Skip to content
Search resources Open menu

Memory in the Courtroom

The Peak-End Rule and Duration Neglect

Published

Snap Shot

  • The question: When a witness describes how bad an extended experience was overall, or how long it lasted, how closely does that retrospective account track what was experienced moment by moment?
  • What the research shows: Retrospective evaluations of an episode are strongly associated with its most intense moment and its ending (the peak-end rule), and the length of the episode adds little to the overall evaluation (duration neglect); a 2022 meta-analysis of 174 effect sizes supported both. Separate research, resting mainly on one 1987 set of laboratory studies (Loftus et al.), found that participants who watched a short videotaped robbery overestimated how long it lasted, more so for a more stressful version.
  • What it does not show: The research concerns how an experience is summarized and evaluated, not whether it occurred or whether its concrete details are remembered accurately. When a witness dwells on the worst moment and the ending, or cannot say precisely how long something went on, that is not evidence of exaggeration; nor does a restrained summary show that an experience was mild.
  • The law: Expert testimony is governed by Rule 702 and Houser, and in Arizona by Rule 702, Chapple, and Nordstrom; experts may describe how retrospective evaluations and duration estimates work but may not opine on a particular witness’s credibility (Lindsey). Duration estimates bear on the opportunity-to-view factor in identification cases (Neil v. Biggers).
  • Issues for each side: Prosecutors can explain why a sincere witness may dwell on the worst moment and the ending and estimate duration imprecisely, without presenting an overall characterization or a stated duration as a measurement; defense counsel can test durations and global characterizations against independent evidence without arguing that an imprecise estimate means an event did not occur.

How a person remembers an extended experience is not a running total of its moments. Two related findings describe what memory tends to keep instead. Under the peak-end rule, a person’s later evaluation of an episode, such as how painful or distressing it was, is closely associated with its most intense moment and with how it ended. Under duration neglect, how long the episode lasted contributes surprisingly little to that later evaluation.

These findings reach court whenever a witness summarizes a prolonged or repeated experience, such as an assault, a confinement, a beating, a course of mistreatment, or a painful medical event, and whenever a witness is asked how long something lasted. A global characterization (“it was the worst night of my life”) and an estimate (“it went on for twenty minutes”) are both retrospective judgments, and both are shaped by how memory summarizes experience. A separate body of research addresses how accurately people estimate the duration of events they witnessed, and it bears directly on testimony about how long a confrontation lasted or how long a witness saw a face.

This paper describes the experiments that established the peak-end rule and duration neglect, the 2022 meta-analysis and later field research, the research on estimating the duration of witnessed events, the limits of this work, and the law that governs testimony about it. Related questions about numerical estimates appear in the companion paper, Anchoring in Witness Estimation.

The Original Experiments

The findings emerged from studies comparing what people experienced moment by moment with what they later reported. In one experiment, participants held a hand in painfully cold water for 60 seconds on one trial and, on another, for the same 60 seconds followed by an additional 30 seconds during which the water was warmed slightly, so that the discomfort eased but did not end. The longer trial contained all the pain of the shorter one and more, yet most participants chose to repeat the longer trial, a preference the authors attributed to its gentler ending (Kahneman et al., 1993).

In a related study, participants watched pleasant and unpleasant film clips of varying lengths, rated their feelings continuously, and later gave overall evaluations. The overall evaluations were associated with the most intense and the final moments of each clip, and the length of the clip had little effect on them (Fredrickson & Kahneman, 1993). The authors called this duration neglect.

The same pattern appeared in medical settings. Patients undergoing colonoscopy (154 patients) or lithotripsy (133 patients) reported their pain intensity in real time during the procedure and afterward rated the total pain of the experience. Their overall ratings were strongly associated with the worst pain and with the pain in the last three minutes of the procedure, and lengthy procedures were not remembered as especially aversive, even though procedures varied widely in length (Redelmeier & Kahneman, 1996). In a randomized trial of 682 colonoscopy patients, those for whom the procedure was extended by a short, less uncomfortable final period rated the experience as modestly less unpleasant than patients whose procedure ended without that period (Redelmeier et al., 2003). In each study, people did not ignore what had happened; their summaries gave disproportionate weight to particular moments.

