The world in color presents a dazzling dimension of phenotypic variation. Biological interest in this variation has burgeoned, due to both increased means for quantifying spectral information and heightened appreciation for how animals view the world differently than humans. Effective study of color traits is challenged by how to best quantify visual perception in nonhuman species. This requires consideration of at least visual physiology but ultimately also the neural processes underlying perception. Our knowledge of color perception is founded largely on the principles gained from human psychophysics that have proven generalizable based on comparative studies in select animal models. Appreciation of these principles, their empirical foundation, and the reasonable limits to their applicability is crucial to reaching informed conclusions in color research. In this article, we seek a common intellectual basis for the study of color in nature. We first discuss the key perceptual principles, namely, retinal photoreception, sensory channels, opponent processing, color constancy, and receptor noise. We then draw on this basis to inform an analytical framework driven by the research question in relation to identifiable viewers and visual tasks of interest. Consideration of the limits to perceptual inference guides two primary decisions: first, whether a sensory-based approach is necessary and justified and, second, whether the visual task refers to perceptual distance or discriminability. We outline informed approaches in each situation and discuss key challenges for future progress, focusing particularly on how animals perceive color. Given that animal behavior serves as both the basic unit of psychophysics and the ultimate driver of color ecology/evolution, behavioral data are critical to reconciling knowledge across the schools of color research.