Sioul: a global pause to take the good moment¶

Research report, started 2026-10-06 13:09, finished 14:31. It builds on capacity-budget.md (cited "CB" with its entry or criterion number), wellbeing-gain.md ("G"), done-for-the-day.md ("DFD"), time-estimation.md ("TE") and meal-prompts.md; what they already cover is pointed to, not repeated. A separate note on an emergency pause (a button for overwhelm) is written at the same time by another agent: emergency-pause.md.

The idea researched (not built). One state in the bottom toolbar, a global pause, puts everything on hold until the person says otherwise, and switches to Leisure, where only the safe list (the people the person chose) may reach them, at any hour. The example: the weather is nice now and will be bad in the evening, so the person goes out to exercise now and moves work to the evening; Sioul's weather applet was made for that. If the pause falls in working hours, the end of the work day moves later by as much. During the pause, tasks for leisure (exercise, art, creativity, massage, well-being) are allowed.

Questions.

1. Taking the good moment: weather, daylight and time outdoors against mood, energy, sleep and cognition; what a short outdoor break gives; the limits in energy-limiting illness.
2. Autonomy over when work happens: job control, flexible schedules, self-chosen breaks, boundary management, neurodivergent workers.
3. Moving work to the evening: the costs to detachment, recovery and sleep; chronotype; how late is too late; guilt and overwork from making up time.
4. Delaying the end of the day by the pause's length: extend, drop or re-plan; caps; what no longer fits; showing it without pressure.
5. What is allowed during the pause: leisure kinds as recovery (DRAMMA); interruptions and notifications during leisure.
6. Prior art: pause, "take the afternoon" and focus modes in planners and operating systems; whether flexible time raises total working time.
7. How the global pause differs from the emergency pause, in purpose and design.

Method. Peer-reviewed studies, meta-analyses and trials first, then guidelines, then lived experience (marked), then opinion (marked). Every DOI was resolved on 2026-10-06 (title, authors, year, venue), on Crossref unless the entry names PubMed or OpenAlex; abstracts were read from OpenAlex, PubMed, Semantic Scholar or the repository named in the entry; "verified (Crossref)" or "verified (PubMed n)" says which, and the entry says when a finding comes from a summary rather than the abstract. "Not checked" marks what could not be confirmed at a primary source. Evidence levels as in the capacity report: strong (meta-analysis, large preregistered or multi-site study, guideline with a systematic review), moderate (several consistent studies, or one large well-run study), weak (one small study, cross-sectional only), opinion (essay, consensus without data, vendor documentation).

Answers so far¶

(Updated 2026-10-06, final. Numbers point to entries below; "GP" to the design criteria.)

1. Taking the good moment: yes, by going out into it, briefly, at a self-chosen pace; never as exercise for energy-limiting illness.

• Weather alone moves daily mood little on average, and people differ: some feel worse in heat and sun (1.2). What helps is going out: pleasant spring weather went with better mood and memory only in people who spent more than 30 minutes outside (1.1).
• Short doses work, with diminishing returns: the largest self-esteem gain at five minutes of green exercise (d = 0.46 self-esteem, 0.54 mood overall), cortisol falling fastest per minute at 20–30 minutes in nature, lunch walks raising restoration and lowering fatigue the same afternoon (d = 0.22–0.58 in autumn), carried by enjoyment and detachment (1.3, 1.5). Nature adds enjoyment more than a proven effect over indoor exercise (1.4).
• Daylight outdoors is at least ten times brighter than indoor light; each extra daily hour outdoors went with less low mood and depression, easier waking, less tiredness and fewer insomnia symptoms in about 400,000 adults (1.6).
• Movement lowers depression (walking or jogging g −0.62, low-certainty trials) and anxiety (−0.58), and lifted energy and motivation in men with ADHD symptoms; pleasure falls above a moderate, self-chosen pace (1.7). Bad weather is a barrier to activity, so a good window is a real opportunity (1.7).
• In ME/CFS and long COVID, exertion can bring post-exertional malaise starting within a day and peaking within three: no exercise suggested, the pause's costs counted (1.8).

2. Autonomy over when work happens helps when the person holds it, and when it is bounded.

• Self-scheduling improved sleep, alertness and mental health; a randomised workplace trial raised schedule control, lowered burnout and stress, added sleep, and did not add hours. Flexibility imposed for the employer's sake harmed, and very loose flextime did less than moderate flextime (2.1).
• Self-chosen breaks restore more than imposed ones (2.2).
• Letting work into the evening suits people who like to mix work and life, under low demands; it harms people who keep them apart and people with little control over their boundaries, through lost recovery (2.3).
• Neurodivergent workers ask for flexible hours, but a late body clock (ADHD) and hyperfocus (ADHD, autism) make evening extensions risky without a firm end (2.4).

3. Moving work to the evening has real costs, mostly to the night.

• Evening work lowers detachment, restoring activities and sleep, most on days of heavy workload and time pressure (3.1).
• The sleep cost comes mainly from a mind still at work: demanding tasks or worry before bed lengthened sleep onset and lowered sleep efficiency in experiments. Screen light matters less than feared in adults, and a bright day may blunt it. No study gives a latest hour for work: the light consensus asks for dim light from three hours before bed, and the arousal studies argue for low-arousal time before the wind-down (3.2, 1.6).
• Late types do moved work as well or better in the evening, early types worse; but evening work and light push a late clock later, and eveningness goes with low mood (3.3).
• "Making up" time risks guilt, an over-full evening (people underestimate tasks), and the pause's gain lost to catch-up demands, as holiday gains fade faster under demands (3.4).

4. Delaying the end of the day: move it by the pause's length, within limits Sioul keeps (inference; no study tests it).

• The extension adds hours, never capacity (GP9). Never into the wind-down; a buffer of low-arousal time before it (guess: one hour); a latest end (guess: two hours after the usual end, earlier for morning people); the evening's "time for you" slot kept (GP10). Only light, non-anxious steps after the usual end, early in the evening, planned lighter (GP12–GP13).
• What no longer fits goes to the next days without a count; a dated step that can no longer fit is said early (GP10, GP14). One key keeps the usual end, no reason asked (GP11).
• Shown in one sentence ("today it runs until 19:00"), never as a countdown, hours owed or steps moved (GP16).

5. During the pause: chosen leisure, and only the chosen people.

• Movement, creative work, massage and body care, nature and absorbing interests each have evidence as recovery, by different routes; they stay offers, never a list to finish (5.1–5.2).
• The safe list only, quietly, fits the evidence better than everyone or no one: interruptions cut attention and enjoyment, total silence made some people anxious. Doses and codes come as always; "nothing at all" as a choice (5.3).
• A line on where the person stopped, and a clear time for work's return, let work be left behind (5.4).

6. Prior art: no product pauses work and moves the end of the day; flexible time often grows.

• iOS and Android focus modes filter people and apps; Motion, Reclaim and Structured re-lay the day when hours change; Sunsama carries tasks forward. None has an outcome study (6.1–6.2).
• Work from home was mostly overtime in a US panel; workdays grew by about 48.5 minutes after the 2020 lockdowns; a third peak of work at 21:00 appeared in developers. Hours did not grow in a trial where control was held and the organisation of the day kept (6.3).

7. The emergency pause is a different thing. The global pause is proactive recovery by a person able to choose: it may ask one choice, and it moves work. The emergency pause is reactive, for states in which choosing and remembering fail: it asks nothing, moves nothing and owes nothing. Different names, places and looks, a status line naming the pause, and one key turning a global pause into one that moves nothing prevent each from being taken for the other (7.1–7.4, GP18–GP20).

Main uncertainties: everything about the extension is inference; the buffer and the latest end are guesses; light against arousal in adults; how often an extension can be used before it shifts the body clock; the right default for people who mix or separate work and life; whether an opt-in weather line helps people with low initiation or pressures them (open questions 1–15).

1. Taking the good moment: weather, daylight, time outdoors¶

Already covered, not repeated: nature as a weekly dose with a plateau (G 2.9, White et al. 2019: 120 minutes a week or more, in any split), movement and affect in daily life (G 2.8, Liao et al. 2015), restorative environments (CB 6.10), pacing and post-exertional malaise (CB 2.6 NICE NG206, 2.8 Chu 2018, 2.9, 6.13).

1.1 Pleasant weather lifts mood only when people are out in it. Keller MC, Fredrickson BL, Ybarra O, Côté S, Johnson K, Mikels J, Conway A, Wager T. 2005. A warm heart and a clear head: the contingent effects of weather on mood and cognition. Psychol Sci 16(9):724–731. doi:10.1111/j.1467-9280.2005.01602.x — verified (Crossref; abstract read on OpenAlex).

• Two correlational studies and one experiment that assigned time outdoors (N = 605 in all): in spring, pleasant weather (higher temperature or barometric pressure) went with better mood, better memory and a broader cognitive style as time spent outside increased. A summary by the British Psychological Society gives the threshold: the benefit appeared only in people who had spent more than 30 minutes outside that day, and for those who had spent less, nicer weather went with poorer mood and cognition, perhaps from resenting being "cooped up indoors" (https://www.bps.org.uk/research-digest/pleasant-thing-it-eyes-behold-sun-ecclesiastes-114). Outside spring the link was absent, and in summer hotter weather went with lower mood. The authors read it as relief after a winter's deprivation.
• Three studies, one experimental. Moderate.
• For the pause: the gain of a fine afternoon is in going out into it, not in the weather itself; a fine day spent indoors may even feel worse. It supports a pause offered when the weather is good, and it says the offer is worth most after a run of bad days and least in summer heat.

1.2 On average weather moves daily mood little, and people differ a lot. Denissen JJA, Butalid L, Penke L, van Aken MAG. 2008. The effects of weather on daily mood: a multilevel approach. Emotion 8(5):662–667. doi:10.1037/a0013497 — verified (Crossref; abstract read). Klimstra TA, Frijns T, Keijsers L, Denissen JJA, Raaijmakers QAW, van Aken MAG, Koot HM, van Lier PAC, Meeus WHJ. 2011. Come rain or come shine: individual differences in how weather affects mood. Emotion 11(6):1495–1499. doi:10.1037/a0024649 — verified (Crossref; abstract read). Lucas RE, Lawless NM. 2013. Does life seem better on a sunny day? Examining the association between daily weather conditions and life satisfaction judgments. J Pers Soc Psychol 104(5):872–884. doi:10.1037/a0032124 — verified (Crossref; abstract read).

• Online diaries of 1233 people linked to weather stations: temperature, wind and sunlight had small average effects on negative affect, and sunlight lowered tiredness; individual sensitivity varied significantly and was not explained by personality, gender or age. Thirty days of diaries in 497 adolescents, replicated in their mothers: four types, "summer lovers" (better mood when warm and sunny), "unaffected", "summer haters" (worse mood when warm and sunny) and "rain haters"; secondary summaries give about 17 %, 48 %, 27 % and 9 % (not checked in the paper). Over a million Americans: daily weather did not reliably affect life satisfaction.
• Large diary studies and a very large survey. Strong for "small on average"; moderate for the types.
• For the pause: Sioul cannot assume that sun is good for this person. Some feel worse in heat and bright sun. The weather can inform an offer the person has asked for; it should not decide that now is the good moment.

