Sexual Selection and the Mismeasure of Color

Abstract

Many long-standing evolutionary hypotheses make predictions about trends in color patterns. Examples of these include crypsis, mimicry and warning coloration, fruit coloration, flower coloration, the handicap principle of honest advertisement, Fisher's runaway process, the parasite theory of sexual selection, and sensory drive theories of signaling. The majority of tests of these hypotheses, particularly with regard to sexual selection, have been conducted on objects that birds perceive visually, with human vision used to assess color. This assumes that birds see color patterns as humans do, an assumption that is seriously flawed. First, birds see very well parts of the spectrum that humans cannot. Second, birds have at least four dimensions to their color vision, compared to only three in humans. Third, birds have a complex system of oil droplets in their retinas, which may alter the number of hues they perceive. Thus, an object will not appear to have the same hues for a human and a bird, and maybe not even the same relative brightness or saturation. Despite this, human vision is routinely, and almost without exception, the method used for assessment of color patterns seen by birds. We argue that the error in this assumption may well be a major reason that support for various evolutionary hypotheses involving color is an area of controversy. We also suggest methods for overcoming the shortcomings of existing studies and suggest which previous approaches are likely to have been most erroneous. As part of this, it is apparent that a research program in color cognition is necessary, for if we wish to understand evolutionary hypotheses involving color we need to understand how animals perceive color. Color is not an inherent property of the object; it is a product of the brain of the animal perceiving the object. Humans can see hues seen by birds as much as knowledge of x and y coordinates can predict the position of a point in three-dimensional space.