Balca Alaybek

Who they are

A meta-analysis pools the results of many separate studies so the combined answer does not rest on any one of them. Hers pooled 174 separate results, which is a large body of evidence for a question of this kind.

Nothing beyond her authorship on this paper has been verified here, so no affiliation is asserted.

Key papers

Alaybek, B., Dalal, R. S., Fyffe, S., Aitken, J. A., Zhou, Y., Qu, X., Roman, A., & Baines, J. I. (2022). "All's well that ends (and peaks) well? A meta-analysis of the peak-end rule and duration neglect", Organizational Behavior and Human Decision Processes, 170, 104149. First author. One hundred and seventy-four effect sizes.

Key ideas

Two results are widely repeated. The peak and the ending together are a strong predictor of how an experience is rated afterwards. How long it lasted predicts essentially nothing, which is Duration Neglect.

The third result is the one nobody repeats, and it is why this page exists. The peak-and-end composite predicted the later rating about as well as the simple average of the whole experience, and no better. It beat the trend, the variability, the beginning and the low point. It did not beat the mean.

So the popular claim that the middle is discarded is not what this meta-analysis found. What is discarded is the clock, not the middle. That distinction is carried at Peak-End Rule, where the strong popular version does not survive.

Both sides in one paper. In favour of the peak-end idea: the association with the later rating is strong and well evidenced. Against it: a plain average does the same job, which removes most of the reason to treat it as a special mechanism. This site keeps the effect and drops the strong claim.

Concepts on this site

Duration Neglect, the half of the pair her paper strengthens

Peak-End Rule, the half it weakens

Rosy Retrospection, what happens to an evaluation with time

---