1.3 Short doses of nature give most of their effect early. Barton J, Pretty J. 2010. What is the best dose of nature and green exercise for improving mental health? A multi-study analysis. Environ Sci Technol 44(10):3947–3955. doi:10.1021/es903183r — verified (Crossref; abstract read in the Essex repository, https://repository.essex.ac.uk/683/). Hunter MR, Gillespie BW, Chen SY-P. 2019. Urban nature experiences reduce stress in the context of daily life based on salivary biomarkers. Front Psychol 10:722. doi:10.3389/fpsyg.2019.00722 — verified (Crossref; abstract read). McMahan EA, Estes D. 2015. The effect of contact with natural environments on positive and negative affect: a meta-analysis. J Posit Psychol 10(6):507–519. doi:10.1080/17439760.2014.994224 — verified (Crossref; abstract read). Bratman GN, Hamilton JP, Hahn KS, Daily GC, Gross JJ. 2015. Nature experience reduces rumination and subgenual prefrontal cortex activation. PNAS 112(28):8567–8572. doi:10.1073/pnas.1510459112 — verified (Crossref; abstract read).

• Barton & Pretty pooled ten of their UK studies (1252 people; walking, gardening, cycling, fishing, boating, riding, farming): self-esteem rose (d = 0.46) and mood improved (d = 0.54); the dose-response curves showed "large benefits from short engagements in green exercise, and then diminishing but still positive returns"; "the largest positive effect on self-esteem came from a five-minute dose" (the authors' press release, https://www.sciencedaily.com/releases/2010/05/100502080414.htm); every green setting helped, water more; the young and people with mental illness gained most.
• Hunter et al.: 36 city dwellers over eight weeks chose when, where and how long to spend at least ten minutes in nature, three times a week. Cortisol fell 21.3 % an hour beyond its daily decline; the benefit per minute was greatest between 20 and 30 minutes, then kept accruing more slowly.
• McMahan & Estes: 32 studies, 2356 people: a moderate rise in positive affect and a smaller, consistent fall in negative affect after brief contact with nature, with high heterogeneity.
• Bratman et al.: 38 young adults randomised to a 90-minute walk in nature or in a city: rumination and activity in a brain region tied to it fell after the nature walk only.
• Multi-study analysis without control groups (moderate for the shape of the curve, weak for the five-minute peak); a small naturalistic study (weak to moderate; its strength is that people chose their own time); a meta-analysis of short studies (moderate); one small randomised study (weak to moderate).
• For the pause: a short time outside, from a few minutes to half an hour, already gives a measurable effect, and each further minute adds less; a pause need not take the afternoon to be worth it. Those who gained most include people with mental illness, and nature lowered rumination, a common thread of anxiety and depression. Freely chosen timing (Hunter) works.

1.4 Outdoors or indoors: the movement may matter more than the place. Thompson Coon J, Boddy K, Stein K, Whear R, Barton J, Depledge MH. 2011. Does participating in physical activity in outdoor natural environments have a greater effect on physical and mental wellbeing than physical activity indoors? A systematic review. Environ Sci Technol 45(5):1761–1772. doi:10.1021/es102947t — verified (Crossref; abstract read). Lahart I, Darcy P, Gidlow C, Calogiuri G. 2019. The effects of green exercise on physical and mental wellbeing: a systematic review. Int J Environ Res Public Health 16(8):1352. doi:10.3390/ijerph16081352 — verified (Crossref; abstract read).

• Thompson Coon: 11 trials, 833 adults, mostly students, each a single walk or run indoors and outdoors: outdoors brought more revitalisation, positive engagement and energy, less tension, confusion, anger and depression, more enjoyment and a greater intent to repeat; calmness may fall. Lahart, updating it with 28 trials: outdoor green exercise may favour affective valence and enjoyment, but the only pooled difference was slightly lower perceived exertion; high risk of bias; "limited evidence" that green exercise beats exercise without nature.
• Systematic reviews of low-quality trials. Weak for an added effect of nature.
• For the pause: the outdoors adds enjoyment and the wish to do it again, which matters because it decides whether it happens at all; but most of the effect on mood comes from moving and from the break. A rainy-day version (indoor movement, a bath, music) is not worthless, and the pause should not be framed as "only worth it outside".

1.5 A midday walk: small, same-afternoon effects, through enjoyment and detachment. de Bloom et al. 2017 (G 3.7: 15 minutes a day for ten days, restoration up and fatigue down in the afternoon, d = 0.22–0.58 in autumn, hardly any effect in spring). 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 — verified (Crossref; abstract read). Thøgersen-Ntoumani C, Loughren EA, Kinnafick FE, Taylor IM, Duda JL, Fox KR. 2015. Changes in work affect in response to lunchtime walking in previously physically inactive employees: a randomized trial. Scand J Med Sci Sports 25(6):778–787. doi:10.1111/sms.12398 — verified (Crossref; abstract read).

• Sianoja: on park-walk days, concentration was better and fatigue lower at the end of the work day, carried by the enjoyment of the walk; relaxation exercises worked through detachment. Thøgersen-Ntoumani: 56 inactive employees, three 30-minute group walks a week for ten weeks: more enthusiasm and relaxation, less nervousness at work in the afternoon.
• Small randomised trials with diary measures. Moderate for "small, short-lived, carried by enjoyment and detachment".
• For the pause: a pause pays back mostly the same afternoon, and only if it is enjoyed and work is left behind during it. A pause spent checking mail on the phone loses what it was for (section 5). Conflict: Keller found weather lifts mood most in spring, de Bloom found walks helped least in spring; seasons change both what people already do outdoors and what is gained.

1.6 Daylight: the strongest light most people get, good for mood, sleep and the clock. Wirz-Justice A, Skene DJ, Münch M. 2021. The relevance of daylight for humans. Biochem Pharmacol 191:114304. doi:10.1016/j.bcp.2020.114304 — verified (Crossref; abstract read). Brown TM, Brainard GC, Cajochen C, et al. 2022. Recommendations for daytime, evening, and nighttime indoor light exposure to best support physiology, sleep, and wakefulness in healthy adults. PLoS Biol 20(3):e3001571. doi:10.1371/journal.pbio.3001571 — verified (Crossref; recommendations read on the publisher's page). Burns AC, Saxena R, Vetter C, Phillips AJK, Lane JM, Cain SW. 2021. Time spent in outdoor light is associated with mood, sleep, and circadian rhythm-related outcomes: a cross-sectional and longitudinal study in over 400,000 UK Biobank participants. J Affect Disord 295:347–352. doi:10.1016/j.jad.2021.08.056 — verified (PubMed 34488088; abstract read). Burns AC, Windred DP, Rutter MK, et al. 2023. Day and night light exposure are associated with psychiatric disorders: an objective light study in >85,000 people. Nat Ment Health 1(11):853–862. doi:10.1038/s44220-023-00135-8 — verified (Crossref; abstract read). Wright KP Jr, McHill AW, Birks BR, Griffin BR, Rusterholz T, Chinoy ED. 2013. Entrainment of the human circadian clock to the natural light-dark cycle. Curr Biol 23(16):1554–1558. doi:10.1016/j.cub.2013.06.039 — verified (PubMed 23910656; abstract read). Hébert M, Martin SK, Lee C, Eastman CI. 2002. The effects of prior light history on the suppression of melatonin by light in humans. J Pineal Res 33(4):198–203. doi:10.1034/j.1600-079X.2002.01885.x — verified (Crossref; abstract read). Golden RN, Gaynes BN, Ekstrom RD, et al. 2005. The efficacy of light therapy in the treatment of mood disorders: a review and meta-analysis of the evidence. Am J Psychiatry 162(4):656–662. doi:10.1176/appi.ajp.162.4.656 — verified (Crossref; abstract read).

• Daylight reaches the brain's clock and mood circuits through melanopsin cells of the retina; light acts like a drug on mood, alertness and sleep (Wirz-Justice). The expert consensus asks for at least 250 lux (melanopic) at the eye through the day, at most 10 lux from three hours before bed, at most 1 lux while sleeping, and notes that indoor lighting is usually at least ten times dimmer than outdoors (Brown).
• UK Biobank, about 400,000 adults spending a median 2.5 hours a day outdoors in daylight: each extra hour went with lower odds of depression and low mood, more happiness, easier waking (OR 1.46–1.49), less tiredness (OR 0.80–0.82), fewer insomnia symptoms and an earlier chronotype; longitudinal links were in the same direction, smaller. With wrist light sensors on 86,772 people: more light at night went with more depression, anxiety, psychosis, bipolar disorder and self-harm; more light by day, independently, with less depression, PTSD, psychosis and self-harm.
• A week of camping in natural light only moved the internal night to start at sunset; late chronotypes moved most (Wright). After a week with more daylight (4.3 hours a day), evening light suppressed melatonin less than after a week in dim light (12 people, large individual differences; Hébert).
• Bright light treats seasonal depression (effect size 0.84, eight trials) and non-seasonal depression (0.53, three trials) about as well as antidepressants in most trials (Golden).
• Review and consensus (moderate; consensus is opinion backed by mechanism studies); very large observational studies (moderate: associations, not causes); small experiments (weak); meta-analysis of few trials (moderate).
• For the pause: going out by day is one of the few things that helps mood, sleep and the clock at once, and daylight is rarely reached indoors. Daytime light may also make evening light less disruptive (Hébert), but that is one small study and no licence for evening screens (section 3).

1.7 Movement: a reliable lift, best at a self-chosen, moderate pace. Reed J, Ones DS. 2006. The effect of acute aerobic exercise on positive activated affect: a meta-analysis. Psychol Sport Exerc 7(5):477–514. doi:10.1016/j.psychsport.2005.11.003 — verified (OpenAlex: title, year and DOI; Crossref did not answer; findings from summaries, abstract not read). Ekkekakis P, Parfitt G, Petruzzello SJ. 2011. The pleasure and displeasure people feel when they exercise at different intensities: decennial update and progress towards a tripartite rationale for exercise intensity prescription. Sports Med 41(8):641–671. doi:10.2165/11590680-000000000-00000 — verified (PubMed 21780850; abstract read). Noetel M, Sanders T, Gallardo-Gómez D, et al. 2024. Effect of exercise for depression: systematic review and network meta-analysis of randomised controlled trials. BMJ 384:e075847. doi:10.1136/bmj-2023-075847 — verified (Crossref; abstract read). Pearce M, Garcia L, Abbas A, et al. 2022. Association between physical activity and risk of depression: a systematic review and meta-analysis. JAMA Psychiatry 79(6):550–559. doi:10.1001/jamapsychiatry.2022.0609 — verified (Crossref; abstract read). Singh B, Olds T, Curtis R, et al. 2023. Effectiveness of physical activity interventions for improving depression, anxiety and distress: an overview of systematic reviews. Br J Sports Med 57(18):1203–1209. doi:10.1136/bjsports-2022-106195 — verified (Crossref; abstract read). Stubbs B, Vancampfort D, Rosenbaum S, et al. 2017. An examination of the anxiolytic effects of exercise for people with anxiety and stress-related disorders: a meta-analysis. Psychiatry Res 249:102–108. doi:10.1016/j.psychres.2016.12.020 — verified (PubMed 28088704; abstract read). Fritz KM, O'Connor PJ. 2016. Acute exercise improves mood and motivation in young men with ADHD symptoms. Med Sci Sports Exerc 48(6):1153–1160. doi:10.1249/MSS.0000000000000864 — verified (Crossref; abstract read). Xu S, Zhao C, Hu L. 2026. The effects of acute and chronic exercise on executive functions and core symptoms in adults with ADHD: a systematic review and meta-analysis. Psychol Sport Exerc 84:103088. doi:10.1016/j.psychsport.2026.103088 — verified (Crossref; abstract not read, findings from https://adhdevidence.org's summary). Tucker P, Gilliland J. 2007. The effect of season and weather on physical activity: a systematic review. Public Health 121(12):909–922. doi:10.1016/j.puhe.2007.04.009 — verified (Crossref, PubMed 17920646).

