: a unified approach to understanding the etiology of depression. Arch Gen Psychiatry 45(10):948–952. doi:10.1001/archpsyc.1988.01800340076012 — theory. Opinion. Loss of social time cues ("persons, social demands, or tasks that set the biological clock") destabilises biological rhythms and can trigger depressive episodes.
  • Monk TH, Flaherty JF, Frank E, Hoskinson K, Kupfer DJ (1990). The Social Rhythm Metric: an instrument to quantify the daily rhythms of life. J Nerv Ment Dis 178(2):120–126. doi:10.1097/00005053-199002000-00007; and Monk TH, Kupfer DJ, Frank E, Ritenour AM (1991). The Social Rhythm Metric (SRM): measuring daily social rhythms over 12 weeks. Psychiatry Res 36(2):195–207. doi:10.1016/0165-1781(91)90131-8
    • Instrument development, n = 50 controls; then 20 remitted depressed vs 15 controls. Weak/moderate. Social rhythms (SRM items include getting out of bed, first contact with another person, starting work, lunch, dinner, an afternoon nap and going to bed) structure the day and act as clock cues. Remitted patients' rhythms varied more day to day and were more "other person" prompted — they relied on outside cues.
  • Frank E, Kupfer DJ, Thase ME, et al. (2005). Two-year outcomes for interpersonal and social rhythm therapy in individuals with bipolar I disorder. Arch Gen Psychiatry 62(9):996–1004. doi:10.1001/archpsyc.62.9.996
    • RCT, n = 175. Strong (for bipolar I). Social-rhythm therapy in the acute phase → longer time without a new episode; "Ability to increase regularity of social rhythms ... was associated with reduced likelihood of recurrence". Regular daily anchors (meals and sleep among them) are protective in mood disorders.

