Purchase this article with an account.
Chuan-Chin Chiao, Charles Chubb; Dimensions of preattentive visual sensitivity in human color space. Journal of Vision 2010;10(7):1359. doi: 10.1167/10.7.1359.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
This study used texture discrimination tasks to investigate preattentive visual sensitivity to equiluminant chromatic variations. Specifically, we looked for evidence of “half-cardinal-axis” mechanisms in DKL space - i.e., mechanisms sensitive exclusively to variations between neutral gray and each of the red and green poles of the L-M axis and the blue and yellow poles of the S-(L+M) axis. Observers strove to discriminate spatially random mixtures of colors called scrambles. A given scramble is characterized by its color histogram. The preattentive sensitivity space of a set C of colors is the space of histogram differences people can discriminate given a brief display. The dimensionality of this space gives the number of preattentive mechanisms sensitive to C variations. In these experiments, briefly presented stimuli comprised alternating bars of scramble differing in histogram, and observers had to judge bar pattern orientation. We used a new method called “iterated seed-expansion” to obtain a basis of the sensitivity space for each of 6 different sets of colors: a set drawn from each of the half-DKL-cardinal-axes in the equiluminant plane and also from each of the full L-M and S-(L+M) cardinal axes. For each of these sets C, the sensitivity space proved to be two-dimensional, with one basis element showing linear and the other parabolic sensitivity for the colors in C. This suggests that associated with each full-cardinal-axis is one linear mechanism L and one second-order mechanism S derived from the L output. Our results support the idea that for any element x in the scramble, S(x) is proportional to the squared difference between L(x) and the mean L-output in the neighborhood of x. These same two mechanisms suffice to account for discrimination of half-axis scrambles; thus, we find no evidence for separate half-axis mechanisms.
This PDF is available to Subscribers Only