How we remember our enjoyment of past experiences isn’t always reliable, according to new research. The study explains that humans tend to remember those that end well as more enjoyable and those that end poorly as less enjoyable, even if the two were equally pleasant.

The findings showcase why we shouldn’t blindly trust our past experiences to inform decisions in the present. If we keep in mind that the last bits of any experience have a disproportionately high effect on our memory of it, we’ll be able to make better choices, the authors hope.

Happy ending?

“When you’re deciding where to go for dinner, for example, you think about where you’ve had a good meal in the past. But your memory of whether that meal was good isn’t always reliable — our brain values the final few moments of the experience more highly than the rest of it,” said Dr Martin Vestergaard, a researcher in the University of Cambridge’s Department of Physiology, Development, and Neuroscience, who led the study.

This preference seems to come built-in in humans, Vestergaard explains. Its effect seems to dampen over time — our memory of something we did a long time ago will factor less into our decision-making than a more recent one.

The process has its roots in two different brain areas which are activated whenever we try to make a decision based on experiences in our memory. However, they also compete with each other while doing so, making us either overvalue experiences that started badly and end well or undervalue experiences that started well and end poorly.

A part of the brain known as the amygdala then uses our memories to work to determine the ‘objective value’ of an experience (such as how tasty a meal was). The other, called the anterior insula, makes older memories progressively less important in our decision-making process. This holds true even among our most recent memories — the further back in time it is, the less it factors into our decisions.

For the study, the team enlisted 27 healthy male volunteers and asked them to estimate which one of two pots of coins on a screen had the greatest total value (these pots were shown one at a time, not side-by-side). They were also shown how coins of varying sizes fell from the pots in quick succession. A functional magnetic resonance imaging (fMRI) machine was used to see what the participants’ brains were doing during the experiment. The task was repeated several times with different sequences of coins.

Participants routinely chose the wrong pot when the coins shown decreased in size by the end of the sequence, the team explains. This suggests that their brains were using this more modest end as a cue to estimate a lower total value. Although the intensity of this effect seemed to vary between participants, only a handful were able to completely bypass it and make rational estimations, according to the authors.

Such results suggest that our current theoretical models on decision-making — chiefly that sub-obtimal decision-making is handled by the amygdala, with higher brain areas handling more complex decisions — is correct. But on a personal level, they showcase to each of us how the final moments of an experience influence our perception of the whole, especially when judging from memory.

“Our attraction to the quality of the final moment of an experience is exploited by politicians seeking re-election; they will always try to appear strong and successful towards the end of their time in office,” said Vestergaard.

“If you fall for this trick, and disregard historical incompetence and failure, then you might end up re-electing an unfit politician. Sometimes it’s worth taking the time to stop and think. Taking a more analytical approach to complement your intuitive judgement can help ensure you’re making a rational decision.”

The paper “Retrospective valuation of experienced outcome encoded in distinct reward representations in the anterior insula andamygdala” has been published in the Journal of Neuroscience.