Categorical perception for red and brown.

Abstract

Recent studies suggest that the widely accepted evidence in support of categorical perception of color may be a confound of effects due to low-level sensory mechanisms that are unrelated to color categories. To reveal genuine category effects, we investigated the category boundary least prone to spurious effects of low-level mechanisms: the boundary between red and brown. We tested for low-level sensory and high-level cognitive effects of categories on color discrimination, while carefully controlling potential factors of color vision that are not related to color categories. First, we established the red-brown boundary through a naming task and measured just-noticeable differences (JNDs) for colors across the boundary. If low-level sensitivity to color differences was categorical, JNDs should decrease toward the boundary. However, this was not the case. Second, we measured performance in terms of response times and error rates in a speeded discrimination task with color pairs that were equalized in discriminability based on the empirical JNDs. There was a boost of performance (lower response times and error rates) for identifying color differences in equally discriminable color pairs, when the colors crossed the boundary. Given the particularity of the red-brown boundary, these results prove that the observed effects were due to color categories rather than low-level visual mechanisms. These findings support the idea that category effects are due to a shift of attention to the linguistic distinction between categories, rather than being a pure product of perception. These category effects do not depend on the hemispheric lateralization of language. (PsycINFO Database Record