Abstract
The ratio of middle (M) to long (L) wavelength sensitive cones exhibits striking variability across individuals. Despite this neuronal variability, previous studies have demonstrated that the spectral location of yellow is little influenced by L to M ratios ranging from 1:3 to 19:1 (Neitz et al., 2002). The extremely small variability in unique yellow despite cone ratio can be explained by the visual system's normalization to the mean spectral distribution of the environment. Here, we explore the question of how to reconcile the existence of a normalization process with the perception of unique green, which is known to vary by as much as 60 nm between individuals. We studied the experience of middle wavelength light using hue cancellation and unique hue measurements. Confirming previous findings, L:M ratio was not a factor for unique yellow. In striking contrast, unique green was highly dependent upon relative cone numerosity. This finding can be explained if the normalization mechanism has a very different effect on the blue-yellow opponent system than the red-green one, owing to differences in neural architecture of the two systems. A physiological model incorporating environmental normalization and L:M ratio accurately described the observed relationships between unique hues and cone ratio. The proposed circuitry combines potentials from short wavelength sensitive cones with L/M opponent signals in the outer retina to produce four lines leaving the eye; the stimulation of each being responsible for the sensation of a particular hue.