peak-end rule error

A very interesting psychological theory I came across is the peak-end rule which plays a very important role in human decision-making. I initially came across this theory while reading Yuval Harari’s ‘Homo Deus’ which is a potential account of how the future might look like where robust data-driven decision-making powered by computer algorithms will supersede human-driven decision-making. He cites the research done by famous psychologist Daniel Kahneman on the psychology of human decision making to make his case. Humans continuously collect and store information based on all the experiences they undergo which aid in their future decision-making process. Daniel Kahneman dichotomized the human-self into the experiencing self and remembering self. The experiencing self is a version of the self that comes into the picture at the time when the person is undergoing the experience and all information in regard to that experience is relevant. The remembering self on the other hand is a version of the self that remembers and records information of only the key moments (i.e. peak and end moment hence peak-end rule) of the experience which is objectively different from the information of the experiencing self. The remembering self plays a dominant role in the decision-making process of humans which is why an experience that was an objectively bad experience on an overall level may appear favorable to the remembering self if the peak and end moments of that experience happened to be favorable. Companies make use of this principle in optimally designing their user experiences to have a good peak and end point so that it can be remembered more favorably in the minds of consumers. This dominance of the remembering self in the decision-making process can be problematic as it is a function of incomplete data. Hence, advanced algorithmic decision-making powered by machines as projected in ‘Homo Deus’ may prove to be a superior alternative to entirely replace human decision-making as it mitigates for the peak-end rule error of the remembering self and ultimately produce objectively accurate decisions that account for all data points of past experiences rather than just the peak and the end.