Decision fatigue, explained after the replication crisis

"Decision fatigue" is one of the most searched ideas in productivity, and nearly every article explaining it still cites the same two things: a parole-board study in which judges supposedly got harsher before lunch, and the claim that willpower is a limited resource that drains through the day. Both have been tested since. This page walks through the original claims, the replications and critiques, what survives, and what to do with a day's worth of decisions once you stop believing in a tank.

The claim and where it came from

The idea has two roots. The first is ego depletion: in a 1998 paper, Baumeister and colleagues reported that an initial act of self-control, such as resisting a plate of cookies and eating radishes instead, left participants persisting about eight minutes less on an unsolvable puzzle afterwards C^[1]. The second is decision fatigue proper: in 2008, Vohs and colleagues extended the model to choice, reporting across seven experiments that making many decisions impaired later self-control and persistence C^[2].

From there the idea travelled into advice columns: wear the same clothes, eat the same breakfast, make your important decisions in the morning. The studies behind it were small by modern standards and sit inside a literature that was about to be tested at scale.

The parole-board study and its critique

The most quoted evidence is Danziger, Levav and Avnaim-Pesso's 2011 analysis of 1,112 rulings by eight Israeli parole judges. Favourable rulings ran at about 65 % right after a break and fell towards zero before the next one C^[3]. It is an observational study, and its headline has been challenged on two fronts.

First, case ordering was not random. Weinshall-Margel and Shapard pointed out that prisoners without legal representation tended to be scheduled last within each session, and unrepresented cases are rejected more often regardless of the hour; ordering by representation and ruling type explains part of the decline B^[4]. Second, the size of the effect is implausible. Glöckner's simulations showed that judges who rationally time breaks around difficult cases, combined with the ordering effects, can reproduce the "hungry judge" curve, and that the true effect, if any, must be far smaller than a drop from 65 % to near zero B^[5]. A single meal does not plausibly remove two thirds of favourable rulings.

The honest status of the parole study: an interesting pattern in observational data with a known confound and an effect size no psychological mechanism can account for. It should not be quoted as proof that people decide worse before lunch.

Ego depletion: the 2015, 2016 and 2021 tests

The depletion model itself has been tested three times at a scale the original studies never approached.

Test	Design	Result	Grade
Carter et al. 2015	Meta-analysis of the depletion literature with corrections for small-study effects	Once publication bias is accounted for, the effect is indistinguishable from zero B[6]	B
Hagger et al. 2016	Preregistered replication, 23 labs, 2,141 participants	d = 0.04, 95 % CI −0.07 to 0.15 A[7]	A
Vohs et al. 2021	Preregistered replication, 36 labs, 3,531 participants, paradigm chosen by proponents	d of about 0.06, below any practically meaningful threshold A[8]	A

Two caveats cut in opposite directions. Each replication tested one paradigm, and proponents argue others might behave differently. And none of this shows that self-control is unlimited; it shows that the specific lab effect, "a bit of self-control now impairs self-control a few minutes later", is tiny or absent. Pignatiello and colleagues' conceptual analysis adds that "decision fatigue" has no agreed definition or measurement, and that the evidence under the label is mixed and dominated by self-report B^[9].

What survives: fatigue, time pressure, choice architecture

If willpower is not a tank, why do afternoons feel worse and why do people make sloppier choices at the end of a long day? Several better-supported mechanisms explain the experience without the metaphor.

Time-on-task fatigue is real, and it is motivational. Performance on demanding tasks declines over hours, and recovers when the reward for continuing rises. Boksem and Tops frame fatigue as a signal about the cost–benefit of continuing rather than an empty store B^[10]. In Hopstaken's two-hour experiment, the physiological and performance signs of disengagement were largely reversed when a reward was introduced B^[11]. Shenhav and colleagues' model of cognitive control describes the same economics: the brain rations effortful control according to its expected payoff B^[12]. One recent study reported that around six hours of demanding cognitive work raised glutamate in the lateral prefrontal cortex and shifted choices towards low-effort options, but it is a single study of 40 people with a novel mechanism, and it needs replication before it supports any rule about workday length C^[13].

Sleep loss degrades the hardware. Partial sleep restriction produces small-to-moderate deficits in sustained attention and executive function that accumulate over nights A^[14], and executive functions are the faculty you use to deliberate A^[15]. A week of short nights predicts worse afternoon decisions without any need for a depletion theory; see productivity and sleep.

Under pressure, people protect the main task and pay elsewhere. Hockey's compensatory-control model describes how performance can look intact under high workload while strain accumulates and secondary tasks suffer B^[16]. The "decision" that goes wrong at 5 pm is often the secondary one.

Choice overload is itself contested. The idea that more options paralyse people has a famous jam-stand anecdote behind it. A meta-analysis of 50 experiments found the average effect of large assortments on choice was essentially zero, with high variability between studies B^[17]. A later meta-analysis argued the effect appears under specific conditions: when the choice is complex, the options hard to compare, people are uncertain about their preferences, or the stakes of choosing well are high B^[18]. Fewer options help in those conditions; they are not a universal rule.

