Skip to content

What the research says, and the rule each finding gives

The detailed notes behind these rules (what each study found, how strong its evidence is, where Sioul applies it, and what was built, left for later or refused, with why) are in research/README.md.

The evidence is not all of one strength: randomised trials and meta-analyses, observational studies, qualitative studies with the people concerned, guidelines, and, where no study tested the point, Sioul's own inference, marked as such in the notes. What none of it shows yet is that Sioul, as built, lightens admin for the people who use it. The research supports the design, not yet its outcome: whether letters are avoided less, starting comes easier, mail intrudes less, fewer things are forgotten and evenings recover better needs testing with the people who use it.

Who the findings are about

Sioul is designed for the people for whom admin costs most. Each group below has the findings that concern it most; most findings serve several groups, and none needs a diagnosis to apply.

  • Autistic people: uncertainty (5), autistic inertia (6), cognitive accessibility (10), visual supports (13), autistic burnout (19), the phone and executive function in admin (29), overwhelm, meltdown and shutdown (32), unexpected calls (34).
  • People with ADHD: support at the point of performance (6), criticism (15), timers (18), "brain weather" (19), timing (24), admin and executive function (29), the two routes of ADHD and PTSD (30), working for oneself (31).
  • People living with PTSD or another trauma: trauma-informed design (9), an office as a reminder of harm (30), what helps and harms in overwhelm (32).
  • Anxiety: avoidance as mood repair (4), uncertainty (5), admin anxiety (30), unexpected calls (34).
  • Depression: shame and guilt (23), timing (24), valued activities as a dose (27), daylight and time outdoors (33).
  • Burnout and exhaustion: autistic burnout (19), detachment from work (21), unfinished tasks (22), recovery (26).
  • ME/CFS, long COVID and other illnesses that limit energy: pacing within the energy envelope (25), the crash after exertion (33), the sides of what an activity costs (36).
  • Eating disorders and irregular eating: meals by the clock, set by the person (28).
  • Everyone who knows admin anxiety: administrative burden (3), avoidance (4), admin anxiety and its routes (30), worry as a cost of its own (36), the portal model (37).
  • People who work for themselves: the admin it adds (31).

