Recognition thresholds for an alphanumeric resolution test object of differing target/background chromaticities and equal luminosities were determined. Combinations of the four psychologically unique hues were studied, and those pairs corresponding to the green-red and yellow-blue opponent neural systems were found to give the highest visibilities. Equivalent achromatic contrasts for each of the presented color pairs were determined, and comparison of the results with achromatic data of other researchers shows clearly that the spectrally opponent neural network in the human visual system is much less efficient in processing spatial information than is the spatially opponent (achromatic) neural network.
Library of Congress Subject Headings
Vision--Testing; Color vision; Visual perception
Department, Program, or Center
School of Photographic Arts and Sciences (CIAS)
Smith, Douglas, "Recognition Thresholds for the Spectrally Opponent Neural Network in the Human Visual System" (1976). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus