Published online by Cambridge University Press: 01 January 2000
Psychophysical (behavioral) detection thresholds and color-discrimination thresholds were determined in a macaque using a two-alternative forced-choice procedure. On a white background, detection thresholds were determined for a white increment and three spectral increments: 618, 516, and 456 nm. Intermixed with detection threshold determinations, color-discrimination thresholds were determined by presenting the white increment, and one of the spectral increments, at 1.0 log units above their respective detection thresholds and dimming both until discrimination performance fell to threshold. The monkey could discriminate the color of the increments at detection threshold because the average color-discrimination threshold was 0.98 ± 0.14 log attenuation. Because the monkey was much more sensitive to the spectral increments than the white increment, we performed an unconventional experiment. We determined the monkey's detection threshold for the white increment alone, and with broadband color filters in the white light path without adjusting the light's intensity. Insertion of several color filters in the light path lowered detection thresholds of both the macaque and six human trichromats. We believe that this improvement in detection thresholds produced by simply inserting color filters in a white light path is a threshold manifestation of the Helmholtz-Kohlrausch effect and suggests that one of color vision's important evolutionary advantages may be improved detection sensitivity.