The findings

  1. Fewer, predictable looks at mail.
    • Checking mail three times a day for a week, against freely, lowered daily stress in a randomised trial of 124 adults; lower stress predicted better well-being (Kushlev & Dunn 2015).
    • Batching notifications three times a day improved attention, mood, productivity and stress in 237 participants, while those given no notifications at all felt more anxious than controls (Fitz et al. 2019).
    • Without email for five days, 13 office workers switched windows half as often, and their heart-rate variability showed less stress (Mark, Voida & Cardello 2012).
    • Rule: admin windows you choose, and the app always says when the next one opens. Predictability, not silence.
  2. Interruptions cost even when ignored. Interrupted people work faster, with more stress, frustration and effort (Mark, Gudith & Klocke 2008). A notification left unanswered still disrupts attention (Stothart, Mitchum & Yehnert 2015). Rule: no pop-ups, sounds or badges; the layout never moves under your eyes.
  3. Administrative burden is made of learning, compliance and psychological costs (Moynihan, Herd & Harvey 2015). It looms larger for people with fewer cognitive resources, executive function, scarcity and health problems among them, as Christensen et al. (2020) argue, naming ADHD; a test followed: in two surveys, people reporting ADHD, anxiety or chronic pain found tax or student-aid rules more burdensome, and those reporting ADHD or pain lost the aid more often for missed requirements (Bell et al. 2023; observational, small effects). Rule: Sioul carries the three costs. Learning: what a letter is and wants. Compliance: drafts, attachments and addresses gathered. Psychological: tone, pacing, no shame.
  4. Avoidance is mood repair, not laziness. Procrastination gives priority to short-term mood (Sirois & Pychyl 2013), and people avoid information they expect to hurt (Sweeny et al. 2010; Golman, Hagmann & Loewenstein 2017). Rule: make looking cheaper. The card comes first, saying who, what it asks, by when, and what happens if nothing is done.
  5. Uncertainty fuels anxiety (Carleton 2012), in autism too (Boulter et al. 2014). Rule: every item states what happens next and who waits on whom; no ambiguous states.
  6. Starting is the hard part.
    • Autistic inertia is a difficulty acting on intentions, eased by scaffolding: "no way out except from external intervention" (Buckle et al. 2021).
    • For ADHD, the support belongs at the point of performance: time, memory and motivation externalised (Barkley 2012). Tools still leave prioritisation, time estimation and task switching to ADHD knowledge workers (Campbell, Çerçi & Cecchinato 2024).
    • "When X, I do Y" plans raise follow-through (Gollwitzer & Sheeran 2006); near subgoals build competence (Bandura & Schunk 1981).
    • Rule: one next step, small, with a when, pre-filled where possible; priorities computed by Sioul.
  7. Company helps. Body doubling is a practice neurodivergent people found by themselves (Eagle & Ringland 2024). Rule: an admin session can have a companion, an AI or a person.
  8. Progress, never streaks. Small wins carry motivation (Amabile & Kramer 2011). Broken streaks lower later engagement, more so when people blame themselves (Silverman & Barasch 2023). Rule: show what got done; no streaks, no pile of "overdue".
  9. Trauma-informed design: expect that trauma may be present, and do not repeat it.
    • SAMHSA's guidance for services asks an organisation to realise how widespread trauma is, recognise its signs, respond by building that knowledge into its procedures, and resist re-traumatisation, through six principles: safety; trustworthiness and transparency; peer support; collaboration and mutuality; empowerment, voice and choice; cultural, historical and gender issues. It takes "a universal precautions approach in which one expects the presence of trauma in lives of individuals being served, ensuring not to replicate it", and asks services to keep people informed of procedures and schedules, "mindful that people who are frightened or overwhelmed may have a difficulty processing information" (SAMHSA 2014, pp. 9–11 and 15).
    • Chen et al. (2022) adapt the principles to computing (safety, trust, peer support, collaboration, enablement, intersectionality): design as if trauma may be at play for anyone, so that nobody has to disclose it; warn before visible changes; explain why something is shown; rethink harsh warnings that force attention; let no assistant judge users or raise their anxiety.
    • Frameworks and design cases: strong as a shared vocabulary, untested as to effect. SAMHSA writes for services, and says nothing of software or paperwork: applying it to admin software is Chen et al.'s step, and Sioul's.
    • Rule, principle by principle:
      • safety: no surprises: no badge, no red, new mail without a sound, nothing moving under your eyes; forged mail marked by a shape and a word, not an alarm (design.md);
      • trust and transparency: the reason shown (why this step now, why this mail was set aside, what a setting does), anything moved, deleted or sent said in one sentence, with ten seconds to undo;
      • collaboration: Sioul proposes and you decide: nothing sent, deleted or paid without you, not even by an AI agent;
      • enablement, voice and choice: your rules and your hours; your own word for your state (a day clear, hazy or foggy), never inferred; each field shown when it is needed;
      • cultural, historical and gender issues: Sioul asks nothing about who you are, needs no diagnosis and infers nothing, so nothing has to be disclosed;
      • peer support is not software's to give; making room for the people who help, a project handed to a helper with what it needs, is not built yet (research/life-admin.md, §3); each project is opened to AI agents on its own, closed until then (ai.md).
  10. Cognitive accessibility (W3C, 2021): help users focus; make processes not rely on memory; use clear language; help users avoid mistakes; support personalisation. For autistic users (UK Home Office, 2016): simple colours, plain English, short sentences and bullets, descriptive buttons, simple and consistent layouts. Rule: key hints on screen, literal labels, confirmations before anything irreversible, everything adjustable.
  11. Tracking is work too. Self-tracking imposes "substantial interpretive and emotional demands" on neurodivergent people (Rudberg Selin, Unéus & Knudsen 2026). Rule: nothing to log by obligation.
  12. AI as a communication aid, with care. Autistic workers preferred an LLM to a human for workplace communication advice, while a job coach found some of that advice questionable (Jang et al. 2024). Rule: the AI drafts and explains, you review, and it never sends.
  13. Visual supports are an evidence-based practice in autism, mostly studied in children and young people (Steinbrenner et al. 2020). Rule: timelines and the Gantt chart on demand, never compulsory.
  14. Too many options weigh when choosing is hard. Across 50 experiments the mean effect of more options was nil (Scheibehenne, Greifeneder & Todd 2010); overload appears when options are hard to compare, preferences unclear and effort to be saved (Chernev, Böckenholt & Goodman 2015): a to-do list on a tired day. Rule: one next step, picked, with its reason; equal steps are picked, not shown as a choice (tasks.md).
  15. A plan date is not a deadline, and lateness is not a debt. Forgiving oneself for procrastinating lowered later procrastination (Wohl, Pychyl & Bennett 2010); adults with ADHD meet more criticism, and lower self-compassion goes with poorer mental health (Beaton, Sirois & Milne 2022). Rule: nothing overdue; a day to start that passed says nothing; the date asked is time left; no counts of what was put off.
  16. Once started, a task is rarely as bad as feared, and an interrupted task pulls to be resumed (Pychyl's experience sampling; Ghibellini & Meier 2025). Rule: a two-minute start where stopping counts.
  17. A plan for where you stopped frees the mind for the next thing. Writing a one-minute ready-to-resume plan reduced attention residue (Leroy & Glomb 2018); a warning before an interruption shortens getting back in (Trafton et al. 2003). Rule: stopping offers one line, shown when the task comes back; a heads-up two minutes before the end.
  18. Time made visible: visual timers and day schedules improved time management in children with ADHD in a randomised trial (Wennberg et al. 2018); a ticking countdown can raise anxiety instead. Rule: a draining disc in a neutral colour, without ticking or sound, open-ended when you prefer.
  19. Capacity is lower on some days, and only you know which. Autistic burnout brings chronic exhaustion and loss of executive skills, eased by reduced expectations (Raymaker et al. 2020); adults with ADHD liked a "brain weather" view and disliked tools that watch them (Chen, Meng & Nie 2026). Rule: the day's weather is chosen, never inferred; fog shows small steps, and one is a full day.
  20. Splitting a task makes it startable and its estimate truer (Kruger & Evans 2004). Rule: steps in one line each, their minutes added up.
  21. Work cues in the evening undo recovery. Detachment from work goes with less exhaustion (r = −.38 across 91 samples: Wendsche & Lohmann-Haislah 2017), and the most exhausted detach least (Sonnentag et al. 2014). Work messages in the evening cost the evening through their tone (Butts, Becker & Boswell 2015), and the mere expectation of checking harms, read or not (Becker et al. 2021). Rule: quiet time outside working hours, during time off and once the day is closed: only the senders you let through then, codes and what you enjoy, with one sentence saying until when; during sleep, no notification but doses (porch.md, areas.md).
  22. An unfinished task stops pulling once it has a place. Unfinished tasks feed affective rumination and poor sleep (Syrek et al. 2017; Wendsche, Weigelt & Syrek 2026); writing where, when and how they will be done quieted intrusions in the lab (Masicampo & Baumeister 2011, not replicated) and raised evening detachment in the field, mostly for people low in self-control (Smit 2016). Rule: closing the day gives everything a place and names the first step, concretely.
  23. Shame, not guilt, feeds avoidance, and repair makes it constructive. Shame tracks depression more than guilt does, as does guilt over what one could not control (Kim, Thibodeau & Jorgensen 2011); shame turns to repair when repair looks possible (Leach & Cidam 2015); feedback about the self rather than the task lowers performance (Kluger & DeNisi 1996). Warm affirmations can threaten the most self-critical (Kirby, Day & Sagar 2019). Rule: no count of what was not done, no history of early stops, no question; kindness in what the software does, in plain words about tasks.
  24. Your guess of how long a task takes carries information, with a bias. The typical error is 20–30 %: short tasks are overestimated, long ones underestimated, and a day's total overruns because overruns are bigger than underruns (Halkjelsvik & Jørgensen 2012). Guessed and real times correlated at r = .77 despite a 1.6× bias (Buehler, Griffin & Ross 1994). People misremember how long past tasks took, and seeing the measured times of similar ones helps (Roy, Christenfeld & McKenzie 2005; Roy, Mitten & Christenfeld 2008). ADHD and depression change timing over seconds, not the length predicted for a task (Marx et al. 2022; Thönes & Oberfeld 2015). Rule: your guess is kept and corrected, never replaced: time spent against time guessed, as a ratio on a log scale, over about two weeks, per kind of task pulled toward your overall ratio; the safety margin held for the whole day, not per task; never a score, never "late", never your estimate pre-filled, never a diagnosis. Detail: time-estimation.md. Built: capacity.md, the kind being the task's area.
