Colour vision

The immense range of colours that we see belies the relative simplicity of the system underlying colour vision. We see colours because each of the three distinct types of cone photoreceptor in our retinae is sensitive to a different range of wavelengths of light. The L, M and S cones, colloquially called red (R), green (G), and blue (B), are selective for long, middle and short wavelengths, respectively. The principle of univariance dictates that once a cone has captured a photon, its response will be the same, regardless of the photon’s wavelength (Fig. 1). Thus, two monochromatic lights at different wavelengths may evoke identical responses from the L cone (Fig. 1b). But because the cone spectral sensitivities are broad and overlapping, the M and S cones will also be activated, to different degrees, by the two lights. The triplet of cone responses will therefore be different for these two monochromatic lights, and so the perceived colours of the two lights will be different. Any two light spectra that evoke the same triplet of cone responses will appear identical in colour; such lights are called metamers. In the natural world, most lights, whether reflected from objects or emanating directly from a light source, are not monochromatic; that is, they contain many wavelengths. Because we have only the three cone types, each responsive to a broad segment of the spectrum, we cannot resolve the individual wavelengths of such natural spectra. Nonetheless, we can discriminate many millions of colours, corresponding to millions of distinct cone-response triplets. This much of colour vision is schoolbook stuff. It does not go much beyond Young’s fundamental insight into human trichromacy in 1802: “Each sensitive filament of the nerve may consist of three portions, one for each principal colour [red, yellow, and blue] . . . each of the particles is capable of being put in motion more or less forcibly by undulations differing less or more from perfect unison.” Since then, our knowledge of the perceptual subtleties of colour vision has vastly increased, as has our understanding of the neurophysiology underlying colour vision. Yet there is still an immense gap between the two; we are a long way from explaining many perceptual phenomena of colour vision in terms of specific neurophysiological mechanisms. In fact, perhaps the only certain link between perception and physiology is that the three cone types in the retina provide the basic trichromatic code (see blue box).