6. Naps and night sleep

  • Werner-Seidler A, Wong Q, Johnston L, O'Dea B, Torok M, Christensen H (2019). Pilot evaluation of the Sleep Ninja: a smartphone application for adolescent insomnia symptoms. BMJ Open 9(5):e026502. doi:10.1136/bmjopen-2018-026502 (PMC6537982)
    • Single-arm pilot, n = 50 aged 12–16; app co-designed with young people. Weak.
    • Same mechanism as Sioul's preparation notice: "Users receive a prompt 1 hour before bed (calculated according to sleep guidelines and their wake-up time) to commence their prebed time routine" and to put devices away. Insomnia improved (d = −0.90). Users asked to change "their wind-down activity choices and the time the wind-down reminder appeared" — the lead time must be user-set.
  • Harvey AG, Dong L, Hein K, et al. (2021). A randomized controlled trial of the Transdiagnostic Intervention for Sleep and Circadian Dysfunction (TranS-C) to improve serious mental illness outcomes in a community setting. J Consult Clin Psychol 89(6):537–550. doi:10.1037/ccp0000650 (PMC9377521)
    • RCT, n = 121 adults with serious mental illness. Strong/moderate. Core module: "establishing regular sleep-wake times including learning a wind-down and wake-up routine". Improved functioning, psychiatric symptoms and sleep (d up to −0.96), reduced bedtime variability. Regular times + a wind-down buffer is a treatment, not just hygiene advice.
  • Windred DP, Burns AC, Lane JM, Saxena R, Rutter MK, Cain SW, Phillips AJK (2024). Sleep regularity is a stronger predictor of mortality risk than sleep duration: a prospective cohort study. Sleep 47(1):zsad253. doi:10.1093/sleep/zsad253
    • UK Biobank accelerometry, n = 60,977. Strong (observational). More regular sleep–wake timing → 20–48% lower all-cause mortality; regularity predicted better than duration. Supports fixed, recurring sleep blocks over duration targets.
  • Kroese FM, De Ridder DTD, Evers C, Adriaanse MA (2014). Bedtime procrastination: introducing a new area of procrastination. Front Psychol 5:611. doi:10.3389/fpsyg.2014.00611
    • Survey, n = 177. Weak/moderate. "Failing to go to bed at the intended time, while no external circumstances prevent a person from doing so"; linked to low self-regulation and to insufficient sleep and fatigue. The evening analogue of working through lunch.
  • Valshtein TJ, Oettingen G, Gollwitzer PM (2020). Using mental contrasting with implementation intentions to reduce bedtime procrastination: two randomised trials. Psychol Health 35(3):275–301. doi:10.1080/08870446.2019.1652753
    • Two RCTs, n = 383 and 221 students. Moderate. A short if-then planning exercise reduced minutes of bedtime procrastination per night. Planning the transition beats reminding of the deadline.
  • Baron KG, Abbott S, Jao N, Manalo N, Mullen R (2017). Orthosomnia: are some patients taking the quantified self too far? J Clin Sleep Med 13(2):351–354. doi:10.5664/jcsm.6472
    • Case series. Weak. Patients developed "a perfectionistic quest for the ideal sleep" from tracker data — the sleep equivalent of calorie-tracking harm. Do not score sleep.
  • Brooks A, Lack L (2006). A brief afternoon nap following nocturnal sleep restriction: which nap duration is most recuperative? Sleep 29(6):831–840. doi:10.1093/sleep/29.6.831
    • Within-subject lab study, n = 24. Moderate. 10-min nap: immediate benefits lasting up to 155 min; 20-min: benefits from 35 min after waking; 30-min: a period of impaired alertness (sleep inertia) before benefits. A nap block needs a recovery buffer after it, longer for longer naps.
  • Hilditch CJ, Dorrian J, Banks S (2017). A review of short naps and sleep inertia: do naps of 30 min or less really avoid sleep inertia and slow-wave sleep? Sleep Med 32:176–190. doi:10.1016/j.sleep.2016.12.016
    • Review. Moderate. Even ≤30-min naps can produce sleep inertia depending on prior sleep and time of day; "30 min or less" is not a guarantee.
  • Dutheil F, Danini B, Bagheri R, et al. (2021). Effects of a short daytime nap on the cognitive performance: a systematic review and meta-analysis. Int J Environ Res Public Health 18(19):10212. doi:10.3390/ijerph181910212
    • Meta-analysis, 11 lab studies, n = 381. Moderate. Afternoon naps improved performance (especially alertness) up to ~120 min after; conflicting results during the sleep-inertia period; early-afternoon naps did better.
  • Milner CE, Cote KA (2009). Benefits of napping in healthy adults: impact of nap length, time of day, age, and experience with napping. J Sleep Res 18(2):272–281. doi:10.1111/j.1365-2869.2008.00718.x — review. Moderate. Nap benefits depend on length, timing, age and napping habit.
  • Monk et al. (1990) (section 5c): the Social Rhythm Metric counts the afternoon nap and bedtime among the daily anchors, alongside meals.

Design criteria

Each criterion is one sentence; sources follow the dash. "Inference" marks criteria that no study tested directly.

Structure and timing

  1. Prompt meals by the clock at times the person sets, never by inferred hunger, because hunger and fullness are unreliable in AN, BN, ARFID, ADHD and depression and eating at planned times is the first core step of CBT-E, CBT-AR and NICE-recommended treatment. — Murphy et al. 2010; NICE NG69 1.4.7, 1.5.5; Thomas, Wons & Eddy 2018; Sivyer et al. 2020; MacDonald et al. 2024; Khalsa et al. 2015; Kutscheidt et al. 2019; Simmons et al. 2016.
  2. Default to three meals including an evening meal, make adding a snack a one-step change, and in the settings view (not in notifications) mark any planned gap longer than about four hours. — Murphy et al. 2010; Zendegui et al. 2014; Ellison et al. 2016; Zunker et al. 2011; Masheb et al. 2011.
  3. Treat preparation and eating blocks as protected time that the planner never fills with work, while letting the person shift a block for today in one action without changing the recurring rule. — Murphy et al. 2010 ("take precedence over other activities" but adjustable "to suit the patients' commitments each day"); Trougakos et al. 2014; Canela et al. 2017.
  4. Make every parameter user-set per block (time, duration, preparation time, heads-up lead time, weekdays, sound, on/off), because wanted frequency and timing varied widely and the same reminder helped one person and obstructed another. — Juarascio et al. 2015; Lindgreen, Lomborg & Clausen 2018; Kruger et al. 2025; Dennison et al. 2013; Werner-Seidler et al. 2019.
  5. Send at most two notices per block (heads-up and start) and never repeat one that was dismissed or ignored, because more automated prompting brought no benefit on average and insistent reminders erode motivation. — Barakat et al. 2019; Juarascio et al. 2023; Kruger et al. 2025; Bakker et al. 2016; Gray et al. 2018.