  25. An average of past load measures habit, not capacity. In sports, the long-term load average predicted injury no better than random numbers (Impellizzeri et al. 2021); one session more than 10 % above the longest of the past 30 days raised injury risk (Frandsen et al. 2025). In ME/CFS, activity is never raised automatically, only after a stable period, and a crash can come days later (NICE NG206, 2021); in long COVID, a structured pacing protocol was followed by fewer crashes in an uncontrolled cohort (Parker et al. 2023). Rule: a daily budget for each cost (thinking, feelings, anxiety, body and senses) and in total, learned from days that went well, with the outcome you give at the end of the day (reviews.md); raised slowly after stable days, lowered at once after a crash; no day planned above the heaviest well-tolerated one plus 10 %; no gauge. Detail: capacity-budget.md. Built: capacity.md.
  26. Willpower is not a tank; recovery is. Large replications found ego depletion near zero (Hagger et al. 2016; Vohs et al. 2021); breaks of more than ten minutes after demanding tasks help (Albulescu et al. 2022). Rule: the plan counts the day's load and its recovery, not minutes of willpower; after a heavy block, a light one or a real break, never two heavy blocks in a row. Built: capacity.md.
  27. What gives back is a floor to reach, not a currency. Scheduling valued activities works as a dose (Jacobson et al. 1996; a self-guided app in a 649-person trial, Dahne et al. 2025); good events raise good mood without lowering bad mood (David et al. 1997), so a gain does not buy room for more cost. Free time helps up to about two hours a day and less past five (Sharif, Mogilner & Hershfield 2021); measuring enjoyment lowers it (Etkin 2016). What gives back: detachment, relaxation, autonomy, mastery, meaning, company (DRAMMA, Kujanpää et al. 2021). Rule: one rating of what it gives back, 0 to 10, given afterwards and rarely; a daily floor kept apart from the costs; something short that gives back right after a heavy block, the evening kept low-key; Sioul keeps the slot free, you choose what goes in it; never a quota. Detail: wellbeing-gain.md. Built: capacity.md.
  28. Meals by the clock, set by you, said plainly, for people with an eating disorder and for people who eat irregularly (burnout, depression, ADHD). Regular eating by the clock, not by hunger (CBT-E: Murphy et al. 2010); every part set by the person (Lindgreen, Lomborg & Clausen 2018); neutral words, no numbers, no "missed" (Eikey 2021; Levinson et al. 2017). Rule: meals, naps and the night are set first and kept free, the work goes in what is left; two notices at most; moved or skipped for today without a word asked; nothing about food, portions or numbers ever recorded or shown. Detail: meal-prompts.md. Built: health.md, "Meals, rest and sleep".
  29. Admin weighs more on neurodivergent people, through executive function and the phone.
    • ADHD's own diagnostic criteria name admin: forms and long papers avoided; calls, bills and appointments forgotten; paperwork lost (American Psychiatric Association 2013). Its executive weaknesses (inhibition, working memory, planning) are of medium size (Willcutt et al. 2005), and procrastination rises with inattention across 11 studies (Suriano 2026). Adults with ADHD notice incoming bills less (Koerts et al. 2023), miss more GP appointments (odds ratio 1.9 in Scottish records: McQueenie et al. 2024), and default on credit more and more from middle age (Swedish registers: Beauchaine, Ben-David & Bos 2020).
    • Most autistic adults avoid the phone (78 % against 38 %), and 62 % find booking a GP appointment by phone hard, against 16 % (Doherty et al. 2022); for services they rank e-mail first and the phone last (Howard & Sedgewick 2021); a third cannot process speech fast enough for real-time talk (Raymaker et al. 2017).
    • Support workers in Sweden say that young adults with autism or ADHD often need help with authorities, calls and documents, and that sorting out their contacts with authorities often comes first (Löthberg et al. 2024).
    • Rule: the software carries the dates, notices the bills and the debits, and turns each letter into one small step; writing before calls (research/life-admin.md, §1.3–1.4; research/call-screening.md).
  30. Admin anxiety is common, and ADHD and PTSD make admin heavier by different routes; no study shows a dread of offices that the two share.
    • "Admin anxiety", a term from a UK policy institute's surveys (grey literature), was found in 23 % of UK adults and 37 % of people with mental health problems, with banks and utilities; the telephone was the hardest channel, text and the web the easiest, and some never open the post they dread (Holkar, Evans & Langston 2018). With benefits, avoiding contact with the agencies was common (Bond, Braverman & Evans 2019). Neither breaks its results down by diagnosis.
    • With ADHD, the route runs through forgetting, avoided effort and procrastination (finding 29).
    • With PTSD, avoidance is of reminders of the trauma (World Health Organization 2024, ICD-11), and an office becomes one when it makes people tell or prove the trauma again: veterans and people in therapy for trauma described benefit assessments and disability claims as feared and re-traumatising (Scullion & Curchin 2022; Roberts et al. 2024; Webermann et al. 2026). Harm done by an institution goes with distrust of it (Smith & Freyd 2014). PTSD also comes with weaker attention, memory and executive function (Scott et al. 2015).
    • The two often come together: PTSD is about three times as frequent with ADHD, and ADHD 1.7 times as frequent with PTSD (Spencer et al. 2016, meta-analysis), though not significantly more than in people with other psychiatric conditions.
    • No study measures a dread of letters, forms or calls from offices in people with ADHD or with PTSD, nor compares the two on admin.
    • Rule: serve both routes. Against forgetting and avoided effort: the card first, saying who, what it asks, by when and what happens if nothing is done; one small step; every date carried by Sioul. Against threat: no surprises, writing before calls, and papers kept once, to be given again without retelling (papers.md). For both: nothing that shames, nothing overdue.
  31. Working for oneself adds admin, and people with a disability or a long-term condition work for themselves more often.
    • In the US in 2025, 9.1 % of employed people with a disability were self-employed, against 5.9 % of those without (U.S. Bureau of Labor Statistics 2026); European data point the same way (Pagán 2009; Kitching 2014). ADHD symptoms went with self-employment in a Swedish population cohort, modestly and through hyperactivity (odds ratio 1.13: Verheul et al. 2016), and a reported diagnosis with plans to start a business, and with starting one, among Dutch students (Lerner, Verheul & Thurik 2019). No study gives a rate for autistic adults. In interviews, some disabled and autistic people left jobs that would not adapt to them (Adams et al. 2019; Davies et al. 2025).
    • The self-employed declare and pay their own taxes, where an employer does it for employees (Milanez & Bratta 2019); compliance costs weigh most on the smallest businesses (Evans 2003); none of 14 founders with ADHD was at ease with bookkeeping (Wiklund, Patzelt & Dimov 2016), and disabled self-employed people who find planning hard called tax planning "especially difficult" (Adams et al. 2019).
    • Rule: carry the small business's admin too, from the client's first mail to the paid invoice and the figures for the accountant, in the same calm window (projects.md, accounting.md).
  32. In overwhelm, less: nothing to answer, nothing new, words that are literal. Acute stress narrows working memory and flexibility (meta-analysis: Shields, Sazma & Yonelinas 2016), and much of what is said is forgotten at once (Kessels 2003); in an autistic shutdown, speech and movement can stop (Paris et al. 2025). Psychological first aid, the consensus for the first minutes, brings immediate relief and does not prevent PTSD (World Health Organization et al. 2011; one trial); single-session debriefing helped nobody, and made PTSD more likely at one year in one trial (Rose et al. 2002, Cochrane); trying to relax under load raises arousal, so "calm down" backfires (Wegner, Broome & Blumberg 1997). No study has tested a pause button. Rule: the Pause asks nothing and shows nothing new: set up beforehand, one press, a few literal sentences about Sioul and never about the person, no question, a breathing guide only if chosen, nothing sent to anyone, the emergency number and a crisis line one tap away; coming back, the rest of the day is lighter (pauses.md; detail: emergency-pause.md, criteria P1–P31).
  33. Taking the good moment: going out into it, briefly, with a say over one's own working time, within the energy envelope. Pleasant weather lifted mood and memory only in people who spent more than half an hour outside (Keller et al. 2005); green exercise gave most of its effect early, the largest gain in self-esteem at five minutes (Barton & Pretty 2010); each extra hour a day outdoors in daylight went with less low mood, easier waking and fewer insomnia symptoms in about 400,000 adults (Burns et al. 2021). Control over one's own working time improves the balance between work and the rest of life (Nijp et al. 2012, 53 studies), while work in the evening costs detachment and sleep (finding 21). In ME/CFS and long COVID, exertion can bring a crash that starts within a day and peaks within three (Stussman et al. 2020; Appelman et al. 2024). Rule: Free time is one press: work waits, only the people you chose reach you, and the work day may end later by as much, never into the wind-down nor past a latest end; movement is offered as a choice, never with "Days kept even", the setting for energy-limiting illness (pauses.md; detail: global-pause.md, criteria GP1–GP21).
  34. The phone is the hardest channel, and an expected call is easier than a surprise. 78 % of autistic adults avoid the phone where they can, and 61 % avoid voicemail too (Doherty et al. 2022); people with admin anxiety find the telephone the hardest channel (Holkar, Evans & Langston 2018); intolerance of uncertainty and anxiety go together in autism (r = 0.62: Jenkinson, Milne & Thompson 2020); text reminders keep appointments as well as calls do (Gurol-Urganci et al. 2013). No study tests call screening itself: the case rests on stated preferences and a plausible mechanism. Rule: on a phone, calls follow the same rules as mail: a call that may not reach you now is declined plainly and goes to voicemail, and the Porch says afterwards who called; Sioul never answers and never listens; the emergency services calling back, and a second call within fifteen minutes, always ring (attention.md; detail: call-screening.md, criteria CS1–CS21).

