Abstract
Once acquired, status as a theoretical virtue is rarely lost. But recent philosophical criticisms of parsimony argue its scope is rather limited and its justificatory basis quite thin. In fact, psychological studies cast a pall on positive assessments of parsimony. They show simplicity considerations frequently derail correct probabilistic inference. We investigate another way theoretical virtues may corrupt reasoning: the terms labeling theoretical virtues themselves produce a framing effect. Our results show merely labeling explanations “simple” exacerbated participants’ neglect of base-rate information. This finding complements recent empirically-oriented criticisms and broaches underexplored issues about whether “theoretical virtues” involve false advertising.
