Hue, Saturation, and Brightness: Cortical Emergence and Modeling of Color Percepts

点击翻译按钮获取中文摘要

What neural processes allow humans to perceive and experience different colors? The operation of human color vision has been among the most extensively investigated topics in the scientific study of human visual perception, yet its neural basis remains elusive. This project aims to investigate a new theory that could help to explain how the perception of different colors emerges from neural activity. Understanding the mechanisms of color perception is important to designing successful artificial vision systems and visual displays, as well as to understanding the basis of visual problems.Though many aspects of color vision are well-understood, attempts to find specific neuronal-level correlates of the various components of a color percept have failed. Specifically, psychophysics research has shown that there are three aspects to the perception of a color: 1) hue, 2) color purity (the difference between a "saturated" color and a pastel one), and 3) brightness, and researchers have long assumed that these properties result from relatively early-stage neural processing, at the level of neurons that specialize in reacting to these properties of a visual stimulus. However, existing research has failed to find consistent evidence that specific neurons specialize in reacting to these properties of a color percept. In this project, a new theory will be investigated that could help to explain how the three properties of color percepts emerge from neural activity. The new theory under investigation in this project is that the properties of color perception (hue, purity, and brightness) emerge from complex interactions among networks of neurons, involving competition in some cases and cooperation in others. In this way, the three color percept properties are emergent properties of complex neural activity, rather than being represented at the level of specialized neurons. In the project, the investigator will carry out sophisticated computational modeling of the proposed complex neural interactions in an effort to explain existing data patterns in the literature.