In 1993, Daniel Kahneman — the psychologist who would later win a Nobel Prize in economics — ran an experiment that sounds like it should be illegal. Participants submerged one hand in 14°C water for sixty seconds. Then they submerged the other hand in the same 14°C water for sixty seconds, plus thirty more seconds while the temperature rose to a barely-less-painful 15°C. Same cold, more time, strictly more discomfort. Then Kahneman asked which trial they’d be willing to repeat. And a large share of participants chose the longer one — the objectively worse one — because it ended better. The paper’s title said it all: When More Pain Is Preferred to Less. Your memory of an experience is not a video recording; it’s a two-point summary written by the most intense moment and the last one. Psychologists call it the peak-end rule.
🧠 The Colonoscopy That Ended Too Early
The cold-water lab result got its most striking real-world confirmation in 1996, when Kahneman and physician Donald Redelmeier studied patients undergoing colonoscopies — an unpleasant procedure by anyone’s standards. Patients reported their pain in real time, and their retrospective ratings of the whole experience tracked almost perfectly with two data points: the worst moment and the final moment. Length barely mattered. Then the researchers went further. They randomly split patients into two groups, and for one group they left the scope in place for three extra minutes at the end — uncomfortable, but not painful. The patients who endured the longer procedure rated the entire colonoscopy as less unpleasant than the patients whose procedure ended sooner, and they were far more willing to return for a repeat screening. More pain, better memory, better future compliance. The peak-end rule is why.
🤔 The Summary That Throws Away the Middle
The counterintuitive core is something psychologists call duration neglect. We assume that a longer, more painful experience is remembered as worse — that the memory is a faithful tally of every moment. It isn’t. When the brain compiles its retrospective summary, the middle of the experience barely counts. Only the peak and the end make it into the final draft. That’s why a vacation with one awful day and a magical finale is remembered as a great vacation, and why the twenty-minute wait at a restaurant is forgotten the moment the waiter appears with an apology and a free dessert. There are boundary conditions, of course — a 2008 study found the rule is a good predictor but not an outstanding one, and day-long experiences like hotel stays seem to escape it. But for events with a clear beginning and end, the two data points rule.
🔗 Design the Ending First
For anyone who designs experiences — a product, a conversation, a story — this is one of the most actionable laws in the book. A session with a digital companion is judged the way a colonoscopy is: by its peak and its end. The last message of a conversation deserves deliberate design, because a warm, grounded goodbye rewrites the memory of everything that came before it. The same physics governs serial fiction: readers remember a chapter by its climax and its closing line, which is why cliffhangers are not a gimmick but a law of memory. And in pricing, the peak-end rule quietly shapes what customers think a product is worth. The internal reference price — the benchmark against which every price is judged — behaves like a weighted average of the highest price seen and the most recent price. That’s why aggressive discounts can permanently erode the price a customer will pay later: the low number becomes the peak in their memory of the brand.
🎲 The Pizza Buffet Test
Here’s a fun way to catch the rule working on you. In a study of pizza buffets, people who paid $4 judged the meal by the taste of the last slice and the best slice; people who paid $8 judged it by the first slice — higher expectations moved the deciding moment to the beginning. Next time you’re at a buffet, notice which bite writes your verdict. And here’s the practical version for a bad day: if today went badly, engineer a better ending. Read two pages of a book you love, listen to an old song, call someone who makes you laugh. Tomorrow, when you remember today, you won’t remember the middle — you’ll remember how it ended. Give the summary something good to work with.