What the Meta-Analysis and Later Research Show

A 2022 meta-analysis gathered 174 effect sizes testing the peak-end rule across laboratory and field studies. The association between the peak-and-end combination and retrospective evaluations was large (r = .58) and held across the conditions the authors examined. It was stronger than the associations for the trend of the experience, its variability, its first moment, or its least intense moment, and it was comparable to the association for the average of all moments. Duration had essentially no effect on evaluations, which supports duration neglect (Alaybek et al., 2022).

Two features of these results matter for testimony. First, the peak and end predicted overall evaluations about as well as the average of the whole experience. The finding is not that retrospective evaluations are unrelated to the experience as a whole; it is that the most intense and final moments carry substantial weight in the summary. Second, the effect is not uniform. In an experience-sampling study of nearly 1,900 people who rated their well-being repeatedly in daily life, retrospective judgments were disproportionately influenced by peak and end experiences, but the size of the effect depended on how the retrospective question was worded and on whether peaks and ends were defined as moments, days, or weeks (Scharbert et al., 2025). Some studies find little peak or end effect at all; in a study of remembered pleasure from meals, duration neglect appeared but peak and end effects were minimal (Rode et al., 2007).

Duration Neglect Is Not the Same as Misjudging Duration

Duration neglect describes how little the length of an episode affects how good or bad it is later judged to have been. It does not describe how accurately a person can say how long the episode lasted, which is a different question. Courts are often interested in both: how bad an experience was and how long it went on.

Research on witnesses’ estimates of duration finds substantial error. In three experiments involving 469 participants who watched a short videotape of a bank robbery, participants consistently overestimated how long it lasted; the more stressful version of the event produced larger overestimates in the third experiment, and women overestimated more than men in two of the experiments. The accuracy of participants’ time estimates was unrelated to how much they recalled or to the accuracy of their memory for the event (Loftus et al., 1987). Reviewing research on identification, Lane and Houston (2021, pp. 30–31) note that longer exposure to a face generally helps later recognition but that no study establishes how long is long enough, and that people misjudge how long they had to view someone.

Two points follow. A witness’s estimate of how long a brief, frightening event lasted may be longer than the event actually was, and the error is not a sign that the rest of the witness’s memory is poor, because duration accuracy and event memory were unrelated. Conversely, an accurate memory of what happened does not guarantee an accurate estimate of how long it took.

Relevance for Testimony

The research bears on several recurring situations. A witness’s overall characterization of an assault or a confinement is likely to reflect its most intense moments and how it ended, rather than a weighted sum of everything that occurred. Two witnesses to the same event may characterize it differently if it ended differently for them, for example if one left before it was over. A witness’s statement about duration, whether of an assault, a struggle, a period of restraint, or the time spent looking at an assailant’s face, is an estimate that may be considerably off and, for brief stressful events, has tended in the available research to run long.

The research also has implications for questioning. Because a global summary compresses the experience, open questions that ask the witness to describe what happened from beginning to end, and what happened next, generally yield more information about the course of an event than questions asking how bad it was overall. Because duration estimates can be substantially in error, independent evidence such as phone records, video, timestamps, or the number and sequence of actions described is generally a better guide to how long something lasted than the estimate itself.

Limits of the Evidence

The foundational studies used cold water, film clips, and medical procedures, experiences that are unpleasant but not threatening in the way a crime is, and the participants knew they were being studied and that the experience would end. The meta-analysis drew largely on laboratory, consumer, and well-being research, not on forensic settings, and the effect varied with how evaluations were requested. Research on duration estimates for witnessed crimes is limited, much of it is decades old, and it typically involves brief filmed events rather than events lasting many minutes or hours. Little research addresses how these effects operate in memory for repeated events, for events experienced while intoxicated or injured, or for events that a person has recounted many times. No study can establish how a particular witness weighted the moments of a particular experience or how far a particular duration estimate departs from the truth.

