Duration neglect
How long an experience lasts barely affects how we remember it.
What it means
Duration neglect is the finding that, when people evaluate past experiences in retrospect, the length of the experience has surprisingly little influence on their overall judgment of how good or bad it was. The mechanism is that the 'remembering self' summarizes an episode mainly by a few salient moments — its emotional peak and its end — rather than by integrating pleasure or pain over its full span, which is why duration drops out of the assessment. It is the close companion of the peak–end rule, and together they explain striking anomalies such as people preferring a longer painful episode with a milder ending to a shorter one that ended at its worst, a violation of the principle that less total suffering should always be preferred. A boundary condition is that duration is not ignored when people are explicitly cued to attend to it or when the decision concerns the experience prospectively rather than in memory, so it is a feature of retrospective evaluation in particular. It matters for the design of services, medical procedures, and customer experiences, and for the deeper puzzle of how to weigh remembered satisfaction against the actual quantity of moment-to-moment experience.
Why duration drops out
The remembering self does not replay an episode second by second. It stores a compressed summary built from a few emotionally charged snapshots, typically the most intense moment and the last one, and reports the average of those as its verdict. Length is a quantitative attribute that would require deliberately integrating feeling over time, work the mind rarely does when a quick global impression is all that is called for. Fredrickson and Kahneman's original film-clip studies showed that whatever effect duration had shrank once real-time affect was accounted for, and shrank again when the rating was made from memory rather than on the spot. So duration is not erased while the experience is happening; it registers moment to moment but fails to survive the translation into a remembered judgment.
What the evidence shows
The two anchor demonstrations are experimental. In the cold-pressor study, adding a stretch of merely less-cold water to the end of an already painful trial made people prefer to repeat the longer, objectively worse ordeal. In a randomized trial of 682 colonoscopy patients, leaving the scope still for an extra gentle interval lowered remembered pain. A 2022 meta-analysis confirmed a large peak-end correlation and a near-zero duration effect, but with a sharp caveat: the simple average of moment-to-moment feeling predicted overall judgments just as well as the peak-end composite. On that reading, the peak-end composite is not the uniquely privileged summary it is often taken to be.
When duration does matter
Duration neglect is a property of quick retrospective evaluation, not a general law. Ariely and Loewenstein showed that when an experience has a clear rising or falling trend, or when people are asked to attend to how long it lasted, length re-enters the judgment. It also surfaces in study design: neglect is cleaner in between-subjects comparisons, where each person rates only one episode, than when the same person weighs a long and a short experience side by side and the difference becomes hard to ignore. Prospective choices, deciding whether to endure something in the future, weight duration more heavily than memories of the past do. Neglect is strongest exactly when a single, holistic, backward-looking impression is all that is asked for.
Using it, and the ethics
For anyone designing an experience, the lever is the shape of the timeline, not its length: put the hardest part early, build toward a genuine high point, and end on relief or a small kindness. Clinics, flights, hold queues, and onboarding flows all trade on this. But the colonoscopy result carries an uncomfortable twist, since the gentler memory was bought by deliberately extending the procedure, adding real if milder discomfort in order to improve the recollection of it. That raises a live ethical question, debated in the medical literature: improving a remembered experience can mean worsening the experienced one, and a patient's later willingness to return may rest on a memory engineered to be rosier than the event.
Examples
Two vacations rated nearly the same despite one being twice as long, because the highlights matched.
In a classic study, patients rated a longer, more drawn-out colonoscopy as less unpleasant than a shorter one, because the longer version simply ended with a gentler final few minutes.
A two-hour meeting that closes on a warm note is remembered more fondly than a forty-minute one that ends in a row, and nobody's rating reflects the eighty minutes saved.
A caller held for twenty minutes who then reaches a warm, competent agent rates the call better than one held two minutes before a curt brush-off; the wait barely enters the rating.
A grueling ninety-minute spin class that closes with an easy cooldown is remembered as more bearable than a brisk thirty-minute one ending at a full sprint, so members re-book the longer session.
First described in Kahneman and colleagues (1993).
Key references
- Alaybek, B., Dalal, R. S., Fyffe, S., Aitken, J. A., Zhou, Y., Qu, X., Roman, A., & Baines, J. I. (2022). All's well that ends (and peaks) well? A meta-analysis of the peak-end rule and duration neglect. Organizational Behavior and Human Decision Processes, 170, 104149. doi.org/10.1016/j.obhdp.2022.104149
- Redelmeier, D. A., Katz, J., & Kahneman, D. (2003). Memories of colonoscopy: A randomized trial. Pain, 104(1-2), 187-194. pubmed.ncbi.nlm.nih.gov/12855328/
- Ariely, D., & Loewenstein, G. (2000). When does duration matter in judgment and decision making? Journal of Experimental Psychology: General, 129(4), 508-523. doi.org/10.1037/0096-3445.129.4.508
- 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. doi.org/10.1111/j.1467-9280.1993.tb00589.x
- Fredrickson, B. L., & Kahneman, D. (1993). Duration neglect in retrospective evaluations of affective episodes. Journal of Personality and Social Psychology, 65(1), 45-55. pubmed.ncbi.nlm.nih.gov/8355141/