Abstract
Purpose :
There is evidence for OFF midget retinal ganglion cells (RGCs) with single S-cone receptive field centers near the fovea in macaque retina. To evaluate the role of RGCs with S vs. (M+L) cone inputs in hue perception, we performed color-matching experiments in which subjects compared the appearance of lights that differ greatly in the extent to which they stimulate S-cones.
Methods :
A Maxwellian view spectrospatiotemporal stimulus generator was used in color-matching experiments in which subjects were asked to adjust a mixture of 470 nm and 640 nm monochromatic lights (10 nm bandwidth) to match the appearance of a 430 nm monochromatic reference light. To minimize influence of the macular pigment, the wavelengths were presented as a split annulus with stimulus at 6.5 +/- 0.5 degrees visual eccentricity. A random-dot pattern white gaussian noise was positioned such that it overlapped all the edges of the annular stimulus, which masked fringe effects resulting from chromatic aberration between the 470 nm and 640 nm mixture lights. Subjects fixated at the annulus center while adjusting the relative amounts of 470 nm and 640 nm lights to get a hue match to the 430 nm light, and then could adjust the intensity of the fixed 470 nm and 640 nm ratio to achieve a brightness match. Subjects were allowed to fine-tune their settings until they were satisfied with their match. The stimuli were calibrated with a Konica Minolta CS-2000 spectroradiometer and the quantal catches associated with the stimuli were calculated by integrating with corneal sensitivities of S-, M-, and L-cones.
Results :
Each participant was able to make color matches between the mixture and comparison lights even though the difference in quantal catch of the three cone types is associated with a large S-cone decrement and L+M cone increment when comparing the mixture to the 430 nm light. This difference in S-cone stimulation produces large differences in activity for both S-OFF midget and small bistratified RGCs, yet subjects couldn't perceive any associated difference in hue.
Conclusions :
The firing of action potentials by S vs. (L+M) RGCs does not contribute to hue perception in a color match. This is consistent with the hypothesis that the visual system uses RGCs with S vs. (L+M) cone inputs for detection of both chromatic and isoluminant boundaries but there are separate parallel RGC pathways for hue perception.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.