  35. A spam filter learns best from one's own mail, and the labels one gives matter more than the model. Filters trained on other people's mail did about a hundred times worse than filters learning inside one mailbox, in the blind tests of the TREC spam tracks (Cormack 2008; Sculley & Wachman 2007); labelling about 1 % of one's mail near the threshold gave 20 to 150 times less missed spam than the provider's own filter (Cormack & Mojdeh 2009); and a pass of SPF, DKIM or DMARC is almost no evidence, since 89 % of unwanted mail passes (Cloudflare 2023). Rule: Sioul's filter learns on your computer, from your own mail and your own word; it judges strangers' mail only, never people you know, codes or projects; until you choose, it moves nothing (spam-filter.md; detail: research/spam-filter.md).

  36. What an activity costs has several sides, and a rating needs words.

    • NICE's guideline for ME/CFS says that energy management "includes all types of activity (cognitive, physical, emotional and social)", and asks that the plan weigh the "effect of environmental factors, including sensory stimulation" (NICE NG206, 2021, recommendations 1.11.2 and 1.11.3; the ME Association reproduces the section in full). Thinking, feelings, and the body and the senses are three of those sides.
    • Admin adds worry: its burden weighs more on people with ADHD, anxiety or chronic pain (finding 3: Christensen et al. 2020; Bell et al. 2023), and dread runs before and after the task itself (capacity-budget.md, criterion 2).
    • A single rating from 0 to 10, with words at fixed numbers, is Borg's CR10 scale (Borg 1982). Session-RPE rates a whole training session on it, and the rating times the minutes tracked a heart-rate load closely (Foster et al. 2001).
    • Rule: four costs, thinking, feelings, worry, and body and senses, and a gain, each from 0 to 10 and unsaid until said; CR10's words beside the number; in each, a concrete question, feelings asked about through what stirs them. Built: capacity.md.

