Is perception probabilistic? Clarifying the definitions

A fundamental question for perception research is what sensory information is available for decision making, or, stated differently, what is the output of perception. One answer that has emerged in the last two decades is that perception is probabilistic, meaning that the brain represents probability distributions over world states. However, despite the apparent simplicity of this statement, there are substantial disagreements on exactly what probabilistic perception is and how it should be tested. In this adversarial collaboration, two proponents (Jehee and Denison) and two skeptics (Rahnev and Block) of probabilistic perception deliberate on the terms of the debate and present arguments for or against the notion that perception is probabilistic. We believe that this collaboration helps clarify the critical issues that need to be considered but that further work is required to reach consensus.