Habits do the work that willpower is credited with. Across 102 studies, trait self-control predicted behaviour more strongly for automatic, habitual behaviours than for effortful inhibition A^[19]. People who seem disciplined mostly have fewer decisions to make.

What is myth

• "Willpower is a limited resource that depletes through the day." Two preregistered multi-site replications found d of 0.04 to 0.06 ^[7][8]. Mental fatigue exists; a draining tank is not what the evidence describes.
• "Judges deny parole before lunch because of decision fatigue." Confounded by case ordering ^[4] and implausibly large ^[5].
• "Make important decisions in the morning." No general evidence supports this. Time-of-day effects on cognition are real but depend on your chronotype, with evening types performing better later ^[20]. Morning is the right time for morning people.
• "Decision fatigue is a recognised clinical condition." It has no consistent definition or measure in the research literature ^[9].
• "Fewer choices are always better." The average choice-overload effect across experiments is near zero; it appears under specific conditions ^[17][18].

Practical defaults that need no willpower theory

The replications do not mean nothing can be done. They change what the sensible actions are, and why they work.

1. Reduce the number of trivial decisions by default, not by discipline. The same breakfast and a standing weekly plan work because they remove decisions, and habits predict behaviour better than inhibition does ^[19]. They do not work because they "save willpower".
2. Use if-then plans for the decisions that matter. Implementation intentions, "if it is 4 pm and I have not started the report, then I close chat and open the draft", have a medium-to-large effect on goal attainment across 94 studies, d = 0.65 A^[21].
3. Put high-stakes choices at your personal peak, not at 9 am. Match the decision to your chronotype rather than to a folk rule ^[20].
4. Treat end-of-day sloppiness as a fatigue and sleep signal. If a reward or a change of task restores you, it was motivational fatigue ^[11]; if nothing does and the week has been short on sleep, it is the hardware ^[14].
5. Structure complex choices rather than shrinking them. When options are hard to compare, give yourself criteria first; that is the condition under which choice overload actually appears ^[18].
6. Audit who makes which decisions. Many "decision fatigue" complaints at work are workload complaints: too many decisions routed to one person. Delegation changes the count; morning routines do not.

A health note: persistent exhaustion, indecision and low mood that do not lift with rest can be features of depression, burnout or sleep disorders. If that describes your weeks rather than your afternoons, see a clinician; this page is about research on normal fatigue, not a diagnosis.

What to do next

The useful version of "decision fatigue" is a design problem: which decisions deserve deliberation, which should be defaults, and which belong to someone else. The AI productivity OS Builder turns thirteen questions about your role, meeting load and priorities into a daily rhythm, a priority system and a set of defaults, so recurring choices are made once rather than every day. Build your defaults there, then revisit them in four weeks.

Related: why cognitive load at work is the better-defined concept for most of what people call decision fatigue, and the broader science of attention.

Frequently asked questions

What does decision fatigue mean?

In popular use, the idea that the quality of your decisions declines after making many of them, because self-control is a limited resource. In the research literature the term has no agreed definition or measure ^[9], and the limited-resource model behind it produced effects of d = 0.04 to 0.06 in two large preregistered replications ^[7][8]. Time-on-task fatigue and sleep loss are the better-supported explanations for worse late-day decisions.

Is decision fatigue real?

The experience is real; the mechanism usually offered for it is not well supported. People do tire over hours of demanding work, and that fatigue behaves like a motivational signal that reverses when the task becomes more rewarding ^[10][11]. What failed replication is the specific claim that a short act of self-control depletes a willpower reserve ^[7][8].

Is decision fatigue linked to ADHD?

ADHD involves differences in executive function, the inhibition, working memory and flexibility used in deliberate choice ^[15], so decisions can cost more effort and feel more draining. That is a separate matter from the ego-depletion debate, and the studies on this page did not test clinical populations. If decision-making difficulty is affecting daily life, a clinician can assess it; this page is not a diagnostic resource.

Why do women report more decision fatigue?

Searches for this often concern the mental load of managing a household on top of paid work. That is a workload and division-of-labour question, and it is plausible that whoever carries most of the recurring decisions experiences more fatigue from them. The ego-depletion and parole-study literature on this page does not address it, and we have not graded evidence on it here. The practical defaults above apply regardless of who holds the load.

Should I make important decisions in the morning?

Only if you are a morning type. Executive performance is better at the time of day that matches your chronotype, and evening types do better later ^[20]. The general advice to decide early comes from the depletion model, which did not replicate ^[7][8].

Do fewer choices really make people happier and more decisive?

Not in general. A meta-analysis of 50 experiments found the average effect of larger assortments on choice to be essentially zero ^[17]. Choice overload shows up under specific conditions, such as complex, hard-to-compare options or uncertain preferences ^[18]. In those cases, setting criteria before choosing helps more than cutting options.