  37. Portals move the work onto the person and gather everyone's data in one store.

    • Mail is encrypted between servers only when both agree, and can be downgraded to clear text (RFC 3207; RFC 8461; Durumeric et al. 2015, in 2015 data); the GDPR asks for security appropriate to the risk, encryption among its first measures (art. 32), and the CNIL calls everyday channels such as mail "rarely" safe for personal data without more (2024). Organisations answered with portals.
    • The Défenseur des droits found that dematerialisation moved work and responsibility onto people, following mail "sur une ou plusieurs messageries" included, and that a personal space not consulted in time risks a "rupture des droits" (2022, pp. 23, 58–59); in its 2025 report (April 2026), 61 % of people reported difficulties with their administrative procedures in 2024 (39 % in 2016), authentication systems "changent sans cesse", and without help from family or friends nearly one in four gives up rights from lost time, discouragement or anxiety. Digital self-service makes citizens do caseworkers' tasks, and can also lighten their burden (Madsen, Lindgren & Melin 2022; qualitative).
    • Remembering passwords and transcribing codes are "a very high or impossible burden" for some people with cognitive disabilities; password managers are a named remedy (W3C, WCAG 2.2 SC 3.3.8; COGA 2021). ADHD, depression, PTSD and burnout come, on average, with weaker working memory, planning or attention (finding 29; Willcutt et al. 2005; Snyder 2013; Scott et al. 2015; Deligkaris et al. 2014).
    • The stores behind portals were breached from 2021 to 2026, often through compromised or usurped logins, or through providers (AP-HP's file-sharing service, the Assurance Maladie's professional portal, a provider of Pôle emploi, Viamedis and Almerys, France Travail and Cap emploi, FICOBA, France Titres, Vacaf); the CNIL counted twice as many breaches affecting over a million people in 2024 as the year before (CNIL 2025).
    • No study measures portal use, or missed portal messages, in people with these conditions: the link is inferred from the burden and executive-function findings, and stated as such.
    • Rule: carry the portals in one place meanwhile (Sites: each logged in once, its news held for the person's hours and said in one line on the Porch), and support what lets letters come back to mail end-to-end encrypted: OpenPGP with Autocrypt and the Web Key Directory, keys on an OpenPGP card (on computers; not yet on a phone), a password manager, FIDO2 keys. The person has one window; no organisation needs one store. Detail: research/secure-mailboxes.md.

