colour vision

The human eye. The retina of the eye contains about 137 million light-sensitive cells in an area of about 650 sq mm/1 sq in. There are 130 million rod cells for black and white vision and 7 million cone cells for colour vision. The optic nerve contains about 1 million nerve fibres. The focusing muscles of the eye adjust about 100,000 times a day. To exercise the leg muscles to the same extent would need an 80 km/50 mi walk.

The ability of the eye to recognize different frequencies in the visible spectrum as colours. In most vertebrates, including humans, colour vision is due to the presence on the retina of three types of light-sensitive cone cell, each of which responds maximally to a different primary colour (red, green, or blue).

Perceived colours are functions of the state of the brain, as well as of physical features of objects. They remain more or less stable, and objects remain recognizable, in spite of the continuously changing illumination in which they are seen, a phenomenon known as colour constancy.

Early discoveries In 1802 the British physicist Thomas Young proposed that the enormous variety of colours in the visual spectrum could be accounted for by only three types of ‘particle’, or cell, in the retina, each corresponding to one of three colours - red, blue, or yellow. His idea was taken up and modified in the mid-19th century by the German physiologist Hermann Helmholtz, who argued that each type of light-sensitive cell, though sensitive to wavelengths over much of the spectrum, is especially sensitive to one of three particular wavelengths - red, green, or blue. The Young-Helmholtz trichromatic, or three-colour, theory has inspired much research and has been particularly useful in explaining inherited colour defects (colour blindness), and the fact that a mixture of three coloured lights can match any other coloured light. However, it cannot explain the phenomenon of colour constancy and says little about what happens behind the retina. The English physicist Isaac Newton demonstrated in the 17th century that ordinary white light is a mixture of lights of different wavelengths, with long wavelengths appearing red, middle wavelengths green, short wavelengths blue, and so on. This led him to assume, not unreasonably, that an object's colour derives from the wavelengths it reflects. According to this theory, a red object, for example, appears red because it mainly reflects long-wavelength, or red, light. This is an oversimplification. A red object appears red in a wide range of lighting conditions (nd) even when it is actually reflecting predominantly middle-wavelength or short-wavelength light.

20th-century discoveries In a remarkable series of experiments, beginning in the 1950s, the US inventor Edwin Land (1909-91) showed that the brain registers the varying intensities of light, or lightnesses, in each of the long-, middle-, and short-wavelength bands from all the coloured surfaces in a scene, compares them simultaneously to produce three ‘lightness records’, and then compares these lightness records to construct colours. Land called his theory the retinex theory. It has received striking support from physiological studies of single cells in the visual cortex, most notably by Semir Zeki of the University of London, and from studies of patients who have become colour blind as a result of injury to the brain.