• One bout: 158 studies, 13,101 people: aerobic exercise raised energetic positive feelings (corrected d = 0.47), most at low intensity, for bouts up to about 35 minutes, and in people who started lowest; the lift lasted at least 30 minutes (Reed & Ones, as summarised). Pleasure falls mainly above the ventilatory threshold; below it most people feel better; self-selected intensity is better tolerated than imposed intensity (Ekkekakis).
• Depression: 218 trials, 14,170 people: walking or jogging (g −0.62), yoga (−0.55), strength training (−0.49) reduced depression against active controls, more when more intense; confidence low to very low because of bias (Noetel). Fifteen cohorts, 191,130 adults: half the recommended weekly activity went with 18 % lower risk of depression, the full amount 25 %, the curve steepest at low volumes (Pearce). An umbrella of 97 reviews: medium effects on depression (−0.43), anxiety (−0.42) and distress (−0.60), mostly from reviews of critically low quality (Singh). Anxiety disorders: six trials, 262 adults, standardised mean difference −0.58 (Stubbs).
• ADHD: 32 young men with ADHD symptoms, 20 minutes of moderate cycling against 20 minutes of rest: more vigour and motivation for mental work, less fatigue, confusion and depression, no change in attention or restlessness (Fritz & O'Connor). A 2026 meta-analysis reports moderate gains in inhibitory control after single bouts and small, consistent gains in core symptoms, from four small studies each (Xu et al., as summarised).
• Weather: across 37 studies (291,883 people), activity varies with the season and poor or extreme weather is a barrier; rain lowers activity (Tucker & Gilliland).
• Meta-analyses of mostly low-quality trials (moderate overall, consistent in direction); large cohort meta-analysis (moderate); single small trials in ADHD (weak).
• For the pause: movement is among the best-supported uses of a free hour for mood, anxiety and the motivation to work afterwards, and even small amounts count. Good weather removes a barrier that bad weather puts back, which is the factual basis of the owner's example. The pace should be the person's own: prescribed intensity costs pleasure.

1.8 When movement harms: post-exertional malaise. NICE NG206 (CB 2.6: no graded exercise; all activity, cognitive, physical, emotional and social, spends energy; PEM delayed by hours or days). Chu 2018 (CB 2.8). Stussman B, Williams A, Snow J, Gavin A, Scott R, Nath A, Walitt B. 2020. Characterization of post-exertional malaise in patients with myalgic encephalomyelitis/chronic fatigue syndrome. Front Neurol 11:1025. doi:10.3389/fneur.2020.01025 — verified (Crossref; abstract read). Appelman B, Charlton BT, Goulding RP, et al. 2024. Muscle abnormalities worsen after post-exertional malaise in long COVID. Nat Commun 15:17. doi:10.1038/s41467-023-44432-3 — verified (Crossref; abstract read).

• Focus groups of 43 people with ME/CFS: PEM varied greatly between people; after an exercise test, 17 of 18 had symptoms start within 24 hours and peak within 72. In long COVID, an exercise test that brought on PEM was followed by worse muscle damage, metabolic disturbance and amyloid deposits in muscle.
• Guideline (strong as a practice standard, weak-to-moderate as evidence); qualitative study (weak, but the best description of timing); case-control study (moderate for mechanism, small).
• For the pause: for a person with energy-limiting illness, a fine afternoon is not a reason to exercise. Sioul must never suggest exercise to them (NICE: no advice to "exercise more" outside a specialist programme), the pause's own activity counts against the day's budgets (G24), and a crash may come one to three days later (CB criterion 20). Sitting outside in daylight may be the right use of the same moment. Heat can be a cost too (1.1: lower mood in summer heat; heat intolerance in some conditions, not researched here).

2. Autonomy over when work happens¶

Already covered, not repeated: autonomy decides whether a lunch break restores (CB 6.9, Trougakos et al. 2014), breaks that are preferred and early restore more (CB 4.11, Hunter & Wu 2016), prompted against self-scheduled well-being breaks (G 3.6, Howe et al. 2022), segmentation norms, evening work messages and telepressure (DFD §2: Park, Fritz & Jex 2011; Derks et al. 2014; Butts et al. 2015; Becker et al. 2021; Barber & Santuzzi 2015).

2.1 Control over one's own working time: better balance and sleep, held by the worker. Karasek RA. 1979. Job demands, job decision latitude, and mental strain: implications for job redesign. Adm Sci Q 24(2):285–308. doi:10.2307/2392498 — verified (Crossref). Häusser JA, Mojzisch A, Niesel M, Schulz-Hardt S. 2010. Ten years on: a review of recent research on the Job Demand–Control (-Support) model and psychological well-being. Work Stress 24(1):1–35. doi:10.1080/02678371003683747 — verified (Crossref; abstract read). Nijp HH, Beckers DGJ, Geurts SAE, Tucker P, Kompier MAJ. 2012. Systematic review on the association between employee worktime control and work–non-work balance, health and well-being, and job-related outcomes. Scand J Work Environ Health 38(4):299–313. doi:10.5271/sjweh.3307 — verified (Crossref; abstract read). Joyce K, Pabayo R, Critchley JA, Bambra C. 2010. Flexible working conditions and their effects on employee health and wellbeing. Cochrane Database Syst Rev (2):CD008009. doi:10.1002/14651858.CD008009.pub2 — verified (Crossref; abstract read). Kelly EL, Moen P, Oakes JM, et al. 2014. Changing work and work-family conflict: evidence from the Work, Family, and Health Network. Am Sociol Rev 79(3):485–516. doi:10.1177/0003122414531435 — verified (Crossref; abstract read). Moen P, Kelly EL, Fan W, et al. 2016. Does a flexibility/support organizational initiative improve high-tech employees' well-being? Evidence from the Work, Family, and Health Network. Am Sociol Rev 81(1):134–164. doi:10.1177/0003122415622391 — verified (Crossref; abstract read). Olson R, Crain TL, Bodner TE, et al. 2015. A workplace intervention improves sleep: results from the randomized controlled Work, Family, and Health Study. Sleep Health 1(1):55–65. doi:10.1016/j.sleh.2014.11.003 — verified (PubMed 29073416; abstract read). Moen P, Kelly EL, Tranby E, Huang Q. 2011. Changing work, changing health: can real work-time flexibility promote health behaviors and well-being? J Health Soc Behav 52(4):404–429. doi:10.1177/0022146511418979 — verified (Crossref; abstract read). Garde AH, Albertsen K, Nabe-Nielsen K, et al. 2012. Implementation of self-rostering (the PRIO-project): effects on working hours, recovery, and health. Scand J Work Environ Health 38(4):314–326. doi:10.5271/sjweh.3306 — verified (Crossref; abstract read). Baltes BB, Briggs TE, Huff JW, Wright JA, Neuman GA. 1999. Flexible and compressed workweek schedules: a meta-analysis of their effects on work-related criteria. J Appl Psychol 84(4):496–513. doi:10.1037/0021-9010.84.4.496 — verified (Crossref; abstract read). Nijp HH, Beckers DGJ, van de Voorde K, Geurts SAE, Kompier MAJ. 2016. Effects of new ways of working on work hours and work location, health and job-related outcomes. Chronobiol Int 33(6):604–618. doi:10.3109/07420528.2016.1167731 — verified (Crossref; abstract read).

• Control at work goes with well-being, added to the effect of demands, in almost every large enough study; longitudinal support is weaker, and control rarely "buffers" demands unless it matches them (Karasek; Häusser, 83 studies).
• Control over working time specifically: moderately strong evidence, from surveys and intervention studies, that it improves the balance between work and the rest of life; no consistent evidence yet on health (Nijp 2012, 53 studies).
• The Cochrane review (10 controlled before-after studies): self-scheduling of shifts and gradual retirement improved blood pressure, heart rate, tiredness, mental health, sleep duration and quality, and alertness, with no harm; flexibility imposed for the employer's sake (fixed-term contracts, involuntary part-time) had equivocal or harmful effects.
• STAR, a group-randomised trial in an IT company (867 workers in the well-being analysis): employees were prompted to reconsider when and where they work, and managers trained to support their personal lives. Schedule control rose; work-family conflict fell modestly; burnout, perceived stress and distress fell over 12 months (in the group surveyed before a merger); work hours and perceived demands did not rise; actigraphy showed 8 more minutes of sleep a night and less sleep insufficiency, carried by more control over work hours. An earlier natural experiment of the same kind (ROWE, 659 employees) found almost an extra hour of sleep on work nights.
• Self-rostering in 28 workplaces (840 questionnaires): need for recovery fell, sleep and distress improved in the moderate version, no harm; the gains were not explained by changes in hours worked.
• Flextime raised productivity, job satisfaction and satisfaction with the schedule, and lowered absenteeism; but highly flexible programmes did less than moderately flexible ones, and the benefits faded over time (Baltes, meta-analysis). New ways of working in a Dutch bank (2912 employees) moved hours home but kept them in the daytime on weekdays; most outcomes did not change and the health score fell (Nijp 2016).
• Reviews of mostly cross-sectional studies (moderate for association, weak for cause); a Cochrane review of controlled before-after studies with bias risks (moderate); one large group-randomised trial with objective sleep data (moderate to strong for its setting); a meta-analysis (moderate).
• For the pause: deciding when one works is a health resource when the person holds the decision, as in self-scheduling, and it acts through control and less conflict rather than through the hours themselves. A pause the person takes is that kind of control. Two cautions: total flexibility did less than bounded flexibility, and flexibility granted for someone else's convenience harmed. The pause should be bounded (the night fixed, the extension capped; section 4) and never become a way for others to claim the evening.

2.2 Self-chosen breaks restore; the choice is part of the rest. Trougakos JP, Hideg I, Cheng BH, Beal DJ. 2014. Lunch breaks unpacked: the role of autonomy as a moderator of recovery during lunch. Acad Manage J 57(2):405–421. doi:10.5465/amj.2011.1072 — verified (Crossref; abstract read). Hunter EM, Wu C. 2016. Give me a better break: choosing workday break activities to maximize resource recovery. J Appl Psychol 101(2):302–311. doi:10.1037/apl0000045 — verified (Crossref; abstract read).

