The research behind the Peak-End Rule
Think about a holiday you took several years ago. You probably don't remember an evenly weighted record of every hour you spent there. Certain moments are much easier to retrieve: an exceptional dinner, an argument, the view from somewhere beautiful, getting lost, the last evening before you went home.
Yet when someone asks whether the trip was good, you can probably give them an answer.
This creates an interesting psychological problem. Experiences unfold continuously, but our memories of them are incomplete. When we later evaluate whether something was enjoyable, painful or worth repeating, we have to construct that judgment from the parts we remember.
Research by Daniel Kahneman, Barbara Fredrickson and their colleagues suggests that some moments can carry considerably more weight in this retrospective evaluation than others. In particular, people often appear to be influenced by especially intense moments and by how an experience ends, a pattern that became known as the Peak-End Rule.
THE EXPERIMENT
One of the most memorable demonstrations came from a 1993 experiment by Kahneman and colleagues involving painfully cold water.
Participants completed two trials. In one, they submerged a hand in cold water for 60 seconds. In another, they experienced the same initial 60 seconds but kept their hand submerged for an additional 30 seconds. During those final 30 seconds, the researchers raised the temperature of the water slightly.
The water was still unpleasant. The longer trial therefore contained everything unpleasant about the shorter trial plus another 30 seconds of discomfort.
If people evaluated the experience by adding together the discomfort of every moment, the choice should have been straightforward. The 60-second trial contained less total discomfort.
But when participants were later asked which trial they would prefer to repeat, a majority chose the longer one.
The extra 30 seconds had not removed any pain. Instead, they had created a less unpleasant ending.
This produced a strange result: adding discomfort to an experience could improve how the experience was remembered, provided the additional discomfort was less intense than what came before it.
Kahneman and colleagues referred to this pattern as a violation of what they called temporal monotonicity. Logically, adding an unpleasant period to an already unpleasant experience should not make the overall experience better. Retrospective evaluation did not always follow that logic.
THE FINDING
The cold-water experiment contributed to a broader body of research examining the difference between an experience as it unfolds and the way that experience is evaluated afterwards.
People do not appear to calculate retrospective enjoyment or discomfort by simply adding together every moment and dividing by the duration. Instead, some parts of an experience can become disproportionately influential.
The peak refers to a particularly intense point in the experience. Depending on the situation, this could be the most pleasurable or most unpleasant moment. The end refers to how the experience concludes.
Duration can sometimes receive surprisingly little weight in comparison.
This does not mean that people literally forget everything except the peak and the ending. Memory is much more complicated than that. Nor does the Peak-End Rule perfectly predict how every experience will be remembered.
The more useful finding is that the remembered quality of an experience is not necessarily an accurate summary of every moment that composed it.
Some moments matter more to the later judgment than their duration alone would suggest.
THE RESEARCH
Researchers subsequently examined similar effects outside the laboratory.
In a well-known study published in 2003, Kahneman and colleagues studied patients undergoing colonoscopy procedures. One group received the standard procedure. For another group, the procedure was deliberately extended by leaving the instrument in place for a short period at the end without additional movement.
This meant the second group experienced a longer procedure. However, the final portion was less uncomfortable than the preceding part.
Patients who experienced this modified ending later evaluated the procedure more favorably overall. They were also more likely to return for a subsequent colonoscopy within the study's follow-up period.
Again, the researchers had not reduced the total duration of discomfort. They had actually increased it.
What changed was the trajectory of the experience and particularly the way it ended.
This distinction led Kahneman to write extensively about the difference between what is sometimes called the experiencing self and the remembering self. One concerns what life feels like while it is happening. The other constructs a story about what happened and uses that story when making future judgments and decisions.
NOT THE SAME THING
The Peak-End Rule does not mean only two moments matter
The theory has become popular in discussions of customer experience, travel, hospitality and behavioral design because it offers an appealing formula: create a great peak and a great ending and people will remember the experience positively.
The actual psychology is less tidy.
Research on retrospective evaluation suggests that peaks and endings can have substantial influence, but their importance varies depending on the type of experience, how it is measured, how long ago it occurred and what information remains available in memory. Duration is not always ignored and other meaningful moments can clearly influence how an experience is remembered.
The Peak-End Rule is therefore better understood as a systematic tendency in retrospective evaluation rather than a rule stating that the brain records only two pieces of information.
It is also important to distinguish memory from experience itself.
A painful medical procedure does not become less painful while it is happening simply because it ends relatively well. A miserable holiday does not retroactively contain fewer miserable hours because the final dinner was wonderful.
The remembered experience may improve.
The lived experience has already happened.
WHY IT MATTERS
This distinction becomes particularly interesting because memories influence future decisions.
You decide whether to return to a restaurant partly according to how you remember the previous visit. You decide whether a holiday destination was worth the money by reconstructing the trip afterwards. You decide whether you enjoyed an event, want to repeat a tradition or would recommend an experience to someone else using a memory that may place unequal weight on different moments.
That can sometimes create a mismatch between what produced the best experience in the moment and what produced the best memory afterwards.
Imagine two holidays. The first contains seven consistently enjoyable days without any extraordinary moment. The second contains more inconvenience and frustration but also one spectacular day that becomes a story you tell for years.
Which holiday was better?
There may not be a single answer. One could have contained more total enjoyment while the other produced the stronger memory.
The same distinction can apply to ordinary periods of life. When you think back on a particular year, you are unlikely to retrieve 365 equally weighted days. A few significant events can dominate the evaluation. A difficult ending to a job can influence how you remember several otherwise good years there. One exceptional summer can shape how you remember an entire period of your life.
This doesn't make memory defective. A perfectly detailed record of every moment would be extraordinarily inefficient. Memory is selective by necessity.
But it does mean that "How good was this experience?" and "How good do I remember this experience being?" are not always the same question.
TRY THIS
Choose an experience from the past year that you have a strong opinion about. It could be a trip, a project, a relationship period, an event or even a particularly difficult month.
First, give it your instinctive evaluation. Was it good, bad, enjoyable, stressful, disappointing or worth repeating?
Then reconstruct more of the experience than you normally would. What were the ordinary parts like? What happened at the beginning? What did most days actually feel like? What was the best or worst moment? How did it end?
Pay particular attention to whether one intense moment or the ending seems to be carrying a disproportionate amount of your overall judgment.
You do not need to correct the memory or decide that your original evaluation was wrong. Peaks and endings can be genuinely important parts of an experience.
The exercise simply separates two pieces of information that are easy to collapse into one: what stands out when you remember something and what it was actually like to live through it.
Both can tell you something useful. They just aren't always telling you the same thing.
References
Fredrickson, B. L., & Kahneman, D. (1993). Duration neglect in retrospective evaluations of affective episodes. Journal of Personality and Social Psychology, 65(1), 45-55.
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.
Redelmeier, D. A., Katz, J., & Kahneman, D. (2003). Memories of colonoscopy: A randomized trial. Pain, 104(1-2), 187-194.