Preparation and transition

  1. Keep the heads-up before preparation and word it about the work, not the food ("No new big task — break in 15 min"), because advance warning lets people park a task and resume it faster. — Trafton et al. 2003; Leroy 2009; Leroy & Glomb 2018.
  2. Offer with the heads-up an optional one-line "where I stopped" note that Sioul shows again when the block ends. — Leroy & Glomb 2018; Trafton et al. 2003.
  3. If the person finishes or switches a task inside the heads-up window, deliver the notice at that breakpoint rather than mid-task. — Iqbal & Bailey 2008; Adamczyk & Bailey 2004.
  4. Keep meal lead times in minutes rather than an hour, because for people with AN the wait before a meal is itself distressing. — Khalsa et al. 2015 (one small study; inference).
  5. Let the person attach each block to an event they choose (closing the laptop, the end of a call) with the clock as fallback, because event-based cues build habits while timer reminders breed dependence on the timer. — Stawarz et al. 2015; Stawarz et al. 2014; Gollwitzer & Sheeran 2006; Adriaanse et al. 2011.

Wording

  1. State time and place in neutral, concrete words ("12:30 — lunch", "Time to start cooking") and never as a command or obligation ("you should", "you must", "don't forget to eat"). — Miller et al. 2007; Rains 2013; Carpenter 2013; Bakker et al. 2016.
  2. Never put clinical, diagnostic or health-justification words in notifications (eating disorder, binge, restrict, recovery, relapse, "your body needs fuel"). — Juarascio et al. 2015; Bakker et al. 2016; Lindgreen, Lomborg & Clausen 2018.
  3. Do not praise, reward or affirm eating (no "well done", badges, confetti, collectibles), because patients using a real app found affirmations childish and condescending, praise did not build habits, and rewards fed obsession. — Lindgreen, Lomborg & Clausen 2018; Stawarz et al. 2015; Eikey 2021.
  4. Let the person rename each block ("Break", "Kitchen", "Pause") so that the words "eat" and "meal" need not appear at all. — Lindgreen, Lomborg & Clausen 2018; Thomas et al. 2017; Eiband et al. 2017; Cordeiro et al. 2015 (inference from these).
  5. Keep each notice to one short line with at most one optional action. — Juarascio et al. 2015; Kruger et al. 2025; Bakker et al. 2016.

What never to show

  1. Never show or ask for calories, portions, weights, body measures, nutrients, food lists or food-quality words ("healthy", "junk", "clean", "good", "bad"). — Levinson et al. 2017; Simpson & Mazzeo 2017; Linardon & Messer 2019; Messer et al. 2021; Eikey 2021; Honary et al. 2019; Cordeiro et al. 2015.
  2. Never show counts, completion rates, streaks, histories, scores or red/green states for meals, naps or sleep. — Eikey 2021; Honary et al. 2019; Silverman & Barasch 2022; Lethbridge et al. 2011; Etkin 2016; Kerner & Goodyear 2017; Baron et al. 2017.
  3. Do not record whether a meal was eaten; a "done" button, if any, only clears the notice and stores nothing. — Murphy et al. 2010 (monitoring is a therapist-reviewed, time-limited tool); NICE NG69 1.3.5; Lindgreen, Lomborg & Clausen 2018; Eikey 2021; Cordeiro et al. 2015.
  4. Never generate text about food, eating, body or self-worth in this feature (no tips, no AI output); use a small fixed set of reviewed messages. — Sharp et al. 2023; Chan et al. 2022; McMaster et al. 2021 (manual review); Fairburn & Rothwell 2015.