• Experience sampling with several sources: social and work activities at lunch increased end-of-day fatigue when the break was not freely chosen, and reduced it when it was; for relaxing breaks, autonomy mattered only when relaxation was low (Trougakos). 95 employees over five days: breaks spent on preferred activities, and breaks taken earlier in the shift, restored more; longer breaks, or frequent short ones, restored more than rare short ones (Hunter & Wu). Self-selected exercise intensity is better tolerated than imposed intensity (1.7, Ekkekakis). System-prompted stress breaks were done more often but did not reduce stress more, and people wanted partial control of their timing (G 3.6).
• Experience-sampling studies (moderate).
• For the pause: the decision to pause must be the person's. Sioul may say that the weather favours now, if the person asked to be told; pressing the pause, and choosing what to do in it, stays theirs. A pause taken after some work, earlier rather than late in the day, restores more than one squeezed in at the end.

2.3 Boundaries: integrators, segmentors, and who is harmed by blurring. Kreiner GE. 2006. Consequences of work-home segmentation or integration: a person-environment fit perspective. J Organ Behav 27(4):485–507. doi:10.1002/job.386 — verified (Crossref; abstract read). Kossek EE, Ruderman MN, Braddy PW, Hannum KM. 2012. Work–nonwork boundary management profiles: a person-centered approach. J Vocat Behav 81(1):112–128. doi:10.1016/j.jvb.2012.04.003 — verified (Crossref; abstract read in the authors' copy). Wepfer AG, Allen TD, Brauchli R, Jenny GJ, Bauer GF. 2018. Work-life boundaries and well-being: does work-to-life integration impair well-being through lack of recovery? J Bus Psychol 33(6):727–740. doi:10.1007/s10869-017-9520-y — verified (Crossref; abstract read in the University of South Florida repository, https://digitalcommons.usf.edu/psy_facpub/2405). Gadeyne N, Verbruggen M, Delanoeije J, De Cooman R. 2018. All wired, all tired? Work-related ICT-use outside work hours and work-to-home conflict: the role of integration preference, integration norms and work demands. J Vocat Behav 107:86–99. doi:10.1016/j.jvb.2018.03.008 — verified (Crossref; abstract read on Semantic Scholar).

• 325 employees: what matters is the fit between how much a person wants to keep work and home apart and how much the workplace lets them; the effects of misfit are asymmetric, which contradicts the older advice that integration lowers conflict for everyone (Kreiner).
• Six boundary profiles, from a person-centred analysis: whatever the direction of interruptions or which role mattered more, the profiles with low perceived control of their boundaries ("reactors", "job warriors") had worse work and family outcomes than the high-control ones ("fusion lovers", "dividers", "family guardians", "eclectics") (Kossek).
• 1916 employees: those who let work into the rest of life reported fewer recovery activities, and through them more exhaustion and less work-life balance (Wepfer).
• 467 working parents: work on a computer outside working hours went with more work-to-home conflict; a preference for integration softened it, but only where workplace norms and demands were low (Gadeyne).
• Cross-sectional surveys with person-environment fit or cluster designs (moderate for association).
• For the pause: moving work into the evening is an act of integration. It suits people who like to mix, under low demands; it harms people who prefer to keep work apart, people under high demands, and people with little control over their boundaries, mostly through lost recovery. What protects everyone is control: the person moves the boundary by choice, once, and the rest of the evening stays guarded as before. areas.md names the people most at risk: those who work from home or for themselves, for whom nothing else marks the end of the day.

2.4 Neurodivergent workers: flexibility asked for, long hours and late clocks as the risks. Coogan AN, McGowan NM. 2017. A systematic review of circadian function, chronotype and chronotherapy in attention deficit hyperactivity disorder. Atten Defic Hyperact Disord 9(3):129–147. doi:10.1007/s12402-016-0214-5 — verified (PubMed 28064405; abstract read). Van Veen MM, Kooij JJS, Boonstra AM, Gordijn MCM, Van Someren EJW. 2010. Delayed circadian rhythm in adults with attention-deficit/hyperactivity disorder and chronic sleep-onset insomnia. Biol Psychiatry 67(11):1091–1096. doi:10.1016/j.biopsych.2009.12.032 — verified (PubMed 20163790; abstract read). 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 — verified (Crossref, PubMed 30267329; abstract read). Lauder K, McDowall A, Tenenbaum HR. 2022. A systematic review of interventions to support adults with ADHD at work: implications from the paucity of context-specific research for theory and practice. Front Psychol 13:893469. doi:10.3389/fpsyg.2022.893469 — verified (Crossref; abstract read). Oscarsson M, Nelson M, Rozental A, Ginsberg Y, Carlbring P, Jönsson F. 2022. Stress and work-related mental illness among working adults with ADHD: a qualitative study. BMC Psychiatry 22:751. doi:10.1186/s12888-022-04409-w — verified (Crossref; abstract read). Lee EAL, Carpini JA, Tomczak MT, Picen T, Tan T, Vinci B, Bölte S, Girdler S. 2025. "At home with my sausage dog": extending the job demands-resources model to autistic employees working flexibly. Autism Adulthood 8(5):970–988. doi:10.1089/aut.2024.0130 — verified (Crossref; findings from Curtin University's summary, https://www.curtin.edu.au/news/media-release/working-from-home-boosts-productivity-and-well-being-for-autistic-workers/). Heinze C. 2025. Workplace accommodations and employment outcomes among employees with autism: a systematic review. Cureus 17(12):e99353 (volume as given on PubMed Central). doi:10.7759/cureus.99353 — verified (Crossref; summary read on PubMed Central).

• ADHD: 62 studies, 4462 patients: consistent evidence of later chronotype, later melatonin onset and later sleep (Coogan & McGowan). In 40 adults with ADHD, the 31 with sleep-onset insomnia had a delayed sleep period and melatonin onset compared with the others and with controls (Van Veen). Two samples (251 and 372 adults): hyperfocus, including losing track of time and difficulty stopping, was more frequent with more ADHD symptoms (Hupfeld). Workplace research is thin: of 143 studies, most tested medication, and few tested anything an employer or a tool could do (Lauder). In 20 interviews, people described stress, the need for structure, and accommodations left to themselves to arrange (Oscarsson).
• Autism: interviews with 14 autistic employees and 4 managers: working flexibly (from home, with flexible hours) reduced the noise, light and social load of the office and helped productivity and well-being; but "a tendency to hyperfocus may also result in autistic employees working long hours", and good flexible work needed clear duties, support and guidelines (Lee). A small review of ten studies found accommodations, flexible hours and remote work among them, linked to better employment outcomes (Heinze; weak). Autistic burnout calls for withdrawal and reduced load (CB 2.17).
• Systematic review of circadian studies (moderate); one small physiological study (weak to moderate); questionnaire studies (weak to moderate); qualitative studies and a small review (weak; lived experience reported through research).
• For the pause: flexibility over hours is among the adjustments neurodivergent people ask for most, and the pause gives it without asking anyone. The same people carry two risks that make an evening extension dangerous: a body clock already late (ADHD), which evening work and light push later still (section 3), and hyperfocus, which makes stopping hard. An extension therefore needs a hard end that Sioul keeps, not one the person must remember.

3. Moving work to the evening: the costs¶

Already covered, not repeated: detachment as the recovery experience most tied to lower exhaustion, the stressor-detachment model, and the recovery paradox, by which the most exhausted detach least (DFD §2: Sonnentag & Fritz 2007, 2015; Wendsche & Lohmann-Haislah 2017; Sonnentag et al. 2014; G 1.4 Sonnentag 2018); work messages in the evening and detachment (DFD §2); unfinished tasks, rumination and sleep, and the plans that quiet them (DFD §1: Syrek et al. 2017; Wendsche et al. 2026; Smit 2016; Leroy & Glomb 2018; Scullin et al. 2018), evening activities and next-morning state (CB 6.7: Sonnentag 2001, ten Brummelhuis & Bakker 2012; G 3.9: Sonnentag et al. 2008), fatigue building over a working day (CB 4.5 Wiehler et al. 2022; CB 7.3 Kratz et al. 2017), the wind-down hour (areas.md, Harvey et al. 2021).

3.1 Work in the evening takes the evening's recovery. Sonnentag S, Bayer U-V. 2005. Switching off mentally: predictors and consequences of psychological detachment from work during off-job time. J Occup Health Psychol 10(4):393–414. doi:10.1037/1076-8998.10.4.393 — verified (Crossref; abstract read). Sonnentag S, Venz L, Casper A. 2017. Advances in recovery research: what have we learned? What should be done next? J Occup Health Psychol 22(3):365–380. doi:10.1037/ocp0000079 — verified (Crossref; abstract read). Barber LK, Jenkins JS. 2014. Creating technological boundaries to protect bedtime: examining work–home boundary management, psychological detachment and sleep. Stress Health 30(3):259–264. doi:10.1002/smi.2536 — verified (Crossref; abstract read). Lanaj K, Johnson RE, Barnes CM. 2014. Beginning the workday yet already depleted? Consequences of late-night smartphone use and sleep. Organ Behav Hum Decis Process 124(1):11–23. doi:10.1016/j.obhdp.2014.01.001 — verified (Crossref; abstract not read, findings from press summaries found by web search).

• 100 teachers over five days: work-related activities in the evening lowered well-being at bedtime; low-effort, social and physical activities raised it (CB 6.7). 87 people over three days: a heavier workload meant less detachment that evening, and detachment went with good mood and low fatigue, most of all on days of high time pressure (Sonnentag & Bayer). The review: recovery activities and experiences go with less exhaustion and more vigour, and both suffer when work stress is high; the same mechanisms work in breaks, evenings and holidays, and natural settings strengthen them (Sonnentag et al. 2017).
• 315 employees: work on devices at home went with shorter, poorer and less regular sleep through lower detachment, but only in people without firm limits on it (Barber & Jenkins). Two studies of managers and employees (82 and 161): work on a smartphone after 9 p.m. went with shorter sleep, more depletion the next morning and less engagement at work the next day (Lanaj et al., as summarised).
• Diary studies, consistent in direction. Moderate.
• For the pause: the price of evening work is mainly what it does to the rest of the evening (less detachment, fewer restoring activities, worse sleep), more than the hours. It is highest on days of heavy workload and time pressure, which are exactly the days with "work to catch up": the recovery paradox applied to one evening, where the person who most needs to switch off is the one the catch-up keeps switched on. Evening work done without a firm end costs sleep; with a firm end, much less.

