Ego Depletion: What Replication Studies Reveal About Willpower
Author: Ukrainian Psychological Hub · Published: October 10, 2026 · Editorial Policy
Ego depletion is the hypothesis that performing one demanding act of self-control temporarily makes later self-control more difficult. The idea gained extraordinary influence because it seemed to explain why resisting temptation, making difficult choices, or regulating emotions might leave someone less able to persist on an unrelated task. Its strongest form treats willpower as a shared, limited resource that is used up through effort and replenished through recovery.
The evidence has changed substantially. Large, preregistered replication projects generally report much smaller effects than early research, and two major projects obtained statistically nonsignificant primary results. A third detected a small effect. The best current interpretation is that the size, reliability, scope, and mechanism of laboratory ego depletion remain contested. Scientific evidence does not establish a universal willpower battery that each decision or act of restraint predictably drains. The central results come from Hagger and colleagues (2016), Dang and colleagues (2021), and Vohs and colleagues (2021).
This distinction preserves an important everyday reality: mental effort can feel demanding, and later performance can change. Experiencing fatigue, boredom, frustration, distraction, or altered motivation does not independently establish the limited-resource mechanism proposed by ego-depletion theory.
What Is Ego Depletion in Psychology?
In laboratory research, ego depletion refers to a possible decline in performance on a second self-control task following a first task designed to demand self-control, compared with performance after a less demanding first task. Ego in this historical research terminology refers to self-regulatory functioning, rather than vanity. Depletion refers to a proposed temporary reduction in capacity.
Two scientific claims need to be separated. The first is an empirical prediction: people assigned to a demanding initial task will perform worse on a later task. The second is a causal account: that performance difference occurs because the first task exhausts a finite common resource. Even a reproducible behavioral effect would not automatically prove the resource explanation.
In the influential experiments by Baumeister, Bratslavsky, Muraven, and Tice (1998), participants who resisted tempting food or engaged in other forms of self-regulation subsequently showed lower persistence or performance. Those findings helped establish a program of work on the strength model of self-control.
Ego depletion is a research construct and an unresolved scientific question. It is not a recognized mental health diagnosis, a personality classification, or a direct measure of moral character.
The Original Strength Model of Willpower
The strength model argues that suppressing impulses, changing emotional expressions, making difficult choices, and sustaining controlled attention draw on a common stock of regulatory strength. The first act uses some of that capacity, leaving less available for subsequent control. Rest or time away from demands is proposed to restore it.
This model was attractive because it connected different kinds of behavior under one mechanism. A person suppressing frustration at work might later find an unrelated decision harder. Someone resisting attractive food might subsequently give up on a frustrating puzzle. The point was generality: self-control demands in one domain could influence performance in another.
The early Hagger et al. (2010) meta-analysis aggregated 83 studies and estimated a medium-to-large effect, often reported as Cohen's d = 0.62. The literature also explored possible moderators such as incentives, task duration, and training. This synthesis was a major reason the model became prominent in textbooks, lectures, and popular accounts of willpower.
The explanatory simplicity created a high evidentiary burden. To demonstrate a common resource, research would have to show that a well-defined change in capacity is responsible for the measured performance differences. A useful metaphor does not in itself identify a physiological substance or a measurable psychological reserve.
How the two-task experiment works
A standard ego-depletion experiment randomly allocates people to two initial conditions. One group performs a task thought to demand more self-control, such as overriding an automatic response or inhibiting an emotional expression. A comparison group performs a less demanding version. Both groups then complete a second task intended to measure persistence, inhibition, or another regulatory process.
The theory predicts poorer performance on the second task in the first group. But the initial activities may differ in boredom, difficulty, frustration, novelty, enjoyment, perceived expectations, and willingness to continue. Second tasks may depend on practice, general cognitive ability, speed–accuracy trade-offs, and familiarity. Researchers must isolate these alternatives before concluding that a general-purpose resource has been depleted.
A null result can have several interpretations, including a genuinely absent or tiny effect, a manipulation that failed to produce the intended psychological state, or an insensitive outcome measure. A positive result also needs scrutiny: it may indicate a task-specific influence rather than a universal drop in self-control capacity.
Measuring an effect and explaining its mechanism
Statistical evidence that two groups differ is evidence about their measured performance under the tested circumstances. Causal explanation requires more. A shift in motivation, attention, response strategy, expectation, or subjective effort can produce a difference resembling depletion without any shared resource becoming empty.
