ArticleNagy AL, Boynton RM.
J Opt Soc Am. 1979 Sep;69(9):1259-65.
A color-naming method was used to examine the large-field red/green discriminations of dichromats screened with standard tests. The stimulus was a 12 degrees annular field with the central 4 degrees removed, flashed for 300 ms. Four wavelengths were equated in brightness for each observer at two retinal illuminance levels, approximately 10 and 100 trolands. The stimuli were then presented in random order and the observer was asked to name each, using one of four color terms. The entire experiment was done with the observers dark-adapted and also with the rods bleached. For all four deuteranopes and two of four protanopes, color names were very significantly related to both illuminance and wavelength in both adaptation conditions. The relationship between name and wavelength was similar to that of a normal trichromat, but the performance of a dichromat was very poor by comparison. Performance was generally somewhat better in the dark than with the rods bleached. However, the result in the bleached condition is consistent with recent evidence that at least some observers who are classified as dichromats with standard small-field screening procedures actually have a weak residual third cone mechanism.