Sources

  • Kushlev & Dunn (2015), Checking email less frequently reduces stress, Computers in Human Behavior 43.
  • Fitz, Kushlev, Jagannathan, Lewis, Paliwal & Ariely (2019), Batching smartphone notifications can improve well-being, Computers in Human Behavior 101.
  • Mark, Voida & Cardello (2012), "A pace not dictated by electrons": an empirical study of work without email, CHI 2012.
  • Mark, Gudith & Klocke (2008), The cost of interrupted work: more speed and stress, CHI 2008.
  • Stothart, Mitchum & Yehnert (2015), The attentional cost of receiving a cell phone notification, J. Exp. Psychol.: Human Perception and Performance 41(4).
  • Moynihan, Herd & Harvey (2015), Administrative burden: learning, psychological, and compliance costs in citizen-state interactions, JPART 25(1).
  • Christensen, Aarøe, Baekgaard, Herd & Moynihan (2020), Human capital and administrative burden: the role of cognitive resources in citizen-state interactions, Public Administration Review 80(1).
  • Sirois & Pychyl (2013), Procrastination and the priority of short-term mood regulation, Social and Personality Psychology Compass 7(2).
  • Sweeny, Melnyk, Miller & Shepperd (2010), Information avoidance: who, what, when, and why, Review of General Psychology 14(4).
  • Golman, Hagmann & Loewenstein (2017), Information avoidance, Journal of Economic Literature 55(1).
  • Carleton (2012), The intolerance of uncertainty construct in the context of anxiety disorders, Expert Review of Neurotherapeutics 12(8).
  • Boulter, Freeston, South & Rodgers (2014), Intolerance of uncertainty as a framework for understanding anxiety in children and adolescents with autism spectrum disorders, JADD 44.
  • Buckle, Leadbitter, Poliakoff & Gowen (2021), "No way out except from external intervention": first-hand accounts of autistic inertia, Frontiers in Psychology 12.
  • Barkley (2012), Executive Functions: What They Are, How They Work, and Why They Evolved, Guilford.
  • Campbell, Çerçi & Cecchinato (2024), ADHD and knowledge work: exploring strategies, challenges and opportunities for AI, INTERACT 2023.
  • Gollwitzer & Sheeran (2006), Implementation intentions and goal achievement: a meta-analysis, Advances in Experimental Social Psychology 38.
  • Bandura & Schunk (1981), Cultivating competence, self-efficacy, and intrinsic interest through proximal self-motivation, JPSP 41(3).
  • Eagle & Ringland (2024), An investigation of body doubling with neurodivergent participants, ACM Transactions on Accessible Computing.
  • Amabile & Kramer (2011), The Progress Principle, Harvard Business Review Press.
  • Silverman & Barasch (2023), On or off track: how (broken) streaks affect consumer decisions, Journal of Consumer Research 49(6).
  • Chen, McDonald, Zou, Tseng, Roundy, Tamersoy, Schaub, Ristenpart & Dell (2022), Trauma-informed computing: towards safer technology experiences for all, CHI 2022.
  • W3C (2021), Making content usable for people with cognitive and learning disabilities, https://www.w3.org/TR/coga-usable/
  • Pun (2016), Dos and don'ts on designing for accessibility, GOV.UK accessibility blog.
  • Rudberg Selin, Unéus & Knudsen (2026), "Chasing shadows": understanding personal data externalization and self-tracking for neurodivergent individuals, CHI 2026.
  • Jang, Moharana, Carrington & Begel (2024), "It's the only thing I can trust": envisioning large language model use by autistic workers for communication assistance, CHI 2024.
  • Steinbrenner et al. (2020), Evidence-Based Practices for Children, Youth, and Young Adults with Autism, NCAEP, University of North Carolina.
  • Scheibehenne, Greifeneder & Todd (2010), Can there ever be too many options? A meta-analytic review of choice overload, Journal of Consumer Research 37(3).
  • Chernev, Böckenholt & Goodman (2015), Choice overload: a conceptual review and meta-analysis, Journal of Consumer Psychology 25(2).
  • Wohl, Pychyl & Bennett (2010), I forgive myself, now I can study: how self-forgiveness for procrastinating can reduce future procrastination, Personality and Individual Differences 48(7).
  • Beaton, Sirois & Milne (2022), Experiences of criticism in adults with ADHD: a qualitative study, PLoS ONE 17(2).
  • Ghibellini & Meier (2025), Interruption, recall and resumption: a meta-analysis of the Zeigarnik and Ovsiankina effects, Humanities and Social Sciences Communications 12.
  • Leroy & Glomb (2018), Tasks interrupted: how anticipating time pressure on resumption of an interrupted task causes attention residue and low performance on interrupting tasks and how a "ready-to-resume" plan mitigates the effects, Organization Science 29(3).
  • Trafton, Altmann, Brock & Mintz (2003), Preparing to resume an interrupted task: effects of prospective goal encoding and retrospective rehearsal, International Journal of Human-Computer Studies 58(5).
  • Wennberg, Janeslätt, Kjellberg, Gustafsson & Hirvikoski (2018), Effectiveness of time-related interventions in children with ADHD aged 9–15 years: a randomized controlled study, European Child & Adolescent Psychiatry 27.
  • Raymaker et al. (2020), "Having all of your internal resources exhausted beyond measure and being left with no clean-up crew": defining autistic burnout, Autism in Adulthood 2(2).
  • Chen, Meng & Nie (2026), "Not just me and my to-do list": understanding challenges of task management for adults with ADHD and the need for AI-augmented social scaffolds, CSCW 2026.
  • Kruger & Evans (2004), If you don't want to be late, enumerate: unpacking reduces the planning fallacy, Journal of Experimental Social Psychology 40(5).
  • Wendsche & Lohmann-Haislah (2017), A meta-analysis on antecedents and outcomes of detachment from work, Frontiers in Psychology 7:2072.
  • Sonnentag, Arbeus, Mahn & Fritz (2014), Exhaustion and lack of psychological detachment from work during off-job time: moderator effects of time pressure and leisure experiences, J. Occup. Health Psychol. 19(2).
  • Butts, Becker & Boswell (2015), Hot buttons and time sinks: the effects of electronic communication during nonwork time on emotions and work-nonwork conflict, Academy of Management Journal 58(3).
  • Becker, Belkin, Conroy & Tuskey (2021), Killing me softly: organizational e-mail monitoring expectations' impact on employee and significant other well-being, Journal of Management 47(4).
  • Syrek, Weigelt, Peifer & Antoni (2017), Zeigarnik's sleepless nights: how unfinished tasks at the end of the week impair employee sleep on the weekend through rumination, J. Occup. Health Psychol. 22(2).
  • Wendsche, Weigelt & Syrek (2026), Unfinished work tasks and work-related thoughts during off-job time: meta-analysis of the Zeigarnik effect in a work-recovery context, Anxiety, Stress & Coping 39(4).
  • Masicampo & Baumeister (2011), Consider it done! Plan making can eliminate the cognitive effects of unfulfilled goals, J. Pers. Soc. Psychol. 101(4).
  • Smit (2016), Successfully leaving work at work: the self-regulatory underpinnings of psychological detachment, J. Occup. Organ. Psychol. 89(3).
  • Kim, Thibodeau & Jorgensen (2011), Shame, guilt, and depressive symptoms: a meta-analytic review, Psychological Bulletin 137(1).
  • Leach & Cidam (2015), When is shame linked to constructive approach orientation? A meta-analysis, J. Pers. Soc. Psychol. 109(6).
  • Kluger & DeNisi (1996), The effects of feedback interventions on performance, Psychological Bulletin 119(2).
  • Kirby, Day & Sagar (2019), The "flow" of compassion: a meta-analysis of the fears of compassion scales and psychological functioning, Clinical Psychology Review 70.
