Abstract
The influence of cognitive processes on low-level perception is a classical question. This study tested whether individual differences in working memory (the ability to hold information in mind during distraction) are related to differences in color memory and/or color constancy. High- and low-working-memory participants were identified using the Aospan and Arspan tasks (Unsworth, Heitz, Schrock, & Engle, 2005). Participants studied a test color for one minute for later recall (paradigm modeled after Jin & Shevell, 1996). In the uniform background condition, the test color was surrounded by a uniform achromatic background that reflected all wavelengths nonselectively; in the complex background condition, the achromatic background had eight different colored sectors embedded within it. The two conditions manipulated context: a complex background typically improves color constancy. During study, each participant saw the simulated (CRT) display under one illuminant (A or C). After study, participants generated random numbers for two minutes (long-delay condition) or ten seconds (short-delay condition) in the dark, to prevent rehearsal of the test color while taxing working memory. A second display was then presented that had either the same illuminant used during study or the other illuminant. Participants adjusted the color of a test patch to appear like the color they had studied (a memory match). The paradigm allowed measurement of both color memory (no illuminant change between study and test) and color constancy (illuminant change between study and test). No significant difference was found for high- and low-working-memory participants for color memory, which depended on only remembering the test stimulus chromaticity. Better color constancy, however, was found for high- than low-working-memory participants in the long-delay condition. This suggests that working memory is involved in the development and/or delayed recall of a color-constant neural representation, but not in simple color memory for a chromaticity always viewed under the same illuminant.
Supported by NIH grant EY-04802.