Why does the visual system adapt and how are adaptation mechanisms realized in the visual system? Answers to this venerable question has proven itself elusive to researchers in the visual sciences. Our assumption is that visual adaptation is the manifestation of a rational system whose sensory responses are optimized with regard to the uncertainty about the extrinsic environment and intrinsic neural mechanisms, in combination with the hard wired physical constraints imposed upon the system through evolutionary development. With this view, we search for Bayesian accounts of visual adaptation as a theoretical approach that can link empirical observations with computational modelling.
For example, an optimized system need not adapt nor change its setting if the visual environment matches that expected by the system. As such, a Bayesian might predict that visual adaptation is a manifestation of an unexpected or unpredicted visual environment. Empirical evidence supports this view. For example, rapidly flickering patterns are somewhat unusual, but ideal visual stimuli from which effects of visual adaptation can be observed empirically. The research has applied benefits, in that migraine suffers have been noted to suffer most in those range of flicker to which the visual system most adapts.