  • Rules 24 to 28, and the details of rules 32 to 35: the full references, with their DOIs, are in the notes they link to.
  • Findings 9 and 29 to 37, with their DOIs; "(abstract only)" marks a source read at its abstract, the others were read in full or, for the older ones, in the notes that cite them:
    • Adams, Cartmell, Foster, Foxwell, Holker, Pearson, Stewart, Cowling & Kitching (2019), Understanding self-employment for people with disabilities and health conditions, Department for Work and Pensions Research Report 974. https://assets.publishing.service.gov.uk/media/5cc70e12ed915d5dc8d3cfcf/understanding-self-employment-for-people-with-disabilities-and-health-conditions.pdf
    • American Psychiatric Association (2013), Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5). doi:10.1176/appi.books.9780890425596 (the criteria read as quoted by the US National Center for PTSD, https://www.ptsd.va.gov/professional/treat/essentials/dsm5_ptsd.asp, and by de la Peña, Pan, Thai & Alisso 2020, Brain Sciences 10(5), doi:10.3390/brainsci10050292)
    • Appelman, Charlton, Goulding et al. (2024), Muscle abnormalities worsen after post-exertional malaise in long COVID, Nature Communications 15. doi:10.1038/s41467-023-44432-3
    • Barton & Pretty (2010), What is the best dose of nature and green exercise for improving mental health? A multi-study analysis, Environmental Science & Technology 44(10). doi:10.1021/es903183r
    • Beauchaine, Ben-David & Bos (2020), ADHD, financial distress, and suicide in adulthood: a population study, Science Advances 6(40). doi:10.1126/sciadv.aba1551
    • Bell, Christensen, Herd & Moynihan (2023), Health in citizen-state interactions: how physical and mental health problems shape experiences of administrative burden and reduce take-up, Public Administration Review 83(2). doi:10.1111/puar.13568
    • Bond, Braverman & Evans (2019), The benefits assault course: making the UK benefits system more accessible for people with mental health problems, Money and Mental Health Policy Institute (report, not peer-reviewed). https://www.moneyandmentalhealth.org/publications/benefits/
    • Borg (1982), Psychophysical bases of perceived exertion, Medicine & Science in Sports & Exercise 14(5). doi:10.1249/00005768-198205000-00012 (abstract only)
    • Burns, Saxena, Vetter, Phillips, Lane & Cain (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, Journal of Affective Disorders 295. doi:10.1016/j.jad.2021.08.056
    • Chen, McDonald, Zou, Tseng, Roundy, Tamersoy, Schaub, Ristenpart & Dell (2022), Trauma-informed computing (above). doi:10.1145/3491102.3517475
    • Cloudflare (2023), Phishing threats report. https://www.cloudflare.com/press-releases/2023/cloudflare-harnesses-the-power-of-its-global-network-to-identify-top
    • CNIL (2024), Sécurité : sécuriser les échanges avec l'extérieur, guide de la sécurité des données personnelles, 14 March 2024. https://www.cnil.fr/fr/securite-securiser-les-echanges-avec-lexterieur
    • CNIL (2025), Violations massives de données en 2024 : quels sont les principaux enseignements et mesures à prendre ?, 28 January 2025. https://www.cnil.fr/fr/violations-massives-de-donnees-en-2024-quels-sont-les-principaux-enseignements-mesures-a-prendre
    • Cormack (2008), Email spam filtering: a systematic review, Foundations and Trends in Information Retrieval 1(4). doi:10.1561/1500000006
    • Cormack & Mojdeh (2009), Autonomous personal filtering improves global spam filter performance, CEAS 2009. https://cormack.uwaterloo.ca/ceas09.pdf
    • Davies, Melinek, Livesey, Killick, Sam, Romualdez, Pellicano & Remington (2025), "I did what I could to earn some money and be of use": a qualitative exploration of autistic people's journeys to career success and fulfilment, Autism 29(4). doi:10.1177/13623613241292177
    • Défenseur des droits (2022), Dématérialisation des services publics : trois ans après, où en est-on ?, report, February 2022, pp. 23 and 58–59. https://www.vie-publique.fr/files/rapport/pdf/283832.pdf
    • Défenseur des droits (2026), Rapport annuel d'activité 2025, April 2026. https://www.defenseurdesdroits.fr/sites/default/files/2026-04/ddd_rapport-annuel-2025_20260407.pdf
    • Deligkaris, Panagopoulou, Montgomery & Masoura (2014), Job burnout and cognitive functioning: a systematic review, Work & Stress 28(2). doi:10.1080/02678373.2014.909545 (abstract only)
    • Doherty, Neilson, O'Sullivan, Carravallah, Johnson, Cullen & Shaw (2022), Barriers to healthcare and self-reported adverse outcomes for autistic adults: a cross-sectional study, BMJ Open 12(2). doi:10.1136/bmjopen-2021-056904
    • Durumeric, Adrian, Mirian, Kasten, Bursztein, Lidzborski, Thomas, Eranti, Bailey & Halderman (2015), Neither snow nor rain nor MITM... An empirical analysis of email delivery security, ACM Internet Measurement Conference 2015. doi:10.1145/2815675.2815695, https://conferences2.sigcomm.org/imc/2015/papers/p27.pdf
    • Evans (2003), Studying the studies: an overview of recent research into taxation operating costs, eJournal of Tax Research 1(1). https://web.archive.org/web/20110218203518id_/http://www.atax.unsw.edu.au/ejtr/content/issues/previous/full_edition_v1n1.pdf
    • Foster, Florhaug, Franklin, Gottschall, Hrovatin, Parker, Doleshal & Dodge (2001), A new approach to monitoring exercise training, Journal of Strength and Conditioning Research 15(1). PubMed 11708692 (abstract only)
    • Gurol-Urganci, de Jongh, Vodopivec-Jamsek, Atun & Car (2013), Mobile phone messaging reminders for attendance at healthcare appointments, Cochrane Database of Systematic Reviews (12). doi:10.1002/14651858.CD007458.pub3
    • Holkar, Evans & Langston (2018), Access essentials: giving people with mental health problems equal access to vital services, Money and Mental Health Policy Institute (report, not peer-reviewed). https://www.moneyandmentalhealth.org/publications/accessessentials/
    • Howard & Sedgewick (2021), "Anything but the phone!": communication mode preferences in the autism community, Autism 25(8). doi:10.1177/13623613211014995
    • Jenkinson, Milne & Thompson (2020), The relationship between intolerance of uncertainty and anxiety in autism: a systematic literature review and meta-analysis, Autism 24(8). doi:10.1177/1362361320932437
    • Keller, Fredrickson, Ybarra, Côté, Johnson, Mikels, Conway & Wager (2005), A warm heart and a clear head: the contingent effects of weather on mood and cognition, Psychological Science 16(9). doi:10.1111/j.1467-9280.2005.01602.x
    • Kessels (2003), Patients' memory for medical information, Journal of the Royal Society of Medicine 96(5). doi:10.1177/014107680309600504
    • Kitching (2014), Entrepreneurship and self-employment by people with disabilities, OECD Local Economic and Employment Development Working Papers. doi:10.1787/a1ef5b0b-en
    • Koerts, Bangma, Mette, Tucha & Tucha (2023), Strengths and weaknesses of everyday financial knowledge and judgment skills of adults with ADHD, International Journal of Environmental Research and Public Health 20(5). doi:10.3390/ijerph20054656 (abstract only)
    • Lerner, Verheul & Thurik (2019), Entrepreneurship and attention deficit/hyperactivity disorder: a large-scale study involving the clinical condition of ADHD, Small Business Economics 53(2). doi:10.1007/s11187-018-0061-1
    • Löthberg, Hirvikoski, Girdler, Bölte & Jonsson (2024), Support in daily living for young adults with neurodevelopmental conditions in Sweden: a qualitative description of current practice, Journal of Autism and Developmental Disorders 54(8). doi:10.1007/s10803-023-06014-6
