Hindsight bias in assessing child sexual abuse

Abstract

ABSTRACT Two studies using large samples of jury-eligible adults tested whether hindsight influences perceptions of potential grooming behaviours. In study 1, participants (n = 371) were presented with vignettes describing five different interactions between an adult and children and rated the likelihood that these behaviours were indicative of sexual abuse in foresight and in hindsight. In hindsight, the ratings of the behaviours doubled in size, evidence of hindsight bias. This finding held even for participants who had received training on how to identify possible sexual abuse. Study 2 used the same stimuli but randomly assigned participants (n = 159) to a hindsight or not condition; estimates of the likelihood that the adult is a child molester were double the size in the former group compared to the latter group. These findings have implications for civil lawsuits occasioned by sexual abuse that occurs within youth-serving organisations.