Abstract
Humans break the color wheel into smaller segments, each labeled by a color name. Some individuals employ more names to segment the color wheel and some, less sophisticated with color names, use fewer names. Here we measured the relationship between the number of color names and color matching performance across individuals. 100 hues (equal saturation and brightness) were evenly sampled from the color wheel and presented to the subjects (n=10, Persian speakers, ages 20-30) in two conditions. In "perceptual matching" condition, a test-patch and a reference-patch were displayed simultaneously. The subjects adjusted the color of the test-patch to match it to the reference-patch. The "memory based matching" condition was the same except that the reference-patch was displayed for 10s, then the test-patch was presented following a 10s interval. In a separate session, the same subjects were presented with the 100 tested hues and asked to type their names (Persian). The number of unique color names (for describing the 100 tested hues) and the average error for color matching (the difference between observed and matched hues) were correlated across subjects for the two conditions. The more color names the subjects possessed the less color matching error they had for memory based matching (r= -0.68, p< 0 .05). This correlation was not significant for perceptual matching (r= -0.02, p=0 .25). Next, we collapsed the data from all subjects and defined "name-density" (ND); the number of color names in Persian for 10-hues-wide bins of the color wheel centered at each hue. The ND of each hue was negatively correlated with its color matching error for memory based matching (r=-0.27, p< 0.05) and not for perceptual matching (r=0.11, p=0.18). Our results reveal the correlation between color memory and color naming, both across individuals and along the hues of the color wheel.
Meeting abstract presented at VSS 2016