Phong Zhong Law

The peak-end rule is a principle of retrospective evaluation: people commonly summarize an episode by averaging the emotional intensity of its peak and its final phase, with little regard for how long the episode lasted. Its decisive significance is that the remembering self can replace the duration-weighted record of the experiencing self, so a longer ordeal may be remembered more favorably than a shorter one if it ends better.

Definition and Operation

The rule emerged from a distinction between two measures of experience. A hedonimeter total treats every moment as part of an area under the curve: pain or pleasure accumulates across time, so two minutes at a given intensity count twice as much as one minute. Retrospective evaluation follows another logic. It gives disproportionate weight to the worst or most intense moment and to the way the episode ends, while duration may have no detectable effect. The resulting memory is therefore not a simple summary of everything that happened.

This difference separates experienced utility from decision utility. The experiencing self communicates moment by moment, whereas the remembering self later assigns a global verdict and uses that verdict in subsequent decisions. The peak-end rule does not claim that every detail disappears; it identifies the particular compression by which two singular moments dominate the remembered whole.

Colonoscopy Evidence

Daniel Kahneman and Don Redelmeier’s colonoscopy study made the divergence visible. In the early-1990s experiment, 154 patients reported their pain every 60 seconds on a scale from zero, meaning no pain, to 10, meaning intolerable pain. One illustrated procedure lasted eight minutes and another 24 minutes, although the larger study included procedures ranging from four to 69 minutes. Moment-by-moment reports showed that the longer procedure exposed its patient to at least as much pain at every level and produced the larger duration-weighted total.

When the procedures ended, however, patients were asked to rate the total amount of pain they had experienced. Their answers did not reproduce the accumulated totals. Retrospective ratings were well predicted by the average of the pain at the worst moment and the pain at the end, while the duration of the procedure had no effect whatsoever. The shorter procedure therefore left the more unpleasant memory in the illustrated comparison: both patients reached a peak of 8, but the final rating was 7 for the eight-minute procedure and only 1 for the 24-minute procedure. Their peak-end averages were consequently 7.5 and 4.5.

The result creates a striking reversal. Patient B endured the worse experience in the ordinary, duration-sensitive sense, yet patient A remembered the episode as worse because the shorter procedure ended at a bad moment. The rule thus turns the ending into a kind of retrospective verdict: gradual relief can matter more to memory than a large reduction in total exposure.

Cold-Hand Experiment and Choice

The rule also shaped behavior in the cold-hand experiment. Participants held a hand in painfully cold water while using the other hand to record their momentary pain continuously. The short episode lasted 60 seconds in water at 14 degrees Celsius. The long episode began identically but lasted 90 seconds; during its final 30 seconds, slightly warmer water raised the temperature by roughly one degree and produced a detectable decrease in pain for most participants.

From the perspective of the experiencing self, the long trial was plainly worse because it contained the same first 60 seconds plus another 30 seconds of discomfort. The peak-end rule, combined with duration neglect, predicted a different retrospective judgment: the long episode would be remembered as less aversive because it ended during relief. When participants were later told that one of the two experiences would be repeated exactly, 80 percent of those who reported a decline in pain during the long episode chose to repeat it. They thereby selected 30 seconds of needless additional pain in the anticipated third trial.

This choice did not reflect a desire for suffering. Participants knew which exposure had lasted longer, and they would have selected the short option if the alternatives had been described as the 90-second immersion versus only its first 60 seconds. Instead, they chose the episode they disliked least in memory. The peak-end rule therefore links retrospective evaluation to decision-making: an ending that improves the remembered story can make people voluntarily repeat an objectively worse experience.

Measurement and Well-Being

The same contrast informs the measurement of well-being. Daniel Kahneman initially approached happiness with suspicion because standard life-satisfaction questions ask the remembering self to evaluate life as a whole. To study the experiencing self, he and colleagues sought measures of momentary feeling rather than relying only on a global judgment.

Experience sampling interrupted people at random intervals through a programmed phone and asked what they were doing, who was present, and how intensely they felt happiness, tension, anger, worry, engagement, or physical pain. Because continuous reporting would prevent ordinary living, the researchers also developed the Day Reconstruction Method. Participants reconstructed the previous day as a sequence of episodes, like scenes in a film, then reported their activities, main focus of attention, social contacts, and feelings on separate scales from zero to six. Episode durations allowed the researchers to calculate duration-weighted daily affect and compare it with life satisfaction, understood as a judgment of the remembering self.

National surveys later extended this distinction to millions of respondents. The Gallup World Poll asked about emotions experienced during the previous day across the United States and more than 150 countries, confirming the importance of situational factors, physical health, and social contact in experienced well-being. The peak-end rule belongs to this larger argument: a retrospective life judgment and a duration-sensitive account of lived moments are related, but they are not interchangeable measures.

Significance

The peak-end rule exposes a persistent cognitive illusion: memory can make an experience seem to have been better or worse than its full temporal record warrants. It explains why a bad final moment can contaminate the memory of a shorter procedure, why gradual relief may be preferable to abrupt relief when future memory matters, and why people can choose to repeat a longer ordeal. Its practical force lies in the separation of two aims. If the objective is to reduce what patients remember, lowering the peak and softening the ending may be more important than minimizing duration; if the objective is to reduce suffering as it is lived, every painful moment and the total time spent in pain remain relevant.