Why Do We Misremember the Pain of a Colonoscopy?

Two patients undergo the same medical procedure. One suffers for 8 minutes; the other, for 24. Days later, the one who suffered less remembers the experience as worse. This paradox, which Daniel Kahneman recounts in Thinking, Fast and Slow (2011), came from a real study he designed with physician and researcher Donald Redelmeier of the University of Toronto, and it shows something counterintuitive about the memory of pain: we don’t remember the total amount of suffering we went through, but a much simpler summary of that experience.

The experiment: 154 patients and a pain scale every 60 seconds

The study, published in 1996, involved 154 patients undergoing a colonoscopy, at a time before the procedure was routinely performed under sedation. It also included 133 lithotripsy patients, the treatment that breaks up kidney stones with shock waves, with similar results.

Every 60 seconds, patients were asked to report their level of pain at that exact moment, on a scale from 0 (no pain) to 10 (unbearable pain).

The length of the procedure varied a lot from one patient to another: the shortest lasted 4 minutes, the longest 69 minutes.

At the end, each patient was asked how much pain, in total, they felt they had experienced during the whole procedure.

Two patients, two very different experiences

Kahneman illustrates the finding by comparing two specific patients from the study:

• Patient A: the procedure lasted 8 minutes. Their last pain rating, just before it ended, was 7 out of 10.
• Patient B: the procedure lasted 24 minutes, three times as long as patient A’s. Their last pain rating, just before it ended, was only 1 out of 10.

If you add up the pain moment by moment across the whole procedure (what Kahneman calls the “area under the curve”), there’s no possible doubt: patient B, exposed for much longer to comparable levels of pain, objectively suffered more in total than patient A.

The result that didn’t fit the logic

Yet when both patients were asked how much pain they remembered feeling in total, the result was the opposite of what you’d expect: patient A remembered the experience as worse than patient B did, despite having suffered, in objective terms, considerably less accumulated pain.

The statistical analysis of the 154 patients revealed two systematic patterns, which Kahneman named:

• Peak-end rule: the retrospective rating of a painful experience is explained mainly by the average of the worst moment (the peak) and the level of pain right at the end (the end).
• Duration neglect: how long the experience actually lasted has almost no influence on how it’s remembered afterward.

Applying this rule to the two patients: both reached the same peak of pain (8 out of 10) at some point during the procedure.

But the average of that peak and the final level was 7.5 for patient A and only 4.5 for patient B.

Patient A simply had the bad luck that their procedure ended at a moment of high pain, and that was enough to leave a worse overall memory, regardless of the fact that in total they had suffered for less time than patient B.

The lab test: more pain, better memory

A study with real patients has one limitation: it can’t control how each procedure ends. That’s why Kahneman, Barbara Fredrickson, Charles Schreiber and Redelmeier (1993) took the idea into the lab with a cold-water experiment.

Each participant went through two trials. In the short one, they kept one hand in water at 14 °C (57 °F) for 60 seconds. In the long one, they kept the other hand in water at 14 °C for 60 seconds and then 30 seconds more while the water slowly warmed to 15 °C: it still hurt, but a little less. Afterward, they had to choose which of the two to repeat.

22 of the 32 participants (69%) chose to repeat the long trial, which contained all the pain of the short one plus another half minute. Among those who clearly noticed the final relief, the proportion rose to 81%. The authors put it bluntly: if people chose to minimize their exposure to pain, that proportion would be zero.

Fredrickson and Kahneman (1993) found the same pattern with pleasant and unpleasant film clips: the length of each clip barely affected how it was rated afterward. That’s what they called duration neglect.

Two different ways of measuring the same thing

Kahneman calls the real moment-by-moment sum of pain (what an outside observer would calculate by adding up all the ratings) hedonimeter totals, and distinguishes them from retrospective assessments, which are what each patient remembers afterward.

His conclusion is that these two measures can differ systematically: the first weighs every minute of the experience equally; the second only really pays attention to two specific instants, the worst and the last.

This led Kahneman to propose a distinction he developed in the rest of the book: the difference between the experiencing self, which lives each moment separately, and the remembering self, which later builds a condensed story of the whole experience, and which in practice is the one that decides whether we’d do something again.

The two “selves” don’t always agree on how much suffering actually occurred.

From the lab to the clinic

The practical consequence seemed clear: if the goal is for the patient to remember the procedure as less painful, how it ends may matter more than how long it lasts. In 2003, Redelmeier, Joel Katz and Kahneman put this to the test in a randomized trial with 682 colonoscopy patients.

For a randomly chosen half, a short interval was added at the end during which the instrument stayed still, without advancing. The results were modest but consistent:

• The final moments were less painful (1.7 versus 2.5 on a scale from 0 to 10).
• The whole experience was remembered as somewhat less unpleasant (4.4 versus 4.9).
• Over a follow-up of about five years, patients in the extended group returned a little more often for their next recommended colonoscopy.

That last finding is the most relevant one, because colonoscopy is a screening test that’s usually repeated at intervals, and a bad memory can weigh on the decision to come back. It’s an unusual example of a memory bias turned into a measurable clinical improvement.

What later research confirms

A meta-analysis by Balca Alaybek and colleagues (2022), covering 174 effect sizes from very different studies, found solid support for the peak-end rule: the combination of peak and end was strongly related to the overall memory (r = 0.58) and weighed more than the beginning of the experience, its mildest moment or its trend. The effect of duration was practically zero.

There’s a nuance the popular version usually leaves out: in that same meta-analysis, the average of the whole experience predicted the memory just as well as the peak and the end. Memory doesn’t ignore the rest of what was lived; what it does is give disproportionate weight to two moments and almost none to how long it lasted.

This same mismatch between what’s experienced and what’s remembered shows up in many everyday decisions that have nothing to do with medicine, from how we rate a trip or a job to how a more analytical profile or a more intuitive one later judges whether an experience was worth it, a judgment that is rarely based on the actual sum of what was lived.

It’s reasonable to think something similar happens when remembering a bad stretch for one’s investments, where a single day of sharp losses can weigh more than how the whole period went, although that particular case isn’t the one Kahneman and Redelmeier studied. Our investor profile quiz explores how you tend to react to swings like that.

Sources

• Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux.
• Alaybek, B., Dalal, R. S., Fyffe, S., Aitken, J. A., Zhou, Y., Qu, X., Roman, A. and 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, 104149. https://doi.org/10.1016/j.obhdp.2022.104149
• Fredrickson, B. L. and 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
• Kahneman, D., Fredrickson, B. L., Schreiber, C. A. and 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
• Redelmeier, D. A. and 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. and 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