3.2 Evening work and sleep: the mind still at work matters more than the screen's light. Harvey AG. 2002. A cognitive model of insomnia. Behav Res Ther 40(8):869–893. doi:10.1016/S0005-7967(01)00061-4 — verified (Crossref, PubMed 12186352). Gross RT, Borkovec TD. 1982. Effects of a cognitive intrusion manipulation on the sleep-onset latency of good sleepers. Behav Ther 13(1):112–116. doi:10.1016/S0005-7894(82)80054-3 — verified (Crossref; abstract read in the Penn State repository, https://pure.psu.edu/en/publications/effects-of-a-cognitive-intrusion-manipulation-on-the-sleep-onset-/). Wuyts J, De Valck E, Vandekerckhove M, et al. 2012. The influence of pre-sleep cognitive arousal on sleep onset processes. Int J Psychophysiol 83(1):8–15. doi:10.1016/j.ijpsycho.2011.09.016 — verified (PubMed 21963535; abstract read). Åkerstedt T, Kecklund G, Axelsson J. 2007. Impaired sleep after bedtime stress and worries. Biol Psychol 76(3):170–173. doi:10.1016/j.biopsycho.2007.07.010 — verified (PubMed 17884278; abstract read). Brosschot JF, Gerin W, Thayer JF. 2006. The perseverative cognition hypothesis: a review of worry, prolonged stress-related physiological activation, and health. J Psychosom Res 60(2):113–124. doi:10.1016/j.jpsychores.2005.06.074 — verified (Crossref). Chang A-M, Aeschbach D, Duffy JF, Czeisler CA. 2015. Evening use of light-emitting eReaders negatively affects sleep, circadian timing, and next-morning alertness. PNAS 112(4):1232–1237. doi:10.1073/pnas.1418490112 — verified (Crossref; abstract read). Rångtell FH, Ekstrand E, Rapp L, et al. 2016. Two hours of evening reading on a self-luminous tablet vs. reading a physical book does not alter sleep after daytime bright light exposure. Sleep Med 23:111–118. doi:10.1016/j.sleep.2016.06.016 — verified (Crossref; abstract read). Hartstein LE, Mathew GM, Reichenberger DA, et al. 2024. The impact of screen use on sleep health across the lifespan: a National Sleep Foundation consensus statement. Sleep Health 10(4):373–384. doi:10.1016/j.sleh.2024.05.001 — verified (PubMed 38806392; abstract read). Virtanen M, Ferrie JE, Gimeno D, et al. 2009. Long working hours and sleep disturbances: the Whitehall II prospective cohort study. Sleep 32(6):737–745. doi:10.1093/sleep/32.6.737 — verified (Crossref; abstract read). Dahlgren A, Kecklund G, Åkerstedt T. 2006. Overtime work and its effects on sleep, sleepiness, cortisol and blood pressure in an experimental field study. Scand J Work Environ Health 32(4):318–327. doi:10.5271/sjweh.1016 — verified (Crossref; abstract read). 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 — verified (Crossref; abstract read). Irish LA, Kline CE, Gunn HE, Buysse DJ, Hall MH. 2015. The role of sleep hygiene in promoting public health: a review of empirical evidence. Sleep Med Rev 22:23–36. doi:10.1016/j.smrv.2014.10.001 — verified (PubMed 25454674; abstract read).

• Arousal. The leading model of insomnia puts worry and cognitive arousal before sleep at its centre (Harvey). 38 women who slept well, told they would give a speech after a daytime nap, took longer to fall asleep and slept less; heart rate and skin conductance did not explain it (Gross & Borkovec). 15 healthy people who did half an hour of demanding cognitive tasks, without emotional content, before bed took longer to fall asleep, measured objectively, with more fast brain activity in deep sleep (Wuyts). Nights after days with more stress and worry at bedtime had lower sleep efficiency (81 % against 85 %) and more waking (Åkerstedt; 33 people, 16 analysed). Worry and rumination prolong the body's stress response after the stressor has gone (Brosschot).
• Light. Twelve adults reading on a tablet for four hours before bed, five evenings running, took longer to fall asleep, had less melatonin, a later clock and less alertness the next morning than when reading print (Chang). Fourteen young adults who had spent 6.5 hours in bright light (about 569 lux) read on a tablet or a book from 21:00 to 23:00: no difference in sleep or melatonin (Rångtell). A national consensus panel (16 experts, 35 experimental studies) agreed that screens, and what they show before sleep, impair sleep in children and adolescents; it reached no consensus for adults (Hartstein).
• Hours. Working more than 55 hours a week, against 35–40, predicted difficulty falling asleep (odds ratio 3.68) and short sleep (1.98) five years later in British civil servants (Virtanen). Sixteen office workers doing one week of 12-hour days against one of 8-hour days slept less and were sleepier and more exhausted by the end of the week, with no change in cortisol (Dahlgren).
• Bedtime. Going to bed later than intended with nothing to prevent it is common, tied to weaker self-regulation, and goes with insufficient sleep (Kroese; 177 people, cross-sectional). Sleep hygiene rules are each linked to sleep, but their direct effects are largely untested outside clinics, and education about them alone does not treat insomnia (Irish).
• Small experiments (weak to moderate each, consistent together for arousal; conflicting for light in adults); a large prospective cohort (moderate); a within-person field experiment (weak to moderate); a consensus statement (opinion based on a review); a cross-sectional survey (weak).
• For the pause: what keeps a person awake after evening work is mostly the mind still working (demanding tasks, worry, an unfinished loop), more than the light of the screen. A worrying letter or a hard task late in the evening is the worst case. Daylight taken during the pause may blunt the evening light's effect (Rångtell, with 1.6 Hébert), but not the arousal. The wind-down protects the last hour; the evidence argues for an extra buffer of low-arousal time between the last work and the wind-down, and for keeping anxious and demanding tasks out of a late extension (inference). Ending close to bedtime also invites "just one more thing" and a later bedtime.

3.3 Chronotype: later work suits late types and costs early ones, but a late clock has its own risk. Facer-Childs ER, Boiling S, Balanos GM. 2018. The effects of time of day and chronotype on cognitive and physical performance in healthy volunteers. Sports Med Open 4:47. doi:10.1186/s40798-018-0162-z — verified (Crossref; abstract read). Chauhan S, Vanova M, Tailor U, et al. 2025. Chronotype and synchrony effects in human cognitive performance: a systematic review. Chronobiol Int 42(4):463–499. doi:10.1080/07420528.2025.2490495 — verified (Crossref; abstract read on the Kingston University repository). Wittmann M, Dinich J, Merrow M, Roenneberg T. 2006. Social jetlag: misalignment of biological and social time. Chronobiol Int 23(1–2):497–509. doi:10.1080/07420520500545979 — verified (Crossref; abstract read). Vetter C, Fischer D, Matera JL, Roenneberg T. 2015. Aligning work and circadian time in shift workers improves sleep and reduces circadian disruption. Curr Biol 25(7):907–911. doi:10.1016/j.cub.2015.01.064 — verified (PubMed 25772446; abstract read). Au J, Reece J. 2017. The relationship between chronotype and depressive symptoms: a meta-analysis. J Affect Disord 218:93–104. doi:10.1016/j.jad.2017.04.021 — verified (PubMed 28463712; abstract read).

• 56 healthy adults tested at 14:00, 20:00 and 08:00: early and late types had different daily profiles of sleepiness, vigilance, executive function and grip strength; late types were impaired in the morning (Facer-Childs). Of 64 studies in adults aged 18 to 45, 29 (45 %) found better performance at a person's own optimal time, mostly in attention, inhibition and memory, and 10 of 12 in older adults; chronotype on its own had no main effect in most studies (Chauhan).
• 501 people: the gap between the body's time and social time ("social jetlag") was largest in late types and went with poorer well-being and more stimulant use (Wittmann). A factory that removed morning shifts for late types and night shifts for early types saw longer, better sleep and better well-being on work days, and an hour less social jetlag (Vetter).
• Eveningness goes with somewhat more depressive symptoms (36 studies, 15,734 people, z = −0.20; Au & Reece). ADHD often comes with a late clock (2.4).
• Laboratory and field studies, a systematic review, a meta-analysis of correlations (moderate overall).
• For the pause: an evening type may do the moved work as well or better at 19:00 than at 15:00; a morning type is more likely to do it worse, and closer to their own bedtime (synchrony effects appeared for both types in about half the studies). Sioul could ask once, in plain words, which suits the person ("mornings", "evenings", "neither"), rather than guess, and give a morning person a shorter limit on the extension (inference). For late types the risk runs the other way: evening work under room or screen light pushes an already late clock later (1.6 Wright; 3.2 Chang), a pattern linked to low mood. Moved work should come early in the evening, not late, whatever the type.

3.4 "Making up" time: guilt, catch-up and lost gains. Clark MA, Michel JS, Zhdanova L, Pui SY, Baltes BB. 2016. All work and no play? A meta-analytic examination of the correlates and outcomes of workaholism. J Manage 42(7):1836–1873. doi:10.1177/0149206314522301 — verified (Crossref; abstract read). Schaufeli WB, Shimazu A, Taris TW. 2009. Being driven to work excessively hard: the evaluation of a two-factor measure of workaholism in the Netherlands and Japan. Cross-Cult Res 43(4):320–348. doi:10.1177/1069397109337239 — verified (Crossref; item wording not re-read today). de Bloom J, Kompier M, Geurts S, de Weerth C, Taris T, Sonnentag S. 2009. Do we recover from vacation? Meta-analysis of vacation effects on health and well-being. J Occup Health 51(1):13–25. doi:10.1539/joh.K8004 — verified (Crossref; abstract read). Kühnel J, Sonnentag S. 2011. How long do you benefit from vacation? A closer look at the fade-out of vacation effects. J Organ Behav 32(1):125–143. doi:10.1002/job.699 — verified (Crossref; abstract read). Fritz C, Sonnentag S. 2006. Recovery, well-being, and performance-related outcomes: the role of workload and vacation experiences. J Appl Psychol 91(4):936–945. doi:10.1037/0021-9010.91.4.936 — verified (Crossref; abstract read). Tonietto et al. 2021 (G 1.9: seeing leisure as wasteful lowers its enjoyment). Buehler, Griffin & Ross 1994 (TE 1.1: the planning fallacy).

• Workaholism, best understood as a compulsion to work, goes with burnout, job stress, work–life conflict and poorer physical and mental health (Clark; meta-analysis). Its compulsive side is measured with items such as feeling guilty when taking time off (Schaufeli's scale, as usually quoted).
• Holidays improve health and well-being (d = 0.43), and the effect fades soon after the return to work (d = −0.38; seven studies). In 131 teachers the benefit was gone within a month; demands after the holiday sped its loss, relaxation in the evenings slowed it (Kühnel & Sonnentag). Workload after a holiday predicted how much of its effect remained (Fritz & Sonnentag; 221 employees).
• People who see leisure as waste enjoy it less (G 1.9). People underestimate how long their tasks take (TE 1.1), and an evening planned to "catch up" is a plan made in that optimism.
• Meta-analyses (moderate); longitudinal field studies (moderate); the transfer from holidays to a pause of a few hours is an inference, supported by the review's finding that recovery works alike across breaks, evenings and holidays (3.1, Sonnentag et al. 2017).
• For the pause: the danger is that the pause becomes a debt the evening collects with interest: guilt spoils the pause, an over-full catch-up evening erases its gain, and a person prone to compulsive work learns that every pause costs an evening. The extension should be offered as room, never as time owed; what does not fit should move to another day without a word about it (DFD); and the evening's catch-up should be planned lighter than the afternoon it replaces (inference from 3.3, TE, and fatigue rising through the day, CB 4.5, 7.3).

4. Delaying the end of the day by the pause's length¶

No study was found that tests what a planner should do with the hours of a pause taken in working time. What follows combines sections 2 and 3 with studies on long days, recovery and unfinished work; most rules drawn from it are inference, and are marked so in the criteria.

4.1 Long or late days prolong the body's work state; that is the cost an extension adds. Geurts SAE, Sonnentag S. 2006. Recovery as an explanatory mechanism in the relation between acute stress reactions and chronic health impairment. Scand J Work Environ Health 32(6):482–492. doi:10.5271/sjweh.1053 — verified (Crossref; abstract read). With 3.2 (Virtanen 2009, Dahlgren 2006, Wuyts 2012, Åkerstedt 2007) and 3.1.

