From Experience to Memory: On the Robustness of the Peak-and-End-Rule for Complex, Heterogeneous Experiences

Abstract

Memory forms the input for future behavior. Therefore, how individuals remember a certain experience may be just as important as the experience itself. The peak-and-end-rule (PE-rule) postulates that remembered experiences are best predicted by the peak emotional valence and the emotional valence at the end of an experience in the here and now. The PE-rule, however, has mostly been assessed in experimental paradigms that induce relatively simple, one-dimensional experiences (e.g., experienced pain in a clinical setting). This hampers generalizations of the PE-rule to the experiences in everyday life. This paper evaluates the generalizability of the PE-rule to more complex and heterogeneous experiences by examining the PE-rule in a virtual reality (VR) experience, as VR combines improved ecological validity with rigorous experimental control. Findings indicate that for more complex and heterogeneous experiences, peak and end emotional valence are inferior to other measures (such as averaged valence and arousal ratings over the entire experiential episode) in predicting remembered experience. These findings suggest that the PE-rule cannot be generalized to ecologically more valid experiential episodes.

Integrating the Peak-end Rule and the Kano Model in Assessing the Product-use Process

“…Average valence, which some previous studies have claimed is a better predictor than peak-end valence for retrospective assessment [22,23], does not be the mediation in our study. However, immediate satisfaction fully mediated the relationship between peak-end valence and retrospective satisfaction, which was unexpected.…”

“…• Positive peak valence: the highest value for the emotional valence during the experience, which can happen at one or several steps during the experience [21,22]. • Negative peak valence: the lowest value for the emotional valence during the experience, which can happen at one or several steps during the experience [21,22]. • End valence: the final value for the emotional valence at the end of the experience.…”

“…• Average valence: the average of the emotional valence for every step during the experience. Some studies have reported that the average valence is a better predictor than the peak-end valence for the recall of an experience [22][23][24][25]. • Immediate satisfaction: the overall evaluation of an experience assessed immediately after completing all steps in the product-use process.…”

More is Not Better: The Emotional Dynamics of an Excellent Experience

“…Peak-end theory posits that the best predictor of overall evaluation is an average of the strongest emotional moment, and the last emotional moment, of that experience (Kahneman, Fredrickson, Schreiber, & Redelmeier, 1993). This theory is largely based on studies of brief, simple experiences of pain, and does not hold for longer, more complex experiences such as workaday life or tourism, where the average of emotions experienced is a better predictor (Kemp, Burt, & Furneaux, 2008;Miron-Shatz, 2009;Strijbosch et al, 2019). These findings suggest that the more positive episodes that exist during an experience, regardless of their order, the more positive the overall evaluation will be.…”

Peak-at-end rule: adaptive mechanism predicts time-dependent decision weighting

“…These observations are insufficient to support the peak-end rule because there is an inherently high correlation between a peak value and the overall average of sequential data 44 , thus hindering reliable conclusions www.nature.com/scientificreports/ about the peak-end rule. In fact, some studies have reported conflicting results that the average score is better than the peak score as a predictor of overall evaluations of experiences 45,46 . In any case, simply calculating the correlation between a single value extracted from an overall dataset and a retrospective value is insufficient to validate the peak-end rule.…”