Categorical color discrimination in the green sunfish, Lepomis cyanellus

Image contrast in the aquatic environment is limited by the amount of available light, by light scattering, and by the spectral transmittance properties of water. There are several ways in which the visual systems of aquatic animals may have evolved to cope with these limitations. Both color- and polarization-opponent processing have been hypothesized to enhance image contrast in aquatic environments. The purpose of these studies was to determine if green sunfish, Lepomis cyanellus, possess these processing capacities. To investigate this, I classically conditioned animals to associate a chromatic or polarized stimulus with shock presentation, and then tested their ability to discriminate that stimulus from another that was never paired with shock. The green sunfish retina possesses two cone receptor types; single cones $(\lambda\sb\max = 532$ nm) and double cones $(\lambda\sb\max = 621$ nm) arranged in a tetradic mosaic (Dearry and Barlow, 1987). In the chromatic experiments, I tested an animal\u27s ability to discriminate a 520 nm light from a 633 nm light. The intensity of each stimulus was systematically varied. Results indicate that these animals can discriminate between these two lights irrespective of relative intensity, and are thus at least dichromats. Moreover, I demonstrated that these animals categorize lights above 610 nm into one group, and those below 580 nm into another. In addition, to test the hypothesis about the double-cone mosaic, experiments were done to determine if green sunfish could learn to discriminate a moving unpolarized test field from a moving test field composed of $2.5\sp\circ$ stripes of alternating linear polarizations. They could not learn this discrimination. Control experiments with chromatically striped patterns of the same size/velocity showed that the inability to discriminate was not due to a general inability to discriminate spatial patterns moving at such velocities. Furthermore, experiments in which I measured an animal\u27s threshold to linearly polarized light showed that sensitivity did not vary with light polarization angle as previously reported (Cameron and Pugh, 1991). In sum, I found no evidence that green sunfish can use the polarization distribution to discriminate targets