Trap/Heuristic/No. 0732
Peak-End Rule
The peak-end rule is a pattern in which people judge some past experiences disproportionately by their most intense moment and their ending. Studied by Daniel Kahneman and colleagues, it is linked to duration neglect, but other moments and the experience’s length can also matter.
Also called Peak-End Effect
- Evidence
- Well established
- Read
- 6 min
- Links
- 7 connections
01You've seen this when…
- in life
Your weekend away includes delayed trains and an uncomfortable hotel room. A spectacular sunset and a leisurely final breakfast dominate the story you tell afterward.
- at work
A daylong workshop gives you several useful techniques. But one humiliating role-play and an abrupt, critical closing leave you remembering the whole day as awful.
- out in the world
You spend much of a city festival standing in lines. One extraordinary performance and a dazzling finale dominate the glowing review you post.
02The idea
When you look back on an experience, a few moments can shape your judgment of the whole. A memorable high or low, followed by how things finished, can carry much more weight than the long stretch in between.
The peak-end rule describes this pattern in retrospective evaluations: judgments of some experiences disproportionately reflect their most intense moment and their ending. It is especially well documented in certain bounded experiences of discomfort. Researchers sometimes model the judgment as an average of peak and ending intensity. The model is a useful description of particular findings, with limits on its use as a formula for other memories.
This matters because your remembered verdict helps decide what you repeat. You might choose an experience that felt worse overall because it left a better memory.
The related idea of duration neglect concerns how little an experience’s length sometimes affects its remembered unpleasantness. The two often appear together, but they’re distinct. Peak-end weighting describes which moments receive extra weight; duration neglect describes information that gets too little.
03Why it happens
There isn’t one settled explanation for every peak-end finding. Several features of memory and judgment can contribute:
- Striking moments and endings offer ready evidence. Intense pleasure or pain stands out against ordinary moments. When you need one overall verdict, that standout can become your main evidence. This overlaps with salience bias and the availability heuristic. The ending is a convenient stopping point. It is recent when an immediate judgment is requested, and it helps frame how the episode turned out. That can give it influence beyond how long it lasted. Recency in memory is relevant to the ending’s influence; the extra weight given to peaks requires a separate explanation.
- A single rating demands compression. Turning ninety minutes of changing feelings into one score requires a shortcut. People can arrive at that score without calculating an average across every minute.
These explanations fit a broader fact: memory is reconstructed as an account assembled from stored details. But peak-end weighting isn’t simply a memory defect. Someone can remember the uneventful middle and still give it little weight in the final judgment.
04A worked example
In a randomized colonoscopy trial published in 2003, researchers changed how the procedure ended. For one group, the colonoscope remained in place briefly after the main examination, without being moved. This added an interval that was uncomfortable but less painful than earlier portions of the procedure.
What it looks like Adding discomfort should make a painful medical experience worse to remember. The patients receive the full examination plus extra time in the procedure.
What’s actually going on Patients given the extended, milder ending reported less unpleasant retrospective evaluations on average. The added interval changed the finish without removing the discomfort already experienced. A better ending improved the remembered verdict.
What would have helped Assess patient welfare by keeping both kinds of information: discomfort reported during the procedure and the patient’s later evaluation. For care teams, the practical lesson is to provide better relief and reassurance and improve how the procedure ends where medically appropriate. Care teams should avoid adding unnecessary discomfort merely to produce a favorable memory. A less unpleasant recollection matters. Pain already endured remains part of the patient’s experience.
05How to spot it
These are prompts to check your judgment. Your judgment may still be sound.
06What to do instead
- Record a few ordinary moments. During a recurring experience, make brief ratings at several points, including both ordinary and remarkable moments. A simple note is enough, and the tracking can be occasional.
- Separate the questions. Rate the overall experience separately from its worst or best moment and its ending. Then compare the answers to build a fuller account than one global score provides.
- Include time when time matters. Two hours of moderate discomfort may matter more to your decision than a brief sharp pain. Write down both intensity and duration.
- Compare your later verdict with your earlier notes. A decision journal can reveal whether your preference changed after the ending. For repeat activities, record your expectations and a few observations during the activity, then add your next-day judgment.
- Improve the experience before polishing its finish. A considerate ending can improve the remembered verdict. Fix long waits and confusing instructions, and relieve sustained discomfort.
Recording observations is a form of cognitive offloading: you give your later self evidence that doesn’t depend entirely on whichever moments are easiest to recall.
07When it isn’t a distorted judgment
A strong reaction to a peak or ending can be a sound judgment. One dangerous incident can reasonably outweigh an otherwise pleasant trip. A final conversation can reveal whether a relationship is repairable. Those moments carry weight because of the important information they contain, beyond the feelings that make them memorable.
The pattern also depends on the experience and the question asked. A short, clearly bounded procedure is different from a month of work or a day containing many unrelated episodes. Research on multiepisode experiences finds limits to peak-end predictions. Average feelings and other parts of the experience can matter substantially.
Pleasant and unpleasant moments also needn’t collapse into one obvious peak. Don’t assume the happiest moment wins, or that the worst moment always dominates.
Finally, memory is itself part of well-being. A comforting conclusion can improve someone’s later experience. The mistake is treating remembered satisfaction as the only thing that counts.
08Roots
In a 1993 experiment, Daniel Kahneman and colleagues asked volunteers to hold a hand in cold water, then choose which ordeal to repeat. One trial lasted sixty seconds. Another lasted ninety: the same initial exposure, followed by thirty seconds during which the water became slightly warmer but remained uncomfortable.
Many volunteers preferred to repeat the longer trial. It contained an additional stretch of discomfort, yet its milder finish made it preferable in retrospect. Work published that year by Barbara Fredrickson and Kahneman also examined retrospective evaluations of emotional film clips. These studies helped establish peak-and-end weighting alongside duration neglect. Research then moved into painful medical procedures, where remembered experience could influence willingness to return for care.
One speculative adaptive account is that intense moments identify what deserves attention, while endings help signal whether an encounter finished safely or should be repeated. That could make peak-end weighting an economical summary strategy. The experiments establish a conditional pattern in judgment. The idea that natural selection produced it for these reasons remains speculative.
09How solid is this?
Controlled experiments and a randomized medical trial support peak-end weighting in some bounded experiences. It is not a universal law: multiepisode experiences show limits, and duration, average feelings, and other moments can still matter.
10Connections
- Often confused withDuration Neglect
- Countered by Decision Journal
- Part of Heuristic, Salience Bias
- See alsoReconstructive Memory, Availability Heuristic, Recency Effect in Memory
11Origin and sources
Daniel Kahneman, Barbara Fredrickson, Charles Schreiber, and Donald Redelmeier described the peak-and-end pattern in 1993; related work by Fredrickson and Kahneman examined duration neglect that year.
- [1]Kahneman, D., Fredrickson, B. L., Schreiber, C. A., & Redelmeier, D. A. (1993). When More Pain Is Preferred to Less: Adding a Better End. Psychological Science, 4(6), 401–405.
- [2]Fredrickson, B. L., & Kahneman, D. (1993). Duration neglect in retrospective evaluations of affective episodes. Journal of Personality and Social Psychology, 65(1), 45–55.
- [3]Redelmeier, D. A., Katz, J., & Kahneman, D. (2003). Memories of colonoscopy: a randomized trial. Pain, 104(1–2), 187–194.
- [4]Miron-Shatz, T. (2009). Evaluating multiepisode events: Boundary conditions for the peak-end rule. Emotion, 9(2), 206–213.
Suggest an edit· Updated 2026-10-02