Non-informative verbal labels can facilitate concept formation when compared to a situation in which no labels are present. Importantly, non-informative nonverbal labels (like movements) do not facilitate concept formation in the same way (Lupyan et al., 2007). Using a different stimulus set, we attempted to conceptually replicate this verbal-label effect. In a training phase, we compared a verbal-label, a nonverbal-label and a no-label group regarding their performance in categorizing two types of fish. All groups received auditory feedback after their categorization response. In addition, the verbal-label group received a written pseudo word for each of the two categories, while the nonverbal-label group saw a specific movement instead of a verbal label. The no-label group received no additional information. In a later test phase, categorization was investigated without feedback or labels. During training, the verbal-label group outperformed the no-label group. At the end of training, this group also outperformed the nonverbal-label group. In the subsequent test phase, however, there was only a tendency towards an advantage of the verbal-label group. Although we could not fully replicate the results by Lupyan et al. (2007), this study provides clear evidence towards a special role of verbal labels during concept formation.

Lupyan, G., et al. (2007). Language is not just for talking: Redundant labels facilitate learning of novel categories. Psychological Science, 18(12), 1077–1083.