Year: 2004
Type: article
Abstract: In this paper, a two by three approach to modeling categorization is presented.Similarity representations based upon a geometric, an additive tree and an additive cluster model are combined with an exemplar model and a prototype model in a single approach.The six models are applied to the categorization of pictorial known and unknown fruits and vegetables (Smits et al., 2002).For novel stimuli, the geometric exemplar model and the cluster models gave the best account, indicating a strategy where people compare stimuli with stored members on more general continua or a limited set of features.For well-known stimuli, the tree-based models gave the best account of the data, suggesting more elaborate taxonomic knowledge.More generally, the results show that different categorization models may perform better for different sets of stimuli, and that a systematic empirical comparison of such models is needed.
Authors: Timothy Verbeemen, Gert Stroms, Tom Verguts
Language: English
Subfield: Experimental and Cognitive Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
Funder: Vlaamse regering
OpenAlex ID: W2525203587