Proceedings of the Annual Meeting of the Cognitive Science Society

How do category systems reflect the information content of their environments? One basic kind of information in a lin-guistic environment is the frequency of objects or meanings: some things are just spoken about more often than others. Agreat deal is known about frequency effects on the evolution of lexical items (e.g. Lieberman et al, 2007); however anal-ogous effects on category systems are not understood. Two theories point in opposite directions: the generalized contextmodel (Nosofsky, 2011) predicts that categories containing high-frequency items will expand over time, while informationtheory (Cover & Thomas, 2012) predicts tighter boundaries around high-frequency items. We explore the impact of fre-quency on the evolution of category systems over time in an iterated category learning experiment that manipulates objectfrequency. How does this manipulation affect category boundaries? Does the result change if transmission is betweendifferent individuals or within the same person over time?