• The review argues, from laboratory and field studies, that recovery after work matters most when there was too little recovery during work, and that two things block it by keeping the body activated: long exposure to work demands, and rumination. The chronic result of incomplete recovery is the path to ill health.
• Theoretical review with empirical support (moderate as a model).
• For the pause: an extension does not add hours to the day's total, but it moves the end of work towards sleep. What it costs is the unwinding time between the last work and the night, and it costs most when the moved work is demanding or worrying. A pause taken during work is itself recovery during work, which lowers the need for recovery after it (Geurts & Sonnentag's "internal" recovery); that is the case for the pause, and for not letting its price be the evening's recovery.

4.2 What extends and what intensifies: flexible time tends to grow. Baltes et al. 1999 (2.1: highly flexible schedules did less than moderately flexible ones). Glass JL, Noonan MC. 2016. Telecommuting and earnings trajectories among American women and men 1989–2008. Soc Forces 95(1):217–250. doi:10.1093/sf/sow034 — verified (Crossref; abstract read). Lott Y, Chung H. 2016. Gender discrepancies in the outcomes of schedule control on overtime hours and income in Germany. Eur Sociol Rev 32(6):752–765. doi:10.1093/esr/jcw032 — verified (Crossref; abstract read). More in section 6.

• In two decades of US panel data, most work done from home was overtime beyond 40 hours: "the capacity to work from home mostly extends the work day and encroaches into what was formerly home and family time" (Glass & Noonan). In the German panel, schedule control went with more overtime, for men with more income, for women in full-time work without it (Lott & Chung).
• Large panel studies (moderate for association; selection into flexible work is a confounder).
• For the pause: an extension that is automatic and unbounded is the mechanism by which flexibility turns into longer days. The evening must not become the place where every pause is paid for; Sioul has to hold the limit, not the person.

4.3 Unfinished work at the end of the day: give it a place, not a count. DFD §1 (Syrek et al. 2017; Wendsche et al. 2026; Smit 2016; Masicampo & Baumeister 2011; Leroy & Glomb 2018; Weigelt & Syrek 2017; Scullin et al. 2018).

• Unfinished tasks feed affective rumination, which costs sleep; writing where, when and how each will be done quiets them, mainly for people who find it hard to switch off; a short "ready-to-resume" note reduces the attention left stuck on an interrupted task; weekend work helped only when it made real progress; writing tomorrow's tasks at bedtime helped people fall asleep faster.
• Field and diary studies with an experimental arm (moderate), laboratory experiments (weak to moderate).
• For the pause: two moments need a resume note: when the pause starts in the middle of a task ("Where I stopped", already built), and when the extended day ends. What does not fit should go to a named place in the plan (a day, a first step), not into a count. The extension is worth what progress it makes on the step that matters most, not how much of the list it clears.

4.4 The options, weighed. Inference from 2.1–2.4, 3.1–3.4, 4.1–4.3; no direct test exists.

• The middle two rows match the evidence best: bounded flexibility (2.1 Baltes), the night and its approach protected (3.2), the person's choice (2.2), and nothing counted (DFD). A planner already re-plans the rest of the day; Structured's "Replan" does this commercially ("takes what you missed and reshuffles it into the time you have left", as a review of the app describes it; section 6).
• The extension adds hours, never capacity: the day's budgets do not grow with it (capacity.md §7–8), and what the pause itself cost, a long walk for someone with energy-limiting illness, counts against them (G24).

4.5 Limits. Inference, with the defaults marked as guesses in the criteria.

• The wind-down is never touched (Sioul's rule; areas.md; Harvey et al. 2021; the light consensus asks for dim light from three hours before bed, 1.6 Brown 2022).
• A buffer before the wind-down in which no work is laid: the experiments on arousal before sleep (3.2: half an hour of demanding tasks lengthened sleep onset; worry at bedtime lowered sleep efficiency), sustained activation (4.1) and bedtime procrastination (3.2) all argue for low-arousal time between the last work and the wind-down. No study gives its length; an hour is a guess.
• The evening's "time for you" slot survives (capacity.md §9): evening detachment and low-effort activities predict the next morning (CB 6.7, G 3.9).
• A latest end of work, a time the person can set, earlier for morning people (3.3); without a setting, a guess such as two hours after the usual end.
• No demanding or anxious step after the usual end: the extension takes light or usual steps, not the dreaded call or the hard letter (3.2: arousal; DFD §2: Butts et al. 2015, one unpleasant message costs the evening).

4.6 Showing it without pressure. TE criterion 24 (no countdown: padded time shown as a target gets absorbed); DFD §8–9 (no counts, no red, plain words; interrupted people compensate by working faster under more stress, Mark et al. 2008); Sonnentag & Bayer 2005 (3.1: time pressure deepens the link between low detachment and fatigue).

• For the pause: one sentence, said once, about when work comes back and until when it lasts ("Work comes back when you do; today it runs until 19:00"); on return, the same sentence and one key to keep the usual end, with no reason asked; no countdown, no "time owed", no number of steps moved. If a dated step can no longer be done before its date, say so plainly and early, as the plan already does (design.md), because that is the one fact the person must not discover late.

5. What is allowed during the pause¶

Already covered, not repeated: DRAMMA, the six things leisure gives (G 2.3: detachment, relaxation, autonomy, mastery, meaning, affiliation; within a person, relaxation, detachment, autonomy and mastery move well-being most), control over leisure as the strongest link to vigour (CB 4.9, Bennett et al. 2018), social contact as gain now and cost later (G 1.11, 2.7; CB 2.17 for autistic burnout), batching and limiting notifications (DFD §8: Kushlev & Dunn 2015; Fitz et al. 2019; Pielot & Rello 2017; Stothart et al. 2015), mind-wandering and savouring (G 2.2 Killingsworth & Gilbert 2010; G 2.14 Jose et al. 2012), measuring enjoyment lowers it (CB 7.6, Etkin 2016), leisure seen as waste (G 1.9).

5.1 The named leisure kinds each have support as recovery, by different routes. Feuerhahn N, Sonnentag S, Woll A. 2014. Exercise after work, psychological mediators, and affect: a day-level study. Eur J Work Organ Psychol 23(1):62–79. doi:10.1080/1359432X.2012.709965 — verified (Crossref; abstract read). Rook JW, Zijlstra FRH. 2006. The contribution of various types of activities to recovery. Eur J Work Organ Psychol 15(2):218–240. doi:10.1080/13594320500513962 — verified (Crossref; abstract read). Eschleman KJ, Madsen J, Alarcon G, Barelka A. 2014. Benefiting from creative activity: the positive relationships between creative activity, recovery experiences, and performance-related outcomes. J Occup Organ Psychol 87(3):579–598. doi:10.1111/joop.12064 — verified (Crossref; abstract read). Conner TS, DeYoung CG, Silvia PJ. 2018. Everyday creative activity as a path to flourishing. J Posit Psychol 13(2):181–189. doi:10.1080/17439760.2016.1257049 — verified (Crossref; abstract read). Moyer CA, Rounds J, Hannum JW. 2004. A meta-analysis of massage therapy research. Psychol Bull 130(1):3–18. doi:10.1037/0033-2909.130.1.3 — verified (Crossref; abstract read). Grove R, Hoekstra RA, Wierda M, Begeer S. 2018. Special interests and subjective wellbeing in autistic adults. Autism Res 11(5):766–775. doi:10.1002/aur.1931 — verified (Crossref; abstract read).

• Movement. 126 employees over five work days: exercise after work raised positive (not negative) affect that evening, through detachment, a sense of belonging and feeling better in one's body (Feuerhahn). 46 people over a week: physical activities predicted recovery from fatigue, while low-effort and social activities did not (Rook & Zijlstra), which conflicts with Sonnentag 2001, where low-effort and social activities raised bedtime well-being (CB 6.7). See also 1.7 and, for its limits, 1.8.
• Art and creativity. Creative activity outside work went with more mastery, control and relaxation, and with better performance at work, in two samples (Eschleman). 658 young adults over 13 days: on days after more creative activity than usual, people felt more energetic positive affect and flourishing; the reverse direction did not hold (Conner).
• Massage and body care. 37 randomised studies: one session lowered state anxiety, blood pressure and heart rate; a course of sessions lowered trait anxiety and depression, its largest effects, similar in size to those reported for psychotherapy (Moyer).
• Absorbing interests. About two thirds of autistic adults surveyed had special interests; motivation to engage in them went with higher well-being and satisfaction with leisure and social contact; very intense engagement went with lower well-being (Grove).
• Nature (1.3–1.5).
• Diary studies (moderate), a two-study survey (weak to moderate), a meta-analysis of small trials (moderate), a survey (weak to moderate).
• For the pause: exercise, creative work, massage and body care, nature and absorbing interests all have evidence as recovery, through detachment, mastery, relaxation or a sense of control. What they share is that they are chosen and absorbing. The pause should allow any item the person marks as leisure, not a fixed list, and Sioul should not rank them. Touch and crowds can be costs for some people (sensory load, CB criterion 4), so nothing should be suggested by default.

5.2 Leisure must stay leisure. G 1.9 (Tonietto et al. 2021), CB 7.6 (Etkin 2016), capacity.md §2 ("what gives back takes no room and is offered, never planned").

• Seeing leisure as wasteful, or measuring its enjoyment, lowers enjoyment.
• For the pause: leisure items may be shown during the pause, but as offers, without completion, timers, ratings asked during it, or a count afterwards. "How was it?" belongs after, rarely, as now (G5).

5.3 Who may reach the person: the chosen few, quietly, rather than everyone or no one. Kushlev K, Proulx J, Dunn EW. 2016. "Silence your phones": smartphone notifications increase inattention and hyperactivity symptoms. Proc CHI 2016:1011–1020. doi:10.1145/2858036.2858359 — verified (Crossref; abstract read). Dwyer RJ, Kushlev K, Dunn EW. 2018. Smartphone use undermines enjoyment of face-to-face social interactions. J Exp Soc Psychol 78:233–239. doi:10.1016/j.jesp.2017.10.007 — verified (Crossref; abstract read on Semantic Scholar). With DFD §2 and §8.

• Against interruption: checking e-mail three times a day, rather than at will, lowered daily stress (Kushlev & Dunn 2015, DFD §8). 221 adults, one week with alerts on and the phone in reach, one week with alerts off and the phone away: more inattention and hyperactivity with alerts on, and inattention predicted lower productivity and well-being (Kushlev 2016). Over 300 people sharing a meal with friends or family, phones on the table or put away at random: with phones present they felt more distracted and enjoyed the meal less, confirmed by experience sampling (Dwyer). A notification alone, unanswered, disrupts attention (Stothart, DFD §8); one unpleasant work message in the evening costs the evening (Butts, DFD §2); the mere expectation of work mail after hours raises anxiety (Becker, DFD §2). People are less happy when their mind wanders (G 2.2).
• Against total silence: notifications batched three times a day made people happier and calmer than constant ones, but none at all made them more anxious than usual (Fitz, DFD §8); a day without notifications left some participants anxious about being reachable and less connected (Pielot & Rello, DFD §8).
• Within-person field experiments and experience sampling (moderate).
• For the pause: the owner's rule (only the safe list, at any hour) fits the evidence better than either extreme. It removes work and admin, which carry the stress, and keeps the few people whose absence would itself worry. It should be quiet (no sound), say once what still gets through, and keep what Sioul already lets through in every quiet time: doses, codes the person just asked for, what they send themselves. A "nothing at all" choice suits people for whom any contact is a cost (autistic burnout, CB 2.17), with doses kept as the existing setting says.

