Proceedings of the Annual Meeting of the Cognitive Science Society

Adults category learning is accelerated by redundant verbal labels (Lupyan et al., 2007). However, it is an open questionhow category representations are affected by labeling. Here, we presented subjects with a learning task that involvedseparating sine wave gratings of differing spatial frequency and orientation into two categories. Categories of easy, mediumand difficult separability were constructed. Participants (N=128) either received only feedback sounds during training, orheard verbal labels in addition. Growth curve analysis (Mirman, 2014) was used, fitting 2nd order polynomials to the dataacross the learning phase. In addition to main effects of difficulty on intercepts and the linear time term, the best-fittingmodel showed an effect of labeling on the linear time term, with steeper learning curves in conditions with labeling. Therewas no interaction of labeling and difficulty, indicating that the impact of labeling is similar across the types of categoriesused here.