Skips, snooze, pauses

  1. When a block passes without acknowledgement, stay silent: no "missed" message, no follow-up, no end-of-day summary. — Grilo & Shiffman 1994; Blythin et al. 2020; Honary et al. 2019; Dennison et al. 2013; Gray et al. 2018.
  2. Offer one-tap "later" (user-set delay, once) and "skip today" with no confirmation, no reason asked and no guilt text on the dismiss button. — Juarascio et al. 2015; Gray et al. 2018; Mathur et al. 2019; Lindgreen, Lomborg & Clausen 2018.
  3. Offer quick mutes (today, for a set time, while in a meeting or with others) that silence the notices but keep the protected time in the plan. — Dennison et al. 2013; Lindgreen, Lomborg & Clausen 2018; Kruger et al. 2025.
  4. Allow different times per weekday and one-off shifts, because rigid schedules felt restricting and CBT-E itself allows daily adjustment. — Murphy et al. 2010; Canela et al. 2017; Bell et al. 2026.

Privacy

  1. Make the default lock-screen and desktop notice safe to be read by someone else (block name and time only) and keep notes and custom text inside the app. — Eiband et al. 2017; Cordeiro et al. 2015; Linardon et al. 2020.
  2. Never share meal, nap or sleep status with anyone (partner, clinician, other people on synced devices) unless the person turns it on for that item. — Dennison et al. 2013; Lindgreen, Lomborg & Clausen 2018; Juarascio et al. 2015.

Naps and sleep

  1. Use the same mechanism for night sleep: bed and wake times set by the person, a wind-down notice before bed with a user-set lead time (about an hour in the cited programmes), and no sleep score. — Harvey et al. 2021; Werner-Seidler et al. 2019; Windred et al. 2024; Baron et al. 2017.
  2. For naps, plan the nap plus a recovery buffer before work resumes, because grogginess after waking can last tens of minutes and is longer after longer naps. — Brooks & Lack 2006; Hilditch et al. 2017; Dutheil et al. 2021.
  3. At setup, offer an if-then plan for the transition ("when the wind-down notice comes, I …") instead of relying on the bedtime notice alone. — Valshtein et al. 2020; Kroese et al. 2014; Gollwitzer & Sheeran 2006.

Scope and safety

  1. Present the feature as planning breaks for everyone and never label or screen the person, because many users will have an undisclosed eating disorder and ADHD carries a three- to fourfold ED risk. — Bakker et al. 2016; Tregarthen et al. 2015; Nazar et al. 2016.
  2. Never suggest foods, portions or meal content; a person following a clinician's meal plan enters its times and Sioul only keeps time. — McMaster et al. 2021 (both papers); Fairburn & Rothwell 2015; NICE NG69; Bell et al. 2026.
  3. Use the eating duration only to protect time in the plan, with no countdown or "time's up" alert during the meal. — Inference from Etkin 2016, Eikey 2021 and Khalsa et al. 2015; no direct study.

