Abstract

Our aim was to investigate whether a Poisson distribution for quantum fluctuation is sufficient for estimating the ratio of long-wavelength-sensitive (L) cones to middle-wavelength-sensitive (M) cones in the human foveal mosaic. For this purpose we used a method of constant stimuli to estimate the psychometric function under two-point presentation. The test stimulus consisted of two 1.7′ spots separated by 30′. The test was presented for 1 msec on a 2° square background. The subject reported seeing 0, 1, or 2 points. The test/background combinations (in nanometers) used were 520/646, 646/520, 625/646, and 520/dark. Stimuli were generated by a two-channel computer-controlled direct-view apparatus with an image rotator for sampling various foveal locations. A total of 580 random trials per session was presented. The data were analyzed using a binomial probability for L and M cone sampling together with the usual Poisson probability distribution for stimulus quantal fluctuation. Our results examine whether the binomial assumption is necessary to accurately estimate the L/M cone ratio.

© 1985 Optical Society of America