Questions to Ask About This Research

When Redelmeier and Kahneman (1996) or Kahneman et al. (1993) is cited to explain why a complainant’s overall description emphasizes the worst moment and the ending, counsel should ask who was studied and what was measured: patients rating medical discomfort and participants rating cold water gave numerical evaluations of experiences they knew would end, which is different from describing a threatening event. When Alaybek et al. (2022) is cited, counsel should ask what exactly was counted, since the peak-end association was about as strong as the association for the average of the whole experience, which limits arguments that a summary is unrelated to what happened overall. When Loftus et al. (1987) is offered to show that a witness’s duration estimate is inflated, counsel should ask whether the conclusion follows for this event: the study used a brief videotaped robbery, and the direction and size of error for a longer event, or one the witness lived through, may differ. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.

The Peak-End Rule and the Law

Federal and Military Standards

Expert testimony about retrospective evaluation and duration estimates is governed by Federal Rule of Evidence 702, as amended December 1, 2023, under which the proponent must show the court that it is more likely than not that the rule’s requirements are met, including that the opinion reflects a reliable application of reliable methods to the case. Daubert v. Merrell Dow Pharmaceuticals, Inc. (1993) made the judge a gatekeeper, and Kumho Tire Co. v. Carmichael (1999) extended that obligation to all expert testimony, including testimony based on technical or other specialized knowledge. In courts-martial, Military Rule of Evidence 702 tracks the federal rule; under Military Rule of Evidence 1102, the Joint Service Committee’s 2025 compilation reflects the 2023 amendment as part of the military rule (Joint Service Committee on Military Justice, 2025), and practitioners should confirm current text. Military judges also apply the six factors of United States v. Houser (1993). In identification cases, the witness’s opportunity to view the perpetrator is one of the reliability factors under Neil v. Biggers (1972) and Manson v. Brathwaite (1977), and that opportunity is often established by the witness’s own estimate of how long the viewing lasted.

Arizona

Arizona Rule of Evidence 702 adopted the federal formulation and the Daubert standard effective January 1, 2012, and was amended effective January 1, 2024, to conform to the 2023 federal amendment (Ariz. Sup. Ct. Order R-23-0004). State v. Chapple (1983) is Arizona’s leading case on expert testimony about eyewitness memory; excluding the expert was an abuse of discretion on its facts, and admission otherwise remains within the trial court’s discretion (State v. Nordstrom, 2001). General, educational testimony that satisfies Rules 702 and 403 is permitted (State v. Salazar-Mercado, 2014), but experts may not opine on whether a particular witness is truthful (State v. Lindsey, 1986, 149 Ariz. at 475). Testimony about the peak-end rule or duration estimates therefore stays at the level of how such judgments generally work.

Special Issues in Legal Practice

The principles described above apply with equal force to every party in a legal proceeding. The science operates the same way regardless of who is asking the questions, and the research summarized here does not by itself favor either side. What differs between the parties is not the science but the vantage point — the particular places where these issues tend to surface, and the particular errors each party is best positioned to guard against. The two summaries below identify those recurring issues. They are offered for orientation, not as litigation strategy; in every case the controlling question is what the evidence in that case will, and will not, support, and the expert’s role is to inform the trier of fact rather than to advocate for either side.

Special Issues for Prosecutors

The research supports several points prosecutors can legitimately present. A witness who dwells on the worst moment and the ending of an assault, and who cannot say precisely how long it lasted, is displaying an ordinary feature of how people remember extended experiences, which is not, by itself, a sign of exaggeration. Errors in estimating duration were unrelated to the accuracy of memory for the event in the leading study, so an inflated estimate of how long something lasted does not, by itself, indicate that the rest of the account is unreliable. Prosecutors are well positioned to elicit a witness’s account through open, sequential questioning that brings out the course of the event, and to establish duration and sequence through independent evidence rather than through the witness’s estimate.

The research also counsels caution. A witness’s overall characterization of severity and a stated duration are retrospective judgments, not measurements, and presenting them as a precise record of what occurred goes beyond the research. Where an element, a sentencing factor, or an identification depends on how long something lasted, the research on overestimation of brief stressful events (one 1987 set of laboratory studies of a videotaped robbery) suggests that the estimate should be corroborated. The same principles apply to prosecution witnesses who estimate how long they observed a suspect.