Open questions / conflicts

  1. Self-monitoring. Recording meals helps inside treatment (regular-meal adherence mediated outcomes in Hildebrandt et al. 2017; CBT-E and NICE use it with a practitioner) but is linked to harm as general tracking (Levinson et al. 2017; Eikey 2021; Lindgreen, Lomborg & Clausen 2018). The only RCT (Hahn et al. 2021) found no harm in women screened as low risk and excluded those at risk. No study tested a minimal "done" tap with no history; criterion 18 takes the cautious side.
  2. Praise. People imagining an app asked for badges and positive messages (Juarascio et al. 2015; Honary et al. 2019; Dennison et al. 2013); people using a real app found affirmations condescending (Lindgreen, Lomborg & Clausen 2018), praise did not build habits (Stawarz et al. 2015) and rewards fed obsession (Eikey 2021). Real use outweighs hypothetical preference here, but all of these are small qualitative studies.
  3. Forgetting versus avoiding. Reminders suit people who forget (hyperfocus, burnout, low appetite) and misread people who avoid eating (AN: "I sometimes don't want to"). No study tested one design for both groups; neutral time-only wording is the least-bad guess.
  4. Skipped meals: silence or a gentle check-in? No study compared them. Shame and lapse research (Blythin et al. 2020; Grilo & Shiffman 1994) argues against any "missed" message; restriction-to-binge research (Zunker et al. 2011) says long gaps matter. Letting the next block's heads-up arrive as normal is a middle path that nobody has tested.
  5. Firmness of the default. CBT-E asks meals to "take precedence over other activities"; routine-based eating correlated with impairment in community women (Bell et al. 2026); adults with ADHD avoid rigid commitments (Canela et al. 2017); autonomy decides whether a lunch break restores (Trougakos et al. 2014). How firmly the planner should defend meal blocks against work is a design choice the evidence does not settle.
  6. Lead time. Support for advance warnings comes from interruption research in healthy adults (Trafton et al. 2003; Leroy & Glomb 2018) and from sleep programmes (one hour before bed). Fifteen minutes before preparation is plausible but untested for meals; long waits may raise anxiety in AN (Khalsa et al. 2015, n = 15).
  7. Prompt dose. On average more prompting adds nothing (Barakat et al. 2019; Juarascio et al. 2023; Kruger et al. 2025), yet just-in-time prompts helped people with high impulsivity or emotion-regulation difficulties (Presseller et al. 2022). A fixed two-notice design may under-serve some people with ADHD; user settings are the hedge.
  8. Mentioning food. No study compared food-free and food wording in notifications; criterion 14 rests on privacy (Eiband et al. 2017), illness reminders (Lindgreen, Lomborg & Clausen 2018), the ARFID fear presentation (Thomas et al. 2017) and stigma (Linardon et al. 2020).
  9. Who was studied. Most evidence comes from women, students and BN/BED samples; little covers men, adults with ARFID or OSFED. No study found tested meal reminders for adults with ADHD, burnout or depression; those criteria rest on mechanism studies (interoception, hyperfocus, self-regulation, social rhythms).
  10. Sleep and naps. Regular timing predicted mortality better than duration, but observationally (Windred et al. 2024). Nap benefits and inertia depend on prior sleep and time of day (Hilditch et al. 2017), so no single buffer length is right; the one-hour wind-down comes from treatment programmes, not from trials comparing lead times.
  11. Strength overall. The strong evidence (NICE, CBT-E trials, Bauer et al. 2012, Harvey et al. 2021) supports regular eating, regular sleep and light-touch contact; the design-specific evidence (Lindgreen, Lomborg & Clausen 2018; Eikey 2021; Juarascio et al. 2015) is qualitative, with 11–41 participants each.

Where Sioul stands (5 October 2026)

  • Built: meals (how many, at what time, how long, preparation time, which days), naps and the night (bedtime, wake time, wind-down), each set by you and kept free in the plan, which lays the work in what is left; a heads-up a quarter of an hour before ("no new big task") and one notice at the time, each once, none during an event; "15 min later" (as often as you like), "Move to…" a time, "Not today", today only, without a word asked, from the notice or from today's list on the Health page; a meal that would fall in an event pushed past it, the time to come back from it counted; one line on where you stopped at each interruption, shown when you are back. Nothing about food, portions or numbers recorded or shown; no "missed", no praise.
  • Not built yet: the heads-up at a breakpoint inside its window (criterion 8), a block tied to an event rather than the clock (10), an if-then plan for the transition (28).