The retinal mosaic, eye movements and colour vision - WCUB, ENWW

Human colour vision is mediated by three different types of photoreceptive cell (long- (L), middle- (M) and short-wavelength (S) cone cells), each sensitive to different, overlapping sections of the visible light spectrum. Different wavelengths of light therefore drive a different balance of responses in these three cell types: the sensation of hue is derived by comparing signals between L, M and S cones. Among people with normal colour vision, who have little measurable difference in hue judgements between them, the ratio of numbers of L cones to M cones ranges from 1.1:1 to 16.5:1. How is it that an observer who has 16 times as many L cones as M cones can consistently agree with the hue percepts of an observer who has roughly equal numbers of L and M cones? The standard answer is that a compensatory neural mechanism adjusts the balance point of the redgreen opponent colour channel, based on visual experience. But other visual tasks - requiring discrimination of fine spatial patterns or rapid flicker - that depend on L and M cones do not show such thorough compensation. I propose to approach this question from two different avenues. Firstly, I will further develop the method for classifying individual cones in the living human retina, in order to bring us closer to our goal of ascertaining the visual percepts arising from the stimulation of single, identifiable cones in the context of their own class and of the classes of their neighbouring cones. Secondly, I will investigate the hypothesis that when a stimulus falls only on one class of cones, one of the tiny eye movements that occur throughout fixation will move the stimulus to an alternate class of cones, thus improving colour perception.

BBSRC priority areas
Our research into the challenges posed to colour vision by the retinal mosaic falls under the Cross Council priority area 'Brain science and mental health'. One of our approaches, to further develop adaptive optics techniques that allow us to image single neurons in the retina, also falls under the BBSRC priority area 'Technology development for the biosciences'.