Sources and evidence

1. Ego depletion: Is the active self a limited resource? — Baumeister, R. F.; Bratslavsky, E.; Muraven, M.; Tice, D. M. (1998), RCT C doi.org/10.1037/0022-3514.74.5.1252 (retrieved 2026-11-09)
2. Making choices impairs subsequent self-control: A limited-resource account of decision making, self-regulation, and active initiative — Vohs, K. D.; Baumeister, R. F.; Schmeichel, B. J.; Twenge, J. M.; Nelson, N. M.; Tice, D. M. (2008), RCT C doi.org/10.1037/0022-3514.94.5.883 (retrieved 2026-11-09)
3. Extraneous factors in judicial decisions — Danziger, S.; Levav, J.; Avnaim-Pesso, L. (2011), observational C doi.org/10.1073/pnas.1018033108 (retrieved 2026-11-09)
4. Overlooked factors in the analysis of parole decisions — Weinshall-Margel, K.; Shapard, J. (2011), observational B doi.org/10.1073/pnas.1110910108 (retrieved 2026-11-09)
5. The irrational hungry judge effect revisited: Simulations reveal that the magnitude of the effect is overestimated — Glöckner, A. (2016), review B doi.org/10.1017/S1930297500004812 (retrieved 2026-11-09)
6. A series of meta-analytic tests of the depletion effect: Self-control does not seem to rely on a limited resource — Carter, E. C.; Kofler, L. M.; Forster, D. E.; McCullough, M. E. (2015), meta-analysis B doi.org/10.1037/xge0000083 (retrieved 2026-11-09)
7. A Multilab Preregistered Replication of the Ego-Depletion Effect — Hagger, M. S.; Chatzisarantis, N. L. D.; et al. (2016), RCT A doi.org/10.1177/1745691616652873 (retrieved 2026-11-09)
8. A Multisite Preregistered Paradigmatic Test of the Ego-Depletion Effect — Vohs, K. D.; Schmeichel, B. J.; Lohmann, S.; et al. (2021), RCT A doi.org/10.1177/0956797621989733 (retrieved 2026-11-09)
9. Decision fatigue: A conceptual analysis — Pignatiello, G. A.; Martin, R. J.; Hickman, R. L. (2020), systematic review B doi.org/10.1177/1359105318763510 (retrieved 2026-11-09)
10. Mental fatigue: Costs and benefits — Boksem, M. A. S.; Tops, M. (2008), review B doi.org/10.1016/j.brainresrev.2008.07.001 (retrieved 2026-11-09)
11. A multifaceted investigation of the link between mental fatigue and task disengagement — Hopstaken, J. F.; van der Linden, D.; Bakker, A. B.; Kompier, M. A. J. (2015), RCT B doi.org/10.1111/psyp.12339 (retrieved 2026-11-09)
12. The Expected Value of Control: An Integrative Theory of Anterior Cingulate Cortex Function — Shenhav, A.; Botvinick, M. M.; Cohen, J. D. (2013), review B doi.org/10.1016/j.neuron.2013.07.007 (retrieved 2026-11-09)
13. A neuro-metabolic account of why daylong cognitive work alters the control of economic decisions — Wiehler, A.; Branzoli, F.; Adanyeguh, I.; Mochel, F.; Pessiglione, M. (2022), RCT C doi.org/10.1016/j.cub.2022.07.010 (retrieved 2026-11-09)
14. The neurocognitive consequences of sleep restriction: A meta-analytic review — Lowe, C. J.; Safati, A.; Hall, P. A. (2017), meta-analysis A doi.org/10.1016/j.neubiorev.2017.07.010 (retrieved 2026-11-09)
15. Executive Functions — Diamond, A. (2013), review A doi.org/10.1146/annurev-psych-113011-143750 (retrieved 2026-11-09)
16. Compensatory control in the regulation of human performance under stress and high workload: A cognitive-energetical framework — Hockey, G. R. J. (1997), review B doi.org/10.1016/S0301-0511(96)05223-4 (retrieved 2026-11-09)
17. Can There Ever Be Too Many Options? A Meta-Analytic Review of Choice Overload — Scheibehenne, B.; Greifeneder, R.; Todd, P. M. (2010), meta-analysis B doi.org/10.1086/651235 (retrieved 2026-11-09)
18. Choice overload: A conceptual review and meta-analysis — Chernev, A.; Böckenholt, U.; Goodman, J. (2015), meta-analysis B doi.org/10.1016/j.jcps.2014.08.002 (retrieved 2026-11-09)
19. Taking Stock of Self-Control: A Meta-Analysis of How Trait Self-Control Relates to a Wide Range of Behaviors — de Ridder, D. T. D.; Lensvelt-Mulders, G.; Finkenauer, C.; Stok, F. M.; Baumeister, R. F. (2012), meta-analysis A doi.org/10.1177/1088868311418749 (retrieved 2026-11-09)
20. A time to think: Circadian rhythms in human cognition — Schmidt, C.; Collette, F.; Cajochen, C.; Peigneux, P. (2007), review A doi.org/10.1080/02643290701754158 (retrieved 2026-11-09)
21. Implementation Intentions and Goal Achievement: A Meta-analysis of Effects and Processes — Gollwitzer, P. M.; Sheeran, P. (2006), meta-analysis A doi.org/10.1016/S0065-2601(06)38002-1 (retrieved 2026-11-09)