Back to overview

The peak-end rule: live fast, love hard, die young

Evaluating the past requires recollection of how we felt during a specific event or episode. With many moments making up an episode, an efficient way is needed to summarize them so that a single evaluation can be reached. One way to achieve this could be to evaluate each moment separately and then average the individual scores in order to reach an overall evaluation rating. Sounds reasonable, right? Wrong. Research shows that it is the extremities of a specific event or episode (that is, peak positive or negative feelings) and the concluding moment (end) that weigh most on our overall evaluation of past experiences [1]. So what implications does this have for brands?

Two interesting consequences of these findings are duration neglect and trend overemphasis. In short, evaluations of past events are highly sensitive to the trend of the experience and not so much to its duration. In theory, adding negative/unpleasant moments should make the overall experience worse, just as adding positive/pleasant moments should make it better – this is known as temporal monotonicity. Duration neglect violates temporal monotonicity. In hindsight, people seem to prefer experiences with an improving trend even if this implies that the duration of an unpleasant event will be extended or that of a pleasant event will be shortened [2].

This counterintuitive prediction was supported by several studies on pleasant and unpleasant experiences [3]. In one such study, participants immersed their hands for 60 seconds in 14°C ice water followed by 30 seconds in 15°C ice water (long duration). They were also asked to do so for 60 seconds in 14°C ice water alone (short duration). When the participants were later asked which of the two conditions they preferred, they chose the longer one seemingly preferring more pain over less (at least in terms of duration). The results suggest that retrospective evaluations are mainly influenced by the peak and ending moments which, in turn, can result in a shorter experience being evaluated as more painful than a longer one with a better ending.

This finding was replicated in real-life settings. Patients rated colonoscopy procedures as being more pleasant when they were extended to include milder pain towards the end [4]. The memory for the event was highly correlated with that judgment thus, increasing patients’ willingness to repeat the procedure. It appears that adding more pain to the end of an episode can actually improve its retrospective evaluation – and its chances to be repeated – as long as the end pain is less than the peak pain.

How about positive experiences? Fredrickson and Kahneman (1993) [1] showed participants a series of video clips that varied in pleasantness and duration. Once again, the overall evaluation of the film clip was better predicted by the peak moment and the ending while duration had no effect. Marketing researchers have also taken interest in whether the peak-and end rule and duration neglect apply to people’s evaluations of televised advertisements. Across three studies, 156 participants watched and later evaluated a series of commercials that varied in duration (from 30 to 90 seconds), intensity of positive emotion, and latency of peak positive affect. Results indicated that viewers most preferred commercials that achieved a high peak positive affect and ended on a strong positive note. They only liked longer adverts more when the added time was used to build to a higher peak experience [5].

An extreme example of the peak-end effect is on the evaluation of one’s life. Diener, Wirtz, and Oishi (2001) [6] presented people with two life scenarios: the first was a wonderful life that ended suddenly while the second was the same life that was extended with some mildly pleasant years. Participants evaluated the shorter life higher than its extended version (the James Dean effect). In 1955, Faron Young had already expressed this idea in his popular song ‘Live Fast, Love Hard, Die Young’. Judging whether someone had a valuable, desirable life is not necessarily affected by longevity (e.g. Alexander the Great, Wolfgang Amadeus Mozart, Joan of Arc) but still we all desire longevity for ourselves. Schreiber and Kahneman (2000) [7] made a clear distinction between judgments made when looking forward as opposed to when looking back. It could be the case that temporal monotonicity applies to ongoing experiences, while it does not apply when these same experiences are later recalled. Looking ahead, we hope for more positive experiences thus we do not exhibit duration neglect and we desire longer experiences. When looking back though, e.g. abstractly evaluating life, we focus more on the experience itself and we value it higher if it tells a good story neglecting its duration.

Overall, striving for a better ending appears to be the key to positive evaluations and objective happiness. Albert & Kessler (1978) [8] have already shown that people intuitively try to end their social interactions on a positive note (e.g. ‘at least we got the hardest part of the job done’, ‘this was fun’, ‘have a good weekend’). Additional research shows that people prioritise the negative experiences over the positive ones in an effort to create happy endings e.g. visiting a close friend after a depressing relative than the other way round [9] [10]. Our life choices – ranging from what we will have for lunch to what career we will follow – reflect our prediction for our future happiness [11]. These predictions are often based on our past experiences and their evaluation. Duration may not always be neglected and the peak-end rule may not be the sole predictor of these retrospective evaluations. Still, those two key moments – the most intense and the final moment – can quite often and quite effectively predict our evaluations of past experiences and guide which experiences we will repeat or avoid.



[1] Fredrickson, B. L., & Kahneman, D. (1993). Duration neglect in retrospective evaluations of affective episodes. Journal of personality and social psychology, 65 (1), 45.

[2] Fredrickson, B. L. (2000). Extracting meaning from past affective experiences: The importance of peaks, ends, and specific emotions. Cognition & Emotion, 14(4), 577-606.

[3] 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.

[4] 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.

[5] Baumgartner, H., Sujan, M., & Padgett, D. (1997). Patterns of affective reactions to advertisements: The integration of moment-to-moment responses into overall judgments. Journal of Marketing Research, 219-232.

[6] Diener, E., Wirtz, D., & Oishi, S. (2001). End effects of rated life quality: The James Dean effect. Psychological Science, 12(2), 124-128.

[7] Schreiber, C. A., & Kahneman, D. (2000). Determinants of the remembered utility of aversive sounds. Journal of Experimental Psychology: General, 129(1), 27.

[8] Albert, S., & Kessler, S. (1978). Ending social encouters. Journal of experimental social psychology, 14(6), 541-553.

[9] Loewenstein, G. F., & Prelec, D. (1993). Preferences for sequences of outcomes. Psychological review, 100(1), 91.

[10] Ross, W. T., & Simonson, I. (1991). Evaluations of pairs of experiences: A preference for happy endings. Journal of Behavioral Decision Making, 4(4), 273-282.

[11] Loewenstein, G., & Schkade, D. (1999). Wouldn’t it be nice? Predicting future feelings. Well-being: The foundations of hedonic psychology, 85-105.

Back to overview

Write a comment

*required field

Leave a Reply

Your email address will not be published. Required fields are marked *

Your comment*