On The Psychology of Experimental Surprises: Outcome Knowledge and the Journal Review Process.

摘要

To merit publication a study must be informative; the reader must not feel that the results could have been confidently predicted prior to conducting the study. Earlier studies of the psychology of hindsight have shown that people overestimate the predictability of past events. This report examines whether they also overestimate the predictability of experimental results. Three experiments strongly suggest that reviewers may overestimate the degree to which they could have predicted the reported results of experiments. The implications of this bias for the journal review process, and some ways in which it might be alleviated, are discussed.