Proceedings of the Annual Meeting of the Cognitive Science Society

When learning to partition the world into categories, peoplerely on a set of assumptions (overhypotheses) about possi-ble category structures. Here we propose that the nature ofthese overhypotheses depends on the presence of a verbal la-bel associated with a given category. We describe a computa-tional model that demonstrates how labels can either acceler-ate or hinder category learning, depending on whether or notthe prior beliefs imposed by their presence align with the truecategory structure. This account provides an explanation forthe phenomena described in prior experimental work (Lupyan,Rakison, & McClelland, 2007; Brojde, Porter, & Colunga,2011) that have remained unexplained by other models. Basedon these results, we argue that the overhypothesis theory of la-bel effects provides a way to formalize and quantify the effectof language on category learning and to develop a more precisedelineation between linguistic and non-linguistic thought.