Purchase this article with an account.
B. Sayim, K.A. Jameson, N. Alvarado; Color Naming Relations in Perceptual Color Space . Invest. Ophthalmol. Vis. Sci. 2003;44(13):1912.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: Research on color representation and categorization frequently uses color naming behavior to define the perceptual color category structures (e.g., Jameson & Alvarado 2003, Guest & Van Laar 2000, Boynton & Olson 1990, 1987). A basic assumption in doing so is that relations among color terms are isomorphic to (or can be proxies for) relations among color percepts. We investigate this assumption by comparing similarity judgments for color words with judgments of color appearances. One goal is to determine the 'color naming function(s)' that link lexical category labels to perceptual color space. Methods: The cognitive tasks used are (1) triadic similarity judgments, between lexical color terms and color appearance stimuli (Shepard & Cooper, 1992), and (2) judgment tasks explicitly linking color terms and appearances. For example, in the lexical triads subjects evaluate three color words and choose the one word that is most different from the others in the triple. This 'odd-man out' judgment is used in separate balanced incomplete-block designs for colors and lexical terms for each participant. The primary hypothesis evaluated is whether symmetry is found in the mapping of color names to color appearance, and color appearances to color names. We define such a mapping as the cognitive 'naming function' between color perceptual representation and linguistic representation. Results: Similarity distance measures are derived from the triad data to produce non-metric multidimensional scaling representations of judged domains for each participant and observer group (e.g., dichromat, anomalous trichromat). For both lexical and color appearances both within-category, and across-category conditions are judged by each participant. Individual and group scaling solutions are assessed to evaluate (i) the degree of correspondence between color similarity scalings for terms and colors, and (ii) the correspondence between scaling representations in similarity tasks compared to explicit naming tasks. Conclusions: Results found for the cognitive 'naming function' of color representation suggest that two different representational models of color experience are warranted. One is the lexical representation which largely conforms to the normative linguistic representation of the population considered, and another is the separate, individually variable, perceptual representation organizing the structural similarities between an observers color sensation experiences. This empirical demonstration of two different representations, lexical and perceptual, call into question the appropriateness of assuming that judgments made in one mode can serve as a valid proxy for the other mode.
This PDF is available to Subscribers Only