    • Madsen, Lindgren & Melin (2022), The accidental caseworker: how digital self-service influences citizens' administrative burden, Government Information Quarterly 39(1), 101653. doi:10.1016/j.giq.2021.101653 (abstract only)
    • McQueenie, Ellis, Williamson & Wilson (2024), Attention-deficit/hyperactivity disorder and serial missed appointments in general practice, PLOS Mental Health 1(2). doi:10.1371/journal.pmen.0000045 (abstract only)
    • ME Association, Energy management: NICE guideline NG206, section 1.11, reproduced in full. https://meassociation.org.uk/nice-guidelines/items/energy-management/ (read 8 October 2026)
    • Milanez & Bratta (2019), Taxation and the future of work: how tax systems influence choice of employment form, OECD Taxation Working Papers 41. doi:10.1787/20f7164a-en
    • NICE (2021), Myalgic encephalomyelitis (or encephalopathy)/chronic fatigue syndrome: diagnosis and management, NICE guideline NG206, recommendations 1.11.2 and 1.11.3. https://www.nice.org.uk/guidance/ng206/chapter/Recommendations (read 8 October 2026)
    • Nijp, Beckers, Geurts, Tucker & Kompier (2012), Systematic review on the association between employee worktime control and work–non-work balance, health and well-being, and job-related outcomes, Scandinavian Journal of Work, Environment & Health 38(4). doi:10.5271/sjweh.3307
    • Pagán (2009), Self-employment among people with disabilities: evidence for Europe, Disability & Society 24(2). doi:10.1080/09687590802652504 (abstract only)
    • Paris, Lodestone, Houser & Lewis (2025), "Shutdowns are like you're stuck on the blue screen of death": a metaphor analysis of autistic shutdowns, Autism in Adulthood 8(4). doi:10.1089/aut.2024.0193
    • Parker, Sawant, Flannery et al. (2023), Effect of using a structured pacing protocol on post-exertional symptom exacerbation and health status in a longitudinal cohort with the post-COVID-19 syndrome, Journal of Medical Virology 95(1). doi:10.1002/jmv.28373
    • Raymaker, McDonald, Ashkenazy, Gerrity, Baggs, Kripke, Hourston & Nicolaidis (2017), Barriers to healthcare: instrument development and comparison between autistic adults and adults with and without other disabilities, Autism 21(8). doi:10.1177/1362361316661261
    • Regulation (EU) 2016/679, the General Data Protection Regulation, articles 5(1)(f), 32 and 34, Official Journal of the European Union L 119, 4 May 2016. https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32016R0679
    • RFC 3207 (2002), SMTP Service Extension for Secure SMTP over Transport Layer Security, section 6. https://www.rfc-editor.org/rfc/rfc3207
    • RFC 8461 (2018), SMTP MTA Strict Transport Security (MTA-STS), section 1. https://www.rfc-editor.org/rfc/rfc8461
    • Roberts, Stuart, Allan & Gumley (2024), "It's like the Sword of Damocles": a trauma-informed framework analysis of individuals' experiences of assessment for the Personal Independence Payment benefit in the UK, Journal of Social Policy 53(4). doi:10.1017/S0047279422000800 (abstract only)
    • Rose, Bisson, Churchill & Wessely (2002), Psychological debriefing for preventing post traumatic stress disorder (PTSD), Cochrane Database of Systematic Reviews (2). doi:10.1002/14651858.CD000560
    • SAMHSA (2014), SAMHSA's Concept of Trauma and Guidance for a Trauma-Informed Approach, HHS Publication No. (SMA) 14-4884, read in SAMHSA's official PDF as archived on 4 September 2025 (SAMHSA no longer serves it). https://web.archive.org/web/20250904005813id_/https://library.samhsa.gov/sites/default/files/sma14-4884.pdf
    • Scott, Matt, Wrocklage, Crnich, Jordan, Southwick, Krystal & Schweinsburg (2015), A quantitative meta-analysis of neurocognitive functioning in posttraumatic stress disorder, Psychological Bulletin 141(1). doi:10.1037/a0038039
    • Scullion & Curchin (2022), Examining veterans' interactions with the UK social security system through a trauma-informed lens, Journal of Social Policy 51(1). doi:10.1017/S0047279420000719
    • Sculley & Wachman (2007), Relaxed online SVMs for spam filtering, SIGIR 2007. doi:10.1145/1277741.1277813
    • Shields, Sazma & Yonelinas (2016), The effects of acute stress on core executive functions: a meta-analysis and comparison with cortisol, Neuroscience & Biobehavioral Reviews 68. doi:10.1016/j.neubiorev.2016.06.038
    • Smith & Freyd (2014), Institutional betrayal, American Psychologist 69(6). doi:10.1037/a0037564 (abstract only)
    • Snyder (2013), Major depressive disorder is associated with broad impairments on neuropsychological measures of executive function: a meta-analysis and review, Psychological Bulletin 139(1). doi:10.1037/a0028727 (abstract only)
    • Spencer, Faraone, Bogucki, Pope, Uchida, Milad, Spencer, Woodworth & Biederman (2016), Examining the association between posttraumatic stress disorder and attention-deficit/hyperactivity disorder: a systematic review and meta-analysis, Journal of Clinical Psychiatry 77(1). doi:10.4088/JCP.14r09479 (abstract only)
    • Stussman, Williams, Snow, Gavin, Scott, Nath & Walitt (2020), Characterization of post-exertional malaise in patients with myalgic encephalomyelitis/chronic fatigue syndrome, Frontiers in Neurology 11. doi:10.3389/fneur.2020.01025
    • Suriano (2026), A systematic review on the association between ADHD and procrastination, Research in Developmental Disabilities 175. doi:10.1016/j.ridd.2026.105342 (abstract only)
    • U.S. Bureau of Labor Statistics (2026), People with a disability: labor force characteristics — 2025, news release USDL-26-0364. https://www.bls.gov/news.release/pdf/disabl.pdf
    • Verheul, Rietdijk, Block, Franken, Larsson & Thurik (2016), The association between attention-deficit/hyperactivity (ADHD) symptoms and self-employment, European Journal of Epidemiology 31(8). doi:10.1007/s10654-016-0159-1
    • W3C (2024), Web Content Accessibility Guidelines (WCAG) 2.2, success criterion 3.3.8, Accessible Authentication (Minimum), and Understanding SC 3.3.8. https://www.w3.org/TR/WCAG22/#accessible-authentication-minimum and https://www.w3.org/WAI/WCAG22/Understanding/accessible-authentication-minimum.html
    • Webermann, Rupcic, Hoffman, Portnoy, Rosen, Monteith, Holliday & Murdoch (2026), "I had to relive all of that again": retraumatization, invalidation, and institutional betrayal among veterans filing military sexual trauma-related disability claims, Journal of Trauma & Dissociation 27(3). doi:10.1080/15299732.2026.2643660 (abstract only)
    • Wegner, Broome & Blumberg (1997), Ironic effects of trying to relax under stress, Behaviour Research and Therapy 35(1). doi:10.1016/S0005-7967(96)00078-2
    • Wiklund, Patzelt & Dimov (2016), Entrepreneurship and psychological disorders: how ADHD can be productively harnessed, Journal of Business Venturing Insights 6. doi:10.1016/j.jbvi.2016.07.001
    • Willcutt, Doyle, Nigg, Faraone & Pennington (2005), Validity of the executive function theory of attention-deficit/hyperactivity disorder: a meta-analytic review, Biological Psychiatry 57(11). doi:10.1016/j.biopsych.2005.02.006 (abstract only)
    • World Health Organization (2024), Clinical descriptions and diagnostic requirements for ICD-11 mental, behavioural and neurodevelopmental disorders. https://iris.who.int/handle/10665/375767
    • World Health Organization, War Trauma Foundation & World Vision International (2011), Psychological first aid: guide for field workers, ISBN 978-92-4-154820-5. https://www.who.int/publications/i/item/9789241548205