For this reason, scientific reporting should avoid moving directly from a significant result to statements such as willpower was used up. The proposed mechanism needs independent measurement or competing predictions that distinguish it from plausible alternatives.
Why Earlier Studies Produced Stronger Effects
Early ego-depletion experiments were frequently relatively small, used many task combinations, and explored numerous behavioral outcomes. Such diversity can generate useful hypotheses, but the scientific record becomes vulnerable when stronger positive results are disproportionately published and when researchers have flexibility in analysis and exclusion choices.
Publication bias occurs when statistically significant findings enter the published record more readily than studies finding little evidence of an effect. Small-study effects arise when smaller experiments tend to report larger estimates. Analytic flexibility can also produce overly optimistic evidence when many decisions about measuring or reporting the outcome are made after results are visible.
Carter, Kofler, Forster, and McCullough (2015) conducted a series of meta-analytic tests addressing weaknesses in the earlier evidence. They concluded that conventional laboratory paradigms offered very little credible support for a general depletion effect. That conclusion conflicted sharply with the large effect estimated by Hagger and colleagues in 2010.
The conflict was productive because it encouraged new, more rigorous experiments. Meta-analyses depend on their inclusion criteria, analytic assumptions, and the quality of available studies. Bias corrections are also model-dependent. Comparing successive syntheses with large, independently organized experiments offers a stronger basis for judging the whole body of evidence.
The review by Friese and colleagues (2019) systematically examined skeptical and supportive arguments. It illustrates why neither a single critical analysis nor a single positive experiment can settle a broad scientific theory.
What the Preregistered Multilab Replications Found
Preregistration commits investigators to hypotheses, procedures, exclusion rules, and planned analyses before results are known. Multilab studies collect data across independent research groups, increasing statistical precision and reducing dependence on one team's procedures. These safeguards do not guarantee a perfect manipulation, but they make it harder to reinterpret outcomes after the fact.
Hagger et al. (2016): 23 laboratories, 2,141 participants
The registered replication report published in 2016 coordinated 23 laboratories using a standardized sequential-task procedure. The total sample included 2,141 participants. Its overall standardized estimate was d = 0.04 with a 95% confidence interval from −0.07 to 0.15.
The estimate was close to zero, and the confidence interval included zero. This result did not reproduce the much larger effects that earlier literature implied for a typical ego-depletion manipulation. The interval indicates that, under the model and design used, the observations were compatible with little or no difference and were inconsistent with a substantial effect of the originally expected size.
Supporters proposed that the chosen first task might not have generated enough regulatory demand or that certain designs are more effective than others. These possibilities are worth studying if they are precisely specified before new data collection. An explanation made after a negative result cannot substitute for a successful independent test.
Dang et al. (2021): 12 laboratories, 1,775 participants
A preregistered multilaboratory replication, published online in 2020 and in a 2021 journal issue, assessed a Stroop-based first task and a later antisaccade performance task across 12 laboratories. In 1,775 participants, it found a small but statistically significant ego-depletion effect, d = 0.10.
The researchers reported a somewhat larger effect, d = 0.16, after excluding people judged potentially to be responding randomly on the outcome task. Both estimates are substantially smaller than the classic meta-analytic figure. They also illustrate why reporting exclusion rules and distinguishing full-sample from restricted-sample analyses matters.
This study is a meaningful counterweight to claims that all high-quality preregistered studies found exactly zero. It supports the possibility of a small effect under some conditions. It does not establish a large or universal resource depletion mechanism.
Vohs et al. (2021): 36 laboratories, 3,531 participants
The multisite preregistered paradigmatic test involved 36 laboratories and 3,531 participants. Laboratories implemented one of two procedures designed to manipulate self-control demands before measuring subsequent performance. The confirmatory effect estimate was d = 0.06 and was not statistically significant.
The preregistered Bayesian comparison reported data approximately four times more likely under its specified null model than under a particular informed alternative model. This finding favors the null relative to that stated alternative; it is not the probability that all possible depletion effects are false.
An exploratory analysis using the full sample rather than the preregistered exclusions yielded a statistically significant but very small d = 0.08. That analysis generates questions for future work but carries a different evidentiary status from the confirmatory findings. The investigators also explored reported fatigue and proposed moderators, finding no consistent confirmatory support for several widely discussed personal characteristics.
