Predicting focal colors with a rational model of representativeness


Best examples of categories lie at the heart of two major debates in cognitive science, one concerning universal focal colors across languages, and the other concerning the role of representativeness in inference. Here we link these two debates. We show that best examples of named color categories across 110 languages are well-predicted by a rational model of representativeness, and that this model outperforms several natural competitors. We conclude that categorization in the contested semantic domain of color may be governed by general principles that apply more broadly in cognition, and that these principles clarify the interplay of universal and language-specific forces in color naming.

Back to Table of Contents