Understanding Higher Order Color: Beyond the Cardinal Mechanisms

Northeastern University, Boston MA

Investigators

Abstract

Humans typically rely on vision as their primary sense in their interactions with the world. Understanding how higher-order cortical brain processes contribute to human visual perception is a critical issue in cognitive science. Color offers a unique opportunity for understanding the brain mechanisms of perception because the very first step in color processing, the absorption of light by cells in the retina (photoreceptors), is fully understood. This knowledge allows the color scientist to manipulate early signals in the nervous system and measure the perceptual result. Perceptual and neurophysiological experiments have shown that signals from the different photoreceptors in the eye are combined, approximately by addition and subtraction. Our ability to see colors and to discriminate one color from another are the result of these neural sums and differences. For about the past half-century, perceptual scientists have tried to understand the number and nature of these neural color mechanisms: Are there few color mechanisms or many? Do they just add and subtract photoreceptor signals or do they perform more complex calculations? The present work combines computational modeling with novel experimental techniques and strategies. In some cases, random visual flickering elements will appear over a test stimulus. If the noise and test are processed through the same color mechanisms, the noise should hinder the ability of a person to see the test color (by a process like camouflage). If the noise and test are processed in separate mechanisms, the noise should have no effect on the ability to see the test at all. The properties of the color mechanisms will be quantitatively studied by varying the color of the noise relative to the test. Understanding how the brain processes color information is an important part of the more general understanding of how we perceive the world around us. There are practical applications of this research as well: Color is an important part of signaling systems and information displays. Having a quantitative model of color vision will help the designer of cockpit and automotive displays, medical information and imaging displays, and even digital television systems, to more efficiently and accurately convey information to users.