Peak–end rule

The effect is among the best established in the psychology of decision, and it is worth distinguishing the three studies that support it, because they do different things. In 1993, Kahneman, Fredrickson, Schreiber and Redelmeier had people plunge a hand into water at 14 degrees for 60 seconds and then repeat the same thing with 30 seconds more, during which the water rose to 15, still painful and slightly less so. When it came to choosing which one to repeat, most chose the longer version, and that is where the rule got its name. In 1996, Redelmeier and Kahneman followed 287 patients through procedures that were already scheduled, recording pain minute by minute, and the memory of the whole tracked the worst moment and the final moment, without longer procedures being remembered as worse. It was observation, with nothing manipulated. In 2003, with Katz, came the test. There were 682 randomised patients, half receiving an extension with the scope held still at the end, final pain 1.7 versus 2.5, and a less bad overall rating. None of the three measured the brain. Episodic memory and retrospective judgement are the ground where the effect lives, and forming and retrieving an episode depends on the hippocampushippocampusThe structure that turns recent experience into lasting memory. Everything your interface asks the person to remember tomorrow passes through it today.See it in Memory →See in the glossary → and the medial temporal cortexmedial temporal cortexThe inner part of the temporal lobe, hippocampus and surroundings, through which fact and episode pass to become lasting memory. Its role is temporary: over time the memory ends up stored in the cortex and stops depending on this region.See it in Memory →See in the glossary →, as SquireSquireOne reference in this work:1991The medial temporal lobe memory systemSee in the bibliography → and Zola-Morgan reviewed in 1991. The link between those structures and the peak–end rule is mine.

an era

Invest in the high point and the ending. The purchase flow that went smoothly and froze on the confirmation screen is remembered for the freeze; the support call that took a long time and ended with the problem solved is remembered for the resolution. In the most-cited figure of the 1996 study, two patients appear. One spent about eight minutes and ended at the peak, the other spent about twenty-four with the pain falling until the end, and it was the second who kept the less bad memory. It is a pair of cases, not an average, because the comparison between groups only came in 2003, with the examination extended by a few minutes with the scope held still. Long journeys can be remembered well, as long as they have a real high point and end well.

• The user decides in the first 50 milliseconds, the first impression settles everything.

  The original finding is about a judgement of appearance formed very quickly, not about a decision to use. And this rule says almost the opposite: what remains of the whole experience is the peak and the end, not the opening. If the first impression settled everything, there would be no reason to look after the ending.