What three large experiments show together
Their main reported estimates, d = 0.04, d = 0.10, and d = 0.06, are all very small compared with the classic estimate near d = 0.62. They cannot be simply pooled without accounting for differences among their tasks and designs. The statistically significant result from Dang and colleagues should be retained, and the nonsignificant primary results from the other two projects should be reported with equal clarity.
The replication record calls for a narrower conclusion than the popular willpower-battery explanation. It leaves open whether there are reproducible small effects in particular tasks and circumstances, while undermining the expectation that a large decrement reliably follows almost any exercise of self-control.
How to Understand Statistical Significance and Effect Size
A statistically significant result suggests that the observed data are relatively surprising under a particular statistical null hypothesis and analysis plan. It does not establish that an effect is large, useful, or caused by a specified mechanism. With large samples, an effect of very small practical importance may be estimated precisely enough to meet a conventional significance threshold.
Cohen's d represents an average difference between groups standardized by variability in the measured outcome. A d of 0.10 signals a small average difference, not a 10 percent loss of willpower. The number cannot be converted into minutes of self-control, units of mental fuel, or the probability that one particular person will make a bad decision.
A nonsignificant result does not logically prove an effect of exactly zero. Confidence intervals and preregistered tests help assess which magnitudes remain plausible. The central challenge is specifying how large an effect the theory predicts and whether a particular observed difference is meaningful beyond its significance label.
The Bayesian comparison in Vohs and colleagues' study evaluates relative predictions of explicitly stated statistical models. It cannot establish that no possible task combination, duration, or population will ever show a sequential-task effect.
What Meta-Analyses Add to the Replication Debate
Hagger et al. (2010) synthesized a research record that appeared strongly supportive. Carter et al. (2015) challenged that interpretation with analyses that addressed selective evidence. Dang (2018) provided an additional updated synthesis in a literature where task selection and heterogeneous results remained important.
The disagreement reflects differences in the evidence that each analysis accepts and the methods used to correct for possible bias. A meta-analysis cannot create missing null studies or ensure that every underlying experimental task measures the intended construct. It can, however, clarify why initial estimates changed and which questions require better research.
For current educational purposes, the large preregistered experiments merit substantial weight because they were specifically organized to assess a controversial effect under transparent conditions. Their primary findings should not disappear beneath an unqualified citation to the more dramatic 2010 estimate.
Is Willpower Actually a Limited Resource?
A resource account remains a scientific hypothesis. Human regulatory behavior can be difficult, variable, and sensitive to prior demands, but that observation does not identify a measurable universal store of willpower. People differ in incentives, familiarity with tasks, strategies, rest, beliefs, available support, and the significance of a goal.
The metaphor of a battery implies a measurable store that predictable actions discharge. Contemporary ego-depletion findings do not provide a validated instrument for estimating such a store or a consistent conversion between prior effort and remaining regulatory ability.
Proponents have refined the original model. In a 2024 review, Baumeister, André, Southwick, and Tice emphasized conservation of resources, more prolonged demands, and possible applications to sustained real-world effort. Their paper documents the ongoing theoretical defense of ego depletion. It should be read as a proponent interpretation alongside preregistered findings that question the magnitude and generality of the effect.
A refined model could gain stronger support by defining in advance which kinds of effort, durations, and outcomes should produce a measurable impairment, and by showing how its predictions differ from motivational or attentional explanations.
What about glucose as mental fuel?
Some early versions proposed glucose as the bodily energy substrate underlying limited self-control. This produced concrete predictions: performing a self-control task should measurably lower glucose; those levels should relate to subsequent performance; and ingesting glucose should restore impaired control.
A meta-analysis by Dang (2016) found no consistent support for those three predictions. Its results challenged the glucose-resource interpretation and raised concerns about publication bias in related research.
Adequate nutrition matters for general health, and low blood glucose can have medical implications in its own right. Neither fact validates the specific claim that sugary drinks replenish an ego-depletion reservoir. The research does not support prescribing sugar as a willpower remedy.
Competing Explanations: Motivation, Attention, and Effort
There are plausible ways an initial demanding task could influence later performance without exhausting a common resource. The proposed mechanisms are themselves subjects of investigation and should be evaluated according to their independent empirical support.