Special Issues for Defense

The research supports several points defense counsel can legitimately present. Retrospective evaluations give disproportionate weight to peak and end moments, so a global characterization may not reflect the full course of an event, and in one 1987 set of laboratory studies (Loftus et al.) participants overestimated the duration of a videotaped robbery, and a more stressful version produced larger overestimates in one experiment. Where a case depends on a stated duration, such as the length of time a witness viewed a face under Biggers, defense counsel are well positioned to test the estimate against video, records, and the sequence of actions described, and general expert testimony can explain why duration estimates are often imprecise.

The research also counsels caution. These findings describe how experiences are summarized and how durations are judged; they do not show that an event did not occur or that the witness’s account of what happened is inaccurate, and arguments that treat an imprecise duration or an emphasis on the worst moment as a sign of fabrication go beyond the research. The meta-analysis found that peak and end were about as predictive as the average of the whole experience, so a witness’s summary is not unrelated to what happened overall. An expert may describe these tendencies but may not tell the factfinder that a particular witness’s characterization or estimate is mistaken.

This paper is for educational purposes only and is not legal advice. The author is a psychologist, not an attorney; summaries of law are general and must be verified against current authority or with a licensed attorney.

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, Article 104149. https://doi.org/10.1016/j.obhdp.2022.104149

Fredrickson, B. L., & Kahneman, D. (1993). Duration neglect in retrospective evaluations of affective episodes. Journal of Personality and Social Psychology, 65(1), 45–55. https://doi.org/10.1037/0022-3514.65.1.45

Joint Service Committee on Military Justice. (2025). Amendments to the Military Rules of Evidence since 1984 (G. E. Maggs, Ed.; 2025 ed.). https://jsc.defense.gov/Portals/99/MRE%20Amendments%202025.pdf

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. https://doi.org/10.1111/j.1467-9280.1993.tb00589.x

Lane, S. M., & Houston, K. A. (2021). Understanding eyewitness memory: Theory and applications. New York University Press.

Loftus, E. F., Schooler, J. W., Boone, S. M., & Kline, D. (1987). Time went by so slowly: Overestimation of event duration by males and females. Applied Cognitive Psychology, 1(1), 3–13. https://doi.org/10.1002/acp.2350010103

Redelmeier, D. A., & Kahneman, D. (1996). Patients’ memories of painful medical treatments: Real-time and retrospective evaluations of two minimally invasive procedures. Pain, 66(1), 3–8. https://doi.org/10.1016/0304-3959(96)02994-6

Redelmeier, D. A., Katz, J., & Kahneman, D. (2003). Memories of colonoscopy: A randomized trial. Pain, 104(1–2), 187–194. https://doi.org/10.1016/S0304-3959(03)00003-4

Rode, E., Rozin, P., & Durlach, P. (2007). Experienced and remembered pleasure for meals: Duration neglect but minimal peak, end (recency) or primacy effects. Appetite, 49(1), 18–29. https://doi.org/10.1016/j.appet.2006.09.006

Scharbert, J., Utesch, K., Reiter, T., ter Horst, J., van Zalk, M., Back, M. D., & Rau, R. (2025). If you were happy and you know it, clap your hands! Testing the peak-end rule for retrospective judgments of well-being in everyday life. European Journal of Personality, 39(1), 55–69. https://doi.org/10.1177/08902070241235969

Legal Authorities

Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).

Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).

Manson v. Brathwaite, 432 U.S. 98 (1977).

Neil v. Biggers, 409 U.S. 188 (1972).

State v. Chapple, 135 Ariz. 281, 660 P.2d 1208 (1983).

State v. Lindsey, 149 Ariz. 472, 720 P.2d 73 (1986).

State v. Nordstrom, 200 Ariz. 229, 25 P.3d 717 (2001).

State v. Salazar-Mercado, 234 Ariz. 590, 325 P.3d 996 (2014).

United States v. Houser, 36 M.J. 392 (C.M.A. 1993).

Fed. R. Evid. 702.

Mil. R. Evid. 702.

Mil. R. Evid. 1102.

Ariz. R. Evid. 702.

Ariz. Sup. Ct. Order R-23-0004 (amending Ariz. R. Evid. 702, effective Jan. 1, 2024).

Related topic papers: Anchoring in Witness Estimation · Emotion, Stress, Trauma, and Memory · Fading Affect Bias · Rosy Retrospection · Eyewitness Memory · Memory for Violence and Weapon Focus

Related topics from Arizona Forensic Services: Counterintuitive Testimony