Decision Fatigue
Learn it → Apply it → Save as Skill
Decision fatigue is the tendency for the quality of your decisions to slip the more decisions you have already made, as the mental effort required for deliberate choice gets used up across a session. Psychologists Kathleen Vohs, Roy Baumeister, and colleagues demonstrated the pattern in a series of studies published in 2008 in the Journal of Personality and Social Psychology, "Making Choices Impairs Subsequent Self-Control," showing that people who had just made a string of choices persisted for less time afterward on an unrelated self-control task than people who had reviewed the same information without having to choose. Field research since then has found a similar drift outside the lab, in courtrooms, medical clinics, and car dealerships, though the exact mechanism behind it is more contested than the popular version of the idea admits.
Dr. Wallace's Favorite Mental Models turns one well-chosen way of thinking at a time into something you can actually use. Each entry gives you a fast overview, an interactive walkthrough, varied real and practical examples, and an AI-ready Markdown skill you can keep.
Instead of only reading about a mental model, you can build your own AI skills library over time: a personal set of portable reasoning tools that help ChatGPT, Claude, and other AI assistants apply the thinking you value.
Decision fatigue shows up anywhere someone has to make choice after choice with little or no break: a parole board hearing case after case, a physician seeing patient after patient, a hiring panel interviewing candidate after candidate, a shopper clicking through option after option in a product configurator. The quality drift is usually not dramatic on any single decision. It shows up as a drift toward whatever requires the least new deliberation: the default option, the easier prescription, the safer no, the shorter interview. Decisions made early in a session, when mental resources are fresh, tend to get more scrutiny than decisions made later in the same session, even when the stakes are identical.
Important: both figures are the published result of a real study of Israeli parole hearings, not an invented illustration. Favorable rulings then jump back up to roughly 65% right after each scheduled food and rest break, tracking the break schedule far more tightly than it tracks the facts of each case.
Use the two axes below, whether the pattern involves one decision or many decisions in a row, and whether it is mainly about the number of options or about mental resources draining over time, to see how decision fatigue relates to three commonly confused neighbors. Select a dot or a chip to read how each one compares.
Each stage builds on the one before it. Use the numbered buttons or Next to move through the walkthrough.
Open an example to see decision fatigue in a specific context. The first five are documented research, including one important replication caveat; the rest are common hypothetical scenarios used to teach the method, clearly labeled as such.
Shai Danziger, Jonathan Levav, and Liora Avnaim-Pesso studied more than 1,100 Israeli parole board rulings and found the percentage of favorable decisions started near 65% right after each scheduled break, declined gradually toward nearly zero by the end of that session, and jumped back to about 65% once the next break began. The case facts did not change on that schedule; the board's position in its own session did.
Jeffrey Linder and colleagues examined more than 21,000 primary care visits for acute respiratory infections across 23 practices over 17 months, published in JAMA Internal Medicine. Inappropriate antibiotic prescribing for infections that typically do not need them was about 5 percentage points higher at the end of a four-hour clinic session than at the beginning of the same session.
Kathleen Vohs, Roy Baumeister, and colleagues published "Making Choices Impairs Subsequent Self-Control" in the Journal of Personality and Social Psychology. Across several studies, people who had just made a series of choices, such as selecting items from a product catalog, persisted for less time on a later, unrelated self-control task than people who had reviewed the same options without having to choose.
Jonathan Levav, Mark Heitmann, Andreas Herrmann, and Sheena Iyengar ran field experiments with real car and suit customers, published in the Journal of Political Economy. The order in which attributes were presented changed how often buyers simply accepted the seller's suggested default for that attribute, with later-presented attributes seeing higher default acceptance as the customization sequence wore on.
Martin Hagger and a large international team ran a preregistered, multi-lab replication of the classic ego-depletion effect, the proposed mechanism behind decision fatigue, published in Perspectives on Psychological Science. Across 23 labs, the overall effect was not statistically significant. The behavioral drift seen in the field studies above has kept showing up, but the specific "limited willpower resource" explanation for why is genuinely contested, not settled science.
A hiring panel schedules nine back-to-back candidate interviews in one day. By the afternoon, interviewers are asking shorter follow-up questions and writing thinner notes, independent of how strong any individual candidate's answers actually were.
Near the end of a long shift, a physician facing an ambiguous case leans toward whichever option requires the least additional deliberation, ordering a familiar test or writing a familiar prescription, rather than working back through the full differential the case actually calls for.
A procurement lead pushes back hard on the first five contract clauses reviewed in the morning, then barely scrutinizes the last five clauses of the afternoon, even though two of the later clauses carry more financial risk than anything reviewed earlier.