Shifting motivation
People may become less willing to continue an unattractive experimental activity after repeated exertion. An immediate reward, alternative task, or opportunity to rest may grow more appealing. If perceived benefits and costs change, a person may allocate less effort to the next task even while retaining the underlying capacity to perform it.
Inzlicht and Schmeichel (2012) developed an influential process account in which initial control demands produce shifts in motivation and attention. This framework predicts circumstances where incentives, importance, or task meaning might influence apparent depletion. Its explanatory proposals also require testing rather than automatic acceptance.
Shifting attention
An initial task can change which cues attract attention, how strongly task rules are maintained, or how people respond to rewards and distractions. Different self-control tasks measure different aspects of cognitive processing. An observed shift in one outcome may reveal changes specific to that measurement rather than a general deficit.
Fatigue, boredom, and perceived effort
The experience of being mentally tired after demanding activity is psychologically meaningful. People may also experience boredom, frustration, and reduced interest in continuing. These states can influence choices and subjective functioning. Yet the presence of fatigue alone neither proves depletion of regulatory capacity nor identifies the original finite-resource mechanism.
Exploratory findings about fatigue in the Vohs et al. study merit additional research, especially when they can be tested as prospective predictions. The overall preregistered result remained very small and nonsignificant.
Task strategy and measurement quality
A participant may change speed–accuracy priorities, learn a strategy, adapt to stimuli, or disengage from instructions. Two tasks that both look like inhibition challenges may rely on different combinations of skills. These methodological factors help explain why one experiment cannot stand in for every possible test.
Can Beliefs About Unlimited Willpower Prevent Ego Depletion?
The idea that people who regard willpower as unlimited are protected from depletion received attention after early research on willpower mindsets. It was especially appealing because it suggested that changing beliefs could reverse a supposedly biological limit.
A preregistered replication by Carruth, Ramos, and Miyake (2023) did not reproduce the influential claim that willpower beliefs reliably moderate the laboratory depletion effect. The Vohs et al. multisite test also did not provide clear support for willpower beliefs as a moderator of its primary effect.
Beliefs and expectations can influence how people interpret effort and engage with tasks. But this broad statement should not be confused with proof that a particular mindset prevents ego depletion. Presenting positive thinking as a guaranteed antidote exaggerates the evidence.
Ego Depletion, Mental Fatigue, and Self-Control Are Different Concepts
Ego depletion is a proposed sequential-task performance effect and, in one formulation, an explanation based on a limited shared resource. Mental fatigue describes an experience of tiredness or reduced readiness after sustained mental effort. Self-control refers to the wider set of processes through which people regulate attention, impulses, behavior, and competing goals.
Someone can feel mentally fatigued without showing the laboratory ego-depletion effect. Someone can perform poorly after an initial task for motivational or task-specific reasons. And a person can successfully exercise self-control by avoiding temptation in advance rather than by repeatedly inhibiting an impulse.
The wider concept is developed in our What Is Self-Control? Psychology, Mechanisms, and Examples article. The distinction prevents one contested experimental theory from defining all of self-regulation.
Burnout, chronic stress, reduced motivation, and clinical difficulties require their own frameworks. Persistent low mood, lack of pleasure, attention difficulties, or substantial impairment may merit professional assessment, but no diagnosis follows from a casual description of low willpower. Our specialized Anhedonia in Depression article examines one clinically important form of reduced pleasure and interest.
What the Evidence Means for Everyday Decisions
Laboratory ego-depletion results cannot establish that making many choices necessarily exhausts a daily quota of willpower, that late-day mistakes are always caused by resource loss, or that an individual's struggle to start a task indicates a depleted store. Everyday situations include additional factors: sleep, obligations, emotions, competing rewards, habits, time pressure, and social support.
It can be productive to ask what specifically changes after a demanding period. Has attention shifted? Is the next task poorly defined? Are distractions especially salient? Is a goal still meaningful? Has the person's opportunity to rest narrowed? Such questions identify practical possibilities without presupposing a contested biological or psychological battery.
Someone distracted by messages while studying may benefit from silencing notifications and specifying a concrete next step. Someone struggling with administrative tasks after work may benefit from a different schedule, smaller steps, or help managing competing demands. The adjustments follow the behavioral problem rather than a universal depletion diagnosis.