5.4 Leaving work behind when the pause starts. DFD §1 (Leroy & Glomb 2018: a ready-to-resume note reduces attention residue; Smit 2016: a plan for unfinished goals helps detachment), G 2.2 (mind-wandering), G 2.14 (savouring), 1.5 (the lunch-walk effect is carried by enjoyment and detachment).

• For the pause: the gain of the pause depends on not ruminating about the work it moved. Two things help, both cheap: one line on where the person stopped (already built), and a clear statement of when work comes back. A pause that leaves the work's return vague invites checking.

6. Prior art: pauses, focus modes, and whether flexible time grows¶

Already covered, not repeated: Structured's Energy Monitor, Tiimo and other planners, with no outcome studies found (CB §5, G 4.1); protected focus time in Outlook (G 3.8, Saha & Iqbal 2023: small, weak); digital well-being apps appreciated but not changing habits (below).

6.1 What products do. Vendor documentation and first-party posts, read on 2026-10-06. Opinion level: none of these features has a published outcome study that was found.

• What none of them does: none was found that pauses work and moves the end of the work day by the pause's length; Motion's "Stop Tasks Early" and Structured's "Replan" re-lay the day's tasks in the hours left (the "keep the usual end, re-plan" option of 4.4); OS focus modes filter people and apps but know nothing of the work plan. A "pause" in these products is mostly either a notification filter or a scheduling change; none was found that ties a filter of chosen people to the work plan at once, as the owner's idea does.
• Reclaim counting time off "in your stats", and a warning that a day is over a threshold of hours, are the kinds of counts and alarms Sioul refuses (capacity.md §11: no score, no "over budget").

6.2 Do digital well-being and focus tools work? Monge Roffarello A, De Russis L. 2019. The race towards digital wellbeing: issues and opportunities. Proc CHI 2019:1–14. doi:10.1145/3290605.3300616 — verified (Crossref; abstract read). Lyngs U, Lukoff K, Slovak P, et al. 2019. Self-control in cyberspace: applying dual systems theory to a review of digital self-control tools. Proc CHI 2019:1–18. doi:10.1145/3290605.3300361 — verified (Crossref; abstract read).

• 42 digital well-being apps, 1128 user reviews, and a three-week field study with 38 people: the apps were appreciated and useful in specific situations, but did not help form new habits and were seen as not restrictive enough (Monge Roffarello). A review of 367 apps and extensions classified their designs; it reports no outcome data (Lyngs).
• HCI reviews and one small field study (weak).
• For the pause: there is no evidence base for pause or focus features as such; Sioul's pause rests on the evidence about recovery, control and interruptions (sections 1–5), not on product precedent.

6.3 Does flexible time increase total working time? Often yes, but not when control is held and bounded. Glass & Noonan 2016 and Lott & Chung 2016 (4.2). Kelliher C, Anderson D. 2010. Doing more with less? Flexible working practices and the intensification of work. Hum Relat 63(1):83–106. doi:10.1177/0018726709349199 — verified (Crossref; abstract read). Mazmanian M, Orlikowski WJ, Yates J. 2013. The autonomy paradox: the implications of mobile email devices for knowledge professionals. Organ Sci 24(5):1337–1357. doi:10.1287/orsc.1120.0806 — verified (Crossref; abstract read). Bloom N, Liang J, Roberts J, Ying ZJ. 2015. Does working from home work? Evidence from a Chinese experiment. Q J Econ 130(1):165–218. doi:10.1093/qje/qju032 — verified (Crossref; abstract read). DeFilippis E, Impink SM, Singell M, Polzer JT, Sadun R. 2022. The impact of COVID-19 on digital communication patterns. Humanit Soc Sci Commun 9:180. doi:10.1057/s41599-022-01190-9 — verified (Crossref; abstract read). Hernandez J, Swain VD, Suh J, et al. 2025. Triple peak day: work rhythms of software developers in hybrid work. IEEE Trans Softw Eng 51(2):344–354. doi:10.1109/TSE.2024.3504831 — verified (Crossref; abstract read on Microsoft Research's page). Chung H. 2022. The Flexibility Paradox: Why Flexible Working Leads to (Self-)Exploitation. Bristol: Policy Press. doi:10.1332/policypress/9781447354772.001.0001 — verified (Crossref; not read, thesis as stated in its title). Kelly et al. 2014 and Nijp et al. 2016 (2.1).

• More hours: professionals working flexibly reported more satisfaction and commitment, and also more intense work, which the authors read as effort given back in exchange for flexibility (Kelliher & Anderson; qualitative and survey). Mobile e-mail gave professionals a sense of control in the short term, and over time raised everyone's expectation of availability until they worked "everywhere/all the time" (Mazmanian; qualitative). A randomised trial of working from home in a call centre: performance rose 13 %, of which 9 points came from working more minutes per shift, through fewer breaks and sick days (Bloom). Meeting and mail data of 3.1 million users after the 2020 lockdowns: the working day lengthened by 8.2 %, about 48.5 minutes, durably (DeFilippis). 65 software developers over a month: a third peak of activity around 21:00, tied to feeling productive on remote days but to more stress on office and hybrid days (Hernandez). Flexible work tends towards self-exploitation (Chung, book).
• Not more hours: the group-randomised STAR trial raised schedule control without raising hours or demands (Kelly 2014); new ways of working in a bank moved hours home but kept them in the daytime on weekdays (Nijp 2016).
• Large panels (moderate), a randomised field experiment (moderate to strong for its setting), very large log data (moderate, descriptive), small log study (weak), qualitative studies (weak for prevalence, strong for mechanism).
• For the pause: flexible time grows when the boundary is left to the person alone under expectations of availability, and does not grow when control is real and the organisation of the day holds. The extension must therefore be bounded by rules Sioul keeps (4.5), must never be recorded as a debt, and the pause must never be suggested on the grounds of productivity.

7. The global pause and the emergency pause: two purposes, two designs¶

The emergency pause is researched in emergency-pause.md ("EP" below): a button for overwhelm that holds everything, may tell a chosen contact, lets only approved people through, covers the window with a calm message, and does not extend the working day. This section says what separates it from the global pause, so that neither is taken for the other.

7.1 Chosen recovery is proactive; a crisis is not the time to choose. Aspinwall LG, Taylor SE. 1997. A stitch in time: self-regulation and proactive coping. Psychol Bull 121(3):417–436. doi:10.1037/0033-2909.121.3.417 — verified (Crossref; abstract read). Arnsten AFT. 2009. Stress signalling pathways that impair prefrontal cortex structure and function. Nat Rev Neurosci 10(6):410–422. doi:10.1038/nrn2648 — verified (Crossref; abstract read on Semantic Scholar). Starcke K, Brand M. 2012. Decision making under stress: a selective review. Neurosci Biobehav Rev 36(4):1228–1248. doi:10.1016/j.neubiorev.2012.02.003 — verified (PubMed 22342781; abstract read). Lewis LF, Stevens K. 2023. The lived experience of meltdowns for autistic adults. Autism 27(6):1817–1825. doi:10.1177/13623613221145783 — verified (Crossref; abstract read). With EP 1.3c (Shields et al. 2016: acute stress impairs working memory and flexibility, meta-analysis) and CB 4.8 (control as a recovery experience).

• Proactive coping is acting ahead of stressors, beginning with gathering resources while conditions allow (Aspinwall & Taylor; theory). Even mild acute stress that feels uncontrollable rapidly impairs the prefrontal abilities used to plan and choose (Arnsten; review of mechanism); stress changes decisions, for better or worse depending on the task (Starcke & Brand). 32 autistic adults described meltdowns as overwhelm by information, senses, social or emotional load, with "losing logic", trouble thinking and remembering, and self-isolation to limit harm (Lewis & Stevens; lived experience gathered by research). Choosing what to do with one's free time is itself part of recovery (CB 4.8).
• Theory and reviews (moderate); a meta-analysis (EP 1.3c, strong); a phenomenological study (weak for prevalence, strong for what the state is like).
• For the design: the global pause is proactive: a well person takes a good moment and decides how the day will bend; it may ask one question and state one trade. The emergency pause is reactive: it must ask nothing, offer no options and hold nothing in mind for the person (EP). The same flow cannot serve both.

7.2 What each does to time. DFD (stopping when resources run out; nothing counts what was not done), CB criteria 20 and 34 (a too-much day lowers what the next days hold; the day after a heavy event holds less), CB 2.6 (PEM delayed hours to days), 3.4 (Kühnel & Sonnentag 2011: demands after recovery speed the loss of its gain), DFD §3–4 (guilt and shame).

• For the design: the global pause moves work: the end of the day moves by the pause's length, within limits (4.5), and the plan keeps its shape. The emergency pause moves nothing and owes nothing: the rest of the day holds less, as a foggy day does, and the next day may too. Turning a crisis into hours to make up would add guilt at the worst moment and repay recovery with demands. Inference.

7.3 Two pauses that look alike invite mode errors. Norman DA. 1981. Categorization of action slips. Psychol Rev 88(1):1–15. doi:10.1037/0033-295X.88.1.1 — verified (Crossref). Sarter NB, Woods DD. 1995. How in the world did we ever get into that mode? Mode error and awareness in supervisory control. Hum Factors 37(1):5–19. doi:10.1518/001872095779049516 — verified (Crossref; abstract read).

• Mode errors happen when a system has several modes and the person acts as if it were in another; when designers multiply modes without making the current one visible, new errors follow (Sarter & Woods, from studies of pilots and cockpit automation; Norman's classification of slips).
• Theory and field studies in another domain (moderate as human-factors principle; transfer to Sioul is inference).
• For the design: the two failures to avoid are a person in distress pressing the global pause and finding their evening taken by work, and a person pressing the emergency pause for a fine afternoon and finding the window locked and a contact told. Hence: different names, different places, different looks; the status line says which one is on and what it does; and from the global pause, one key turns it into a pause that moves nothing (the "keep my usual end" of 4.6), so a pause pressed in a bad moment never costs an evening.

7.4 Side by side. Inference from 7.1–7.3, sections 1–6, and the emergency-pause note.

• Shared ground: both switch to a time in which only chosen people reach the person, using the same lists and the same quiet rules, so the person learns one rule ("only the people I chose") rather than two.

Design criteria for Sioul¶

Numbered GP1–GP21. They keep every rule of the other notes (CB, G, TE, DFD) and of design.md. "Inference" marks a rule built from several sources rather than shown by one; numbers marked "guess" are starting values to tune, not findings.

