The talk will identify several domains in which people overweight negative domains of an experience in their overall judgments.

I will begin by examining the peak-end rule, and demonstrate how, in multi-episode events (namely, days) the overall evaluation is based on the average of episode evaluations, and on the presence of a low, much more so than a peak. Peaks and lows are measured separately of the episode ratings. Ends do not contribute to the prediction of overall ratings, Thus, lows are rarer and more diagnostic than peaks, and certainly more so than ends.

Parsing overall evaluations into the evaluations of specific emotions, work done with Daniel Kahneman and Arthur Stone will demonstrate how people are more prone to peak-code negative emotions. This effect is exacerbated the higher the variance of the evaluations. From a personality perspective, peak-coding of negative emotions is more extreme on people high in negative affective disposition. To compare, those high in positive affective disposition peak-code positive events, but not to the same degree.

Finally, using another domain where experiences are extended, namely consumer behavior, I will examine how online reviewers rate their hotel stays and car ownerships. Here too we will find that the presence of a low, but not so much a peak, contributes to the prediction of the overall ratings.

To sum, the talk will display boundary conditions for the peak-end rule. In addition, I would like to claim that the presence of lows and of negative emotions overrides that of peaks and positive emotions in extended experiences.

Registration required:

No

Free:

Yes