Evidence-Based Alternatives to Relying on Willpower Alone
A broad review by Duckworth, Gendler, and Gross (2016) distinguishes self-control strategies that operate earlier in a response process from direct attempts to resist a fully activated impulse. Selecting situations, modifying environments, directing attention, changing interpretations, and managing responses provide different intervention points.
Duckworth, Milkman, and Laibson (2018) further synthesize strategies such as planning, commitment devices, and changing choice conditions to reduce failures of self-control. These approaches differ in evidence strength and suitability across domains, and they should be matched to actual obstacles.
Practical options include reducing cues that trigger an unwanted behavior, arranging a desired action at a predictable time, preparing materials in advance, using a clear plan for interruptions, and arranging social or environmental support when appropriate. These strategies do not depend on knowing how much willpower remains in a hypothetical tank.
Rest and recovery remain worthwhile when a person is tired or overburdened. The rationale can be the immediate experience of fatigue, the importance of sleep and functioning, or the demands of a particular situation. No claim about a demonstrated self-control resource is necessary. For a more complete practical discussion, see How to Improve Self-Control: Evidence-Based Strategies.
How Better Research Could Resolve the Dispute
Future experiments should specify what form of ego depletion they intend to test. Researchers need reliable measures of initial demand, strong comparison conditions, valid second-task outcomes, preregistered analysis plans, sufficient statistical precision, and independent replication.
If the theory predicts a common limited resource, studies should identify its measurable consequences and distinguish those predictions from changes in motivation, attention, fatigue, and strategy. If proponents argue that only sustained exertion can create the effect, duration and intensity thresholds should be specified in advance and replicated.
Field research could examine real-world sequences of demands and choices, but would also need to address confounding factors such as workload, timing, sleep, incentives, and prior habits. A correlation between a hard afternoon and a subsequent lapse does not alone identify a depleted resource.
Clear reporting of positive, null, and ambiguous results is crucial. The goal is a cumulative understanding of when preceding demands change later performance, how large such changes are, and which mechanisms best explain them.
Scientific Evidence: Established, Limited, and Contested
Established: the original strength model made a clear prediction about sequential self-control tasks; early meta-analytic findings appeared large; subsequent bias-sensitive reviews and large preregistered collaborations substantially challenged the original expected effect size.
Supported in specified conditions: the Dang et al. preregistered collaboration detected a small statistically significant effect. This finding belongs in the evidence base alongside the small nonsignificant primary results of Hagger et al. and Vohs et al.
Contested: whether there is a generalizable sequential-task phenomenon with meaningful everyday effects, which manipulations reliably produce it, and whether a common finite resource explains any observed performance changes. Refined resource theories and motivational or attentional theories all require appropriately discriminating tests.
Unsupported as a universal prescription: that every choice drains willpower, that sugar restores an exhausted control resource, that a positive belief invariably prevents depletion, or that a person's symptoms or habits can be diagnosed through the ego-depletion metaphor.
Frequently Asked Questions
Is ego depletion real?
There is mixed evidence for a small sequential-task effect. Major preregistered studies found estimates close to zero, with some nonsignificant primary results and one small significant finding. The existence of a large, general depletion effect caused by a finite willpower resource is not established.
Has ego-depletion theory been disproven?
The classic strong formulation has suffered serious replication challenges. This does not establish that every possible sequential-task effect must be exactly zero. The narrower question of which procedures produce small reproducible changes remains open.
What did the major replication studies show?
Hagger et al. (2016) estimated d = 0.04 across 23 labs. Dang et al. (2021) found a small significant d = 0.10 across 12 labs. Vohs et al. (2021) obtained a nonsignificant confirmatory d = 0.06 across 36 labs.
Does making many decisions drain willpower?
The evidence does not establish a fixed number of decisions that exhaust a universal resource. Decision difficulty, rewards, time, fatigue, and context can matter without requiring the finite-resource explanation.
Is ego depletion the same as mental fatigue?
No. Mental fatigue is a subjective experience and research topic in its own right. Ego depletion is a proposed pattern of performance after an earlier self-control task and, in the original model, a particular causal theory.
Can glucose or sugar restore depleted willpower?
The 2016 glucose meta-analysis did not support the central physiological predictions of this account. A sugary drink is not an established intervention for replenishing willpower.
Can adopting an unlimited-willpower mindset eliminate depletion?
There is insufficient reproducible evidence for that specific moderator. The 2023 preregistered replication did not reproduce the influential original claim.
Why does resisting temptation sometimes feel harder later in the day?
