How can we determine which aspects of a given experience are represented perceptually, as opposed to cognitively? I explore perceptual adaptation as an empirically based method for distinguishing perception from cognition. I focus on vision. The basic idea is that, if a represented property shows adaptation, then it is perceptually represented. I explore why, and under which conditions this is true. In short, if a given property shows adaptation, then we must rule out that the effect (1) can be explained by a cognitive shift in category boundary, and (2) that it is a result of adaptation to other (low-level) visual features of the stimuli, before we can conclude that the property is perceptually represented. I discuss methods for excluding these possible explanations. I also discuss whether we should expect all perceptually represented properties to show adaptation. As evidence suggests that this is plausible, I conclude that the adaptation methodology has the potential to determinably distinguish perception from cognition.