Animal Color Vision

Critical Definitions and Underlying Biology

Light reaches the eye directly from illuminants, such as a computer monitor, or as reflected from surfaces of objects, such as an apple sitting on the [Page 49]desk. Ignoring changes that may occur across space or over time, light can be physically characterized as varying along just two dimensions—wavelength and intensity. The presence of color vision is formally defined as the capacity of an individual to successfully discriminate between lights or surfaces based on differences in their wavelengths, irrespective of what their relative intensities may be. Note that the definition asks nothing about the quality of what is seen, only whether there is consistent discrimination. If any animal can make such discriminations, it has color vision.

Once the presence of color vision is established, it becomes possible to further characterize the nature of the capacity, for example, by asking what particular wavelengths can and cannot be successfully discriminated. Each of the large range of such tests can provide useful information about the color vision of any given animal. A particularly important description of color vision comes from using a testing procedure called color matching. In such a test, lights of different wavelengths are added together in an attempt to find out which mixtures of lights appear identical to other lights. Although the analysis of such matches can be a bit complex, what results is a description of the dimensionality of color vision. In such an account, if an animal lacks color vision completely, it is characterized as being monochromatic (i.e., it experiences a world that is devoid of color variation), an animal with a single dimension of color vision is called dichromatic, two dimensions are trichromatic, three dimensions tetrachromatic, and so on. As the dimensionality of color vision increases, there are corresponding and dramatic increases in the number of separate colors that can be discriminated. For example, it is estimated that a human having dichromatic color vision (a common form of color blindness) can discriminate among some 10,000 surface colors, whereas a human with trichromatic color vision, normal for our species, has the capacity to discriminate something in excess of one million colors.

There is a compelling link between the dimensionality of color vision and the biology of the eye. In the eye, light is absorbed by photopigments that are located in the photoreceptors—in vertebrates, these receptors are called cones. Different types of photopigments vary according to the wavelengths to which they are most sensitive. The number of types of photopigment that are active in daylight viewing is typically directly related to the dimensionality of color vision—monochromats have only a single type of cone pigment, dichromats have two, trichromats have three, and so on. Because of this linkage, it is possible to infer the nature of an animal's color vision from an analysis of the number of types of photopigment contained in eye or, because the photopigment proteins (opsins) are specified by single genes, from direct studies of these opsin genes. Both of these shortcuts allow scientists to learn something about the nature of color vision in a large number of animals for which behavioral tests of color vision may be impractical for one reason or another. This tactic can even be used to infer what color vision must have been like in some ancestral species.

...