The Secret Agent Effect: A self-serving discrepancy in estimates of covert peer-to-peer learning

Abstract

This study sought to examine the differences between how one assesses his own knowledge of his peers and how he assesses his peers’ knowledge of him. The experiment tested the hypothesis that in general, people perceive themselves as being more informed about their peers’ involvements and affiliations than those peers are about theirs. The research provides evidence strongly supporting the existence of such a bias, dubbed the “secret agent effect.” 49 Princeton undergraduates evaluated their own knowledge of classmates depicted in randomly selected photographs, then estimated the classmates’ knowledge of them. Perceived knowledge about Princeton eating club membership, athletic participation, academic focus, and romantic history was gauged, among perceived knowledge of other categories. The discussion explores implications of the secret agent effect throughout everyday social contexts, and poses questions about its manifestations for further research.