Starting the pause

• GP1. The pause is the person's act. One control in the bottom toolbar, pressed by the person. Sioul never starts it, never proposes it from behaviour, and never presents it as a way to work better. Worker-held control over time helps, control imposed or framed for others' benefit does not, and the choice is part of the rest. — 2.1 Joyce 2010, Kelly 2014, Moen 2016; 2.2 Trougakos 2014, Hunter & Wu 2016; 6.3 Mazmanian 2013, Kelliher & Anderson 2010; design.md (nothing guessed from behaviour).
• GP2. The weather informs; it does not prompt. If the person asks for it, the weather applet may say when the next hours are better for going out than the evening ("Dry and mild until 16:00; rain from 18:00"). It is off by default, never a notification, and never "nice weather!": weather lifts mood only for people who go out in it, mostly after bad spells, not for everyone, and heat lowers mood for some. — 1.1 Keller 2005; 1.2 Denissen 2008, Klimstra 2011, Lucas & Lawless 2013; 1.7 Tucker & Gilliland 2007; G 3.6 Howe 2022 (prompts raise uptake, not benefit). Inference.
• GP3. Leave work behind in two lines. Starting the pause offers the existing "Where I stopped" line, and says once when work comes back and until when it runs. — 4.3 and 5.4: Leroy & Glomb 2018, Smit 2016; 1.5 Sianoja 2018 (the gain is carried by detachment and enjoyment).
• GP4. No minimum, no timer. A short pause outside already gives a measurable effect, and longer adds less per minute; Sioul shows no length, target or countdown for it. — 1.3 Barton & Pretty 2010, Hunter 2019, McMahan & Estes 2015; 1.5 de Bloom 2017; TE criterion 24.

During the pause

• GP5. Leisure time, with only the chosen people. The pause is Leisure at any hour: the safe list's mail and messages come, quietly, without sound; doses, codes the person just asked for, and what they send themselves come as in every quiet time; work and admin wait, including their sites and calls. One sentence says what still gets through. — 5.3 Kushlev 2016, Dwyer 2018, DFD §8 (Fitz 2019, Pielot & Rello 2017, Stothart 2015), DFD §2 (Butts 2015, Becker 2021); areas.md.
• GP6. "Nothing at all" as a choice, for people for whom any contact is a cost; doses keep their own setting. — 5.3; CB 2.17 (autistic burnout). Inference.
• GP7. Leisure items are offers, not a list. Any item the person marks as leisure (movement, art and creativity, massage and body care, nature, absorbing interests) may show during the pause, without completion, timers, ratings asked during it, or a count after. Sioul does not rank the kinds or suggest one by default. — 5.1 Feuerhahn 2014, Eschleman 2014, Conner 2018, Moyer 2004, Grove 2018; 5.2 G 1.9 Tonietto 2021, CB 7.6 Etkin 2016; G21.
• GP8. Energy-limiting illness: no exercise suggested, costs counted, a crash expected late. When the person's settings say an illness limits their energy, the pause never suggests movement, what is done in it counts against the day's budgets like any item, and the next days are lightened as the capacity rules already do after a heavy day. — 1.8 NICE NG206, Stussman 2020, Appelman 2024; G24; CB criteria 20, 34.

The end of the work day

• GP9. The extension adds hours, never capacity. The day's budgets do not grow with it; the room after the pause is laid within what remains of them. — 4.4; capacity.md §7–8. Inference.
• GP10. Move the end of work by the pause's length, within limits Sioul keeps. By default the end of work moves later by the length of the part of the pause that fell in working hours, but never past the earliest of: (a) the start of the wind-down less a buffer of low-arousal time (guess: one hour); (b) the person's latest end of work, a setting (guess without one: two hours after the usual end; earlier for someone who says mornings suit them); (c) the evening's "time for you" slot (capacity.md §9). What the remaining hours cannot hold goes to the next days, without a count. — 3.1 Sonnentag & Bayer 2005, Barber & Jenkins 2014; 3.2 Gross & Borkovec 1982, Wuyts 2012, Åkerstedt 2007, Kroese 2014; 3.3 Facer-Childs 2018, Chauhan 2025; 4.1 Geurts & Sonnentag 2006; 4.2 Glass & Noonan 2016, Baltes 1999; 1.6 Brown 2022; areas.md (the wind-down is sleep's). Inference; the numbers are guesses.
• GP11. One key keeps the usual end. On return, "Keep my usual end" (no reason asked) leaves the end where it was and lays the remaining hours anew; the person may also make either behaviour the default, once. — 2.2 autonomy; 2.3 Kreiner 2006, Kossek 2012, Wepfer 2018, Gadeyne 2018 (integrators and segmentors); 4.4. Inference.
• GP12. Nothing hard or worrying after the usual end. In the extension the plan lays only light or usual steps: no step rated high on anxiety, no dreaded call, no new letter from the Porch, nothing that opens a new project. — 3.2 Wuyts 2012, Åkerstedt 2007, Brosschot 2006; DFD §2 Butts 2015; 4.5. Inference.
• GP13. Early in the evening, and lighter. Moved work is laid at the start of the evening, never at its end, and evening hours are planned as holding less than daytime hours, whatever the person's estimate. — 3.3 Facer-Childs 2018, Wittmann 2006, 1.6 Wright 2013, 3.2 Chang 2015; 3.4 TE 1.1 Buehler 1994; CB 4.5 Wiehler 2022, CB 7.3 Kratz 2017. Inference.
• GP14. Say early what a deadline can no longer hold. If, after the pause, a dated step cannot be done before its date, say so plainly and at once, with the plan's usual offer to drop, shrink or hand over. — design.md; 4.6.
• GP15. Close the extended day like any other. "Done for today" ends it, at the extended end or earlier, without comment; what remains gets a place (a day, a first step), not a number. — 4.3 Smit 2016, Leroy & Glomb 2018, Scullin 2018, Weigelt & Syrek 2017; DFD.

Showing it

• GP16. Time in one sentence, never as a race. The status line names the pause and when work comes back ("Work comes back when you do; today it runs until 19:00"). No countdown, no "hours owed", no number of steps moved, no red. — 4.6: TE criterion 24, DFD §8 (Mark 2008), Sonnentag & Bayer 2005; capacity.md §11.
• GP17. Nothing becomes a debt or a record of the person. Pauses are not counted, charted, compared or used to infer anything about the person. — 3.4 Clark 2016, Schaufeli 2009, de Bloom 2009, Kühnel & Sonnentag 2011; DFD (streaks); design.md; 6.1 (Reclaim's time-off statistics as the counter-example).

The two pauses

• GP18. Different names, places and looks, and a status line that says which pause is on and what it does. — 7.3 Norman 1981, Sarter & Woods 1995.
• GP19. The global pause asks at most one choice and can always become one that moves nothing; the emergency pause asks none and moves nothing. — 7.1 Arnsten 2009, Starcke & Brand 2012, EP 1.3c Shields 2016, Lewis & Stevens 2023; 7.2; 7.3.
• GP20. One rule for who gets through. Both pauses use the same lists and quiet rules (safe senders, doses, codes), the emergency pause possibly narrowing them further. — 7.4. Inference.

Neurodivergent people

• GP21. Hold the end firmly for late clocks and hyperfocus. Sioul, not the person, ends the extended work time, and for people with ADHD or a late clock the extension stays early in the evening. — 2.4 Coogan & McGowan 2017, Van Veen 2010, Hupfeld 2019, Lee 2025; 3.2 Kroese 2014; 3.3 Au & Reece 2017. Inference.

To check over time (on the person's own record, never shown as a score; in the manner of capacity.md §12)

• How often the extension reaches its limit, and how often "Keep my usual end" is chosen: if the limit is often reached, the default moves too much work.
• Days said "too much" and nights said "heavy" after extended days, against other days.
• Whether pauses followed by an extension are followed by a later bedtime (where the night is recorded).

Open questions / conflicts¶

1. No direct evidence on the extension. No study tests moving the end of a work day by a pause's length; every rule about it (GP9–GP15) is inference from studies of recovery, long and late days, and flexible time.
2. The buffer and the latest end. One hour of low-arousal time before the wind-down, and two hours past the usual end, are guesses; the arousal experiments (3.2) say "some", not how much.
3. Light or arousal. Chang 2015 found evening tablet light delayed the clock and sleep; Rångtell 2016 found no effect after a bright day; the National Sleep Foundation panel reached consensus for children and adolescents only (3.2). The design rests on arousal, which is better supported, and treats light as secondary.
4. Which leisure restores. Sonnentag 2001 found low-effort and social activities restoring; Rook & Zijlstra 2006 found only physical ones were (5.1). Left to the person.
5. Nature or movement. Barton & Pretty 2010 credit green settings; Lahart 2019 finds limited evidence that green exercise beats indoor exercise (1.3, 1.4). The pause need not be outdoors to be worth it.
6. Seasons. Keller 2005 found weather lifts mood most in spring; de Bloom 2017 found lunch walks helped least in spring (1.1, 1.5).
7. Does flexible time grow? Panels, a randomised trial and log data say it often does (Glass & Noonan, Lott & Chung, Bloom, DeFilippis); the STAR trial and a Dutch bank say it need not (6.3). The difference seems to be whether control is held and bounded; that is a reading, not a test.
8. Late types. Evening work suits a late type's performance on the day (3.3), but evening light and activity push the clock later over days, and eveningness goes with low mood (Au & Reece 2017). How often an extension can be used before it shifts the clock is unknown.
9. Integrators and segmentors. The right default (move the end, or keep it) differs between them (2.3). Asking once is the inference; whether people know which they are is untested.
10. Safe list or silence. Total silence made some people anxious (Fitz 2019, Pielot & Rello 2017); for some autistic people any contact is a cost (CB 2.17). Default and option (GP5–GP6) are a compromise.
11. Telling the person the weather is good. Prompts raise uptake without more benefit (G 3.6 Howe 2022), and Sioul's rules forbid pushing. Whether an opt-in weather line helps people with low initiation (ADHD, depression) more than it pressures them is untested.
12. Repeated use. If the pause and its extension become daily, evenings and the clock erode (3.1, 3.3); a weekly limit has no evidence, and nagging is ruled out.
13. Other people's hours. Meetings, clients and offices keep their times; the pause can only move what the person controls. Whether the pause should ever be visible to others (as iOS shares a Focus status) is a social question not researched here; availability norms raise expectations (Mazmanian 2013, Gadeyne 2018).
14. Planning the pause ahead ("if it is dry tomorrow at 14:00") was not researched.
15. Nothing here has been tested with Sioul's users.

Sources checked (verification log)¶

• Method (2026-10-06, 13:09 onwards): every DOI above was resolved on Crossref (title, authors, year, venue), except Reed & Ones 2006, matched on OpenAlex when Crossref timed out; PubMed records were checked where an entry names them; abstracts were read from OpenAlex, PubMed, Semantic Scholar or the named repository. One author's copy of a full text was read (Kossek et al. 2012, first pages). Product facts come from vendor documentation and first-party posts, with their URLs.
• Findings not read in the primary source, marked in their entries: Keller 2005's 30-minute threshold (British Psychological Society summary); Klimstra 2011's type percentages (secondary summaries); Barton & Pretty 2010's five-minute peak (press summary); Reed & Ones 2006 (summaries; Crossref did not answer, OpenAlex matched title, year and DOI); Xu et al. 2026 (summary); Lanaj et al. 2014 (press summaries); Lee et al. 2025 (university summary); DUWAS item wording (Schaufeli 2009); Chung 2022 (not read); several product details (secondary reviews, marked).
• Not peer-reviewed, marked as such: vendor documentation (Apple, Google, Motion, Reclaim, Sunsama, Structured, Tiimo); Chung 2022 is an academic book; NICE NG206 is a guideline; the National Sleep Foundation statement is a consensus.
• No reference in this file was invented.