The evaluation of diagnostic explanations for inconsistencies

When individuals detect an inconsistency between a fact and their beliefs, they revise their beliefs. They also use their causal knowledge to create explanations of what led to the inconsistency. According to the theory in the present paper, an ideal explanation is a chain of a cause and an effect, where the effect explains the inconsistency. Two experiments corroborated this account. When participants evaluated explanations for inconsistencies, they rated a conjunction of a cause and its effect as more probable than the cause alone, which they rated as more probable than the effect alone. This trend violates the laws of probability – it is an instance of the "conjunction fallacy". It also violates the common assumption that individuals make minimal changes to their beliefs.