Understanding how neural responses contribute to the diversity of avian colour vision

The past 20 years have seen a surge of interest in how animals perceive colour, setting the stage for a much more detailed examination of how colour perception differs among species, how a species’ colour perception relates to its environment and how it all fits into the framework of animal communication. We need to address two major questions: first, how do general mechanisms of signal processing work within whole clades of animals, and second, how do these mechanisms modulate differences among related species within clades? The receptor noise-limited (RNL) model (Vorobyev & Osorio, 1998) has made a critical advance in the field. Relevant parameters of the model, including screening pigments in the eye, visual pigments and relative numbers of the different receptor cells, can be measured to predict how species detect objects in their environment, based on wavelength. Details of the opponent channels, however, which compare the outputs of the retinal receptors and can determine the signals sent to the brain, are unknown for most species and are not required by the model. Here, we unpack the RNL model, focusing on experiments in humans and birds, and explore the impact of including specific opponent channels in the model. Incorporating opponent channels into the RNL model could help us better understand the selective forces and coevolutionary processes that shape the visual system and determine visual adaptations. Present evidence shows that the RNL model works as a good first approximation and points to critical parameters we need to measure, such as noise in receptor cells.