This is an illustrative, invented scenario designed to teach the method. The figures are the director's own working notes, not documented research statistics.
Expand each item for the research and reasoning behind the model.
Journalist John Tierney popularized the phrase in a 2011 New York Times Magazine article, "Do You Suffer From Decision Fatigue?" The piece drew heavily on Roy Baumeister's decades of research into self-control and willpower, including Baumeister, Bratslavsky, Muraven, and Tice's 1998 study "Ego Depletion: Is the Active Self a Limited Resource?" which first proposed that acts of self-control, including making choices, draw on one shared, limited mental resource.
No. Liora Weinshall-Margel and John Shapard published a 2011 reply arguing that part of the drop in favorable rulings could reflect how cases were scheduled, including how attorneys and the parole board itself ordered cases, rather than fatigue alone. A later 2016 simulation paper by Andreas Glockner argued the published effect size may be larger than the underlying pattern actually warrants. Danziger and Levav published their own reply defending the original interpretation. The direction of the effect has been widely cited since, but the exact size and cause are still debated among researchers.
Not reliably. Martin Hagger led a large, preregistered replication across 23 independent labs, published in 2016 in Perspectives on Psychological Science, and found no statistically significant overall ego-depletion effect. This is one of the more consequential replication failures in social psychology, and it means the specific mechanism of a literal limited willpower resource is considerably less certain than the field-level behavioral pattern itself.
No. Field data from real Israeli parole hearings, real United States primary care visits for respiratory infections, and real car and suit customization purchases all show decisions drifting in the same direction as a session continues, independent of any laboratory self-control task.
Choice Overload is about a single decision point: too many options in front of you at once makes that one choice harder, slower, or less satisfying. Decision fatigue needs no complicated menu at all. A parole ruling and an antibiotic prescription are both effectively binary choices, yet both still drifted measurably worse purely as a function of how many similar decisions had already been made earlier in the same session.
The specific "ego depletion" mechanism failed to replicate in a 23-lab study. The behavioral pattern of worse later decisions is well documented; the brain-level explanation for why is still genuinely disputed.
Choice Overload is about too many options at one decision point. Decision fatigue shows up even in simple, binary choices made repeatedly, purely from how many came before in the same session.
Published critiques argue case-scheduling patterns, not fatigue alone, may explain part of the drop in favorable rulings, and the original authors have disputed that reply in turn.
The pattern is a statistical tendency visible across many decisions, not a guarantee about the quality of any one specific case.
The documented fixes involve restructuring the decision queue itself, its order and its breaks, not simply trying harder to concentrate through it.
The antibiotic-prescribing and product-customization research both found the pattern in routine, low-drama choices, not only in life-altering decisions.
If decision quality predictably erodes across a session, the highest-stakes item belongs earliest in the queue, not scheduled for the most depleted stretch of the day.
it describes a pattern across many decisions in a session, not evidence about whether any single ruling, prescription, or approval was actually wrong.
the proposed "limited willpower resource" mechanism behind it failed a large, preregistered, 23-lab replication, so describe the behavioral pattern, not a confirmed brain mechanism.
the model is about accumulation across a sequence; a well-rested first choice of the day has nothing yet to accumulate.
the fix is restructuring when and in what order decisions happen, not abandoning careful judgment on the decisions that matter.
This browser tool structures your thinking. It does not send your information anywhere or perform AI analysis.
Save as Skill converts the model into portable Markdown instructions that teach an AI how to spot a fatigue-prone decision queue, ask the right follow-up questions, and avoid the common mistakes above.
Save the skills you find useful and organize them into your own reusable library. The Markdown is intentionally model-neutral, so it can be used wherever an AI system accepts uploaded files, project knowledge, custom instructions, or reusable text context. Platform capabilities differ, so this is a portable skill file rather than a claim of one-click installation.
Describe a specific decision queue or sequence you are working through. This builds a separate, personalized prompt that asks an AI to apply decision fatigue directly to your situation, distinct from the prompt builder above.
No. It means the position of a decision in a sequence of similar decisions is worth managing on purpose. Reordering the queue so high-stakes decisions land earlier, or adding a real break beforehand, addresses the pattern directly, rather than avoiding the afternoon altogether.
The parole judges in the original study and the physicians in the prescribing study were both experienced professionals operating well within their expertise. The drift tracked their position in the session, not a lack of skill or seniority, which is why the fix is structural, changing the queue, rather than a matter of training harder.
Before building a long queue of similar decisions, place the highest-stakes item as early as realistically possible, and schedule a genuine break before any decision you cannot afford to get wrong. Where a decision is low-stakes and repetitive, consider turning it into a standing default or checklist so it stops competing for the same mental budget as the decisions that actually need judgment.