Possible influences include accumulated demands, sleepiness, incentives, boredom, context, and shifting attention or motivation. A late-day lapse does not prove that a measurable store of control has been emptied.
Is procrastination or ADHD caused by ego depletion?
No such causal conclusion follows from this literature. Procrastination and ADHD-related difficulties require their own conceptual and, when relevant, clinical assessment. A sequential-task research hypothesis cannot diagnose an individual.
How can people improve self-control despite scientific uncertainty?
Identify the conditions that make desired behavior difficult, reduce predictable distractions, prepare clear actions, use workable routines, and seek appropriate rest or support. These approaches draw on broader self-regulation research rather than an assumed willpower reservoir.
Related Articles
Motivation, Habits, and Self-Control: Psychology of Behavior Change — a comprehensive overview of motivation, learned habits, self-control, and long-term behavior change.
What Is Self-Control? Psychology, Mechanisms, and Examples — the broader psychological construct underlying debates about ego depletion.
How to Improve Self-Control: Evidence-Based Strategies — strategies for everyday choices that do not depend on the finite-resource claim.
High vs Low Conscientiousness: Organization, Self-Control, and Flexibility — individual differences in organization and regulation, distinct from temporary experimental effects.
For the environmental and timing mechanisms behind proactive control, see Situational Self-Control: How Environment and Timing Shape Choices.
References
Baumeister, R. F., André, N., Southwick, D. A., & Tice, D. M. (2024). Self-control and limited willpower: Current status of ego depletion theory and research. Current Opinion in Psychology, 60, 101882. DOI: 10.1016/j.copsyc.2024.101882
Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74(5), 1252–1265. PubMed: 9599441
Carruth, N. P., Ramos, J. A., & Miyake, A. (2023). Does willpower mindset really moderate the ego-depletion effect? A preregistered replication of Job, Dweck, and Walton (2010). PLOS ONE, 18(6), e0287911. DOI: 10.1371/journal.pone.0287911
Carter, E. C., Kofler, L. M., Forster, D. E., & McCullough, M. E. (2015). A series of meta-analytic tests of the depletion effect: Self-control does not seem to rely on a limited resource. Journal of Experimental Psychology: General, 144(4), 796–815. DOI: 10.1037/xge0000083
Dang, J. (2016). Testing the role of glucose in self-control: A meta-analysis. Appetite, 107, 222–230. DOI: 10.1016/j.appet.2016.07.021
Dang, J. (2018). An updated meta-analysis of the ego depletion effect. Psychological Research, 82, 645–651. DOI: 10.1007/s00426-017-0862-x
Dang, J., et al. (2021). A multilab replication of the ego depletion effect. Social Psychological and Personality Science, 12(1), 14–24. DOI: 10.1177/1948550619887702
Duckworth, A. L., Gendler, T. S., & Gross, J. J. (2016). Situational strategies for self-control. Perspectives on Psychological Science, 11(1), 35–55. DOI: 10.1177/1745691615623247
Duckworth, A. L., Milkman, K. L., & Laibson, D. (2018). Beyond willpower: Strategies for reducing failures of self-control. Psychological Science in the Public Interest, 19(3), 102–129. DOI: 10.1177/1529100618821893
Friese, M., Loschelder, D. D., Gieseler, K., Frankenbach, J., & Inzlicht, M. (2019). Is ego depletion real? An analysis of arguments. Personality and Social Psychology Review, 23(2), 107–131. DOI: 10.1177/1088868318762183
Hagger, M. S., et al. (2016). A multilab preregistered replication of the ego-depletion effect. Perspectives on Psychological Science, 11(4), 546–573. DOI: 10.1177/1745691616652873
Hagger, M. S., Wood, C., Stiff, C., & Chatzisarantis, N. L. D. (2010). Ego depletion and the strength model of self-control: A meta-analysis. Psychological Bulletin, 136(4), 495–525. DOI: 10.1037/a0019486
Inzlicht, M., & Schmeichel, B. J. (2012). What is ego depletion? Toward a mechanistic revision of the resource model of self-control. Perspectives on Psychological Science, 7(5), 450–463. DOI: 10.1177/1745691612454134
Vohs, K. D., et al. (2021). A multisite preregistered paradigmatic test of the ego-depletion effect. Psychological Science, 32(10), 1566–1581. DOI: 10.1177/0956797621989733