Full Program »
Is Cognitive and Neural Processing Bayesian?Current Bayesian models of psychological and neural processing have an usually large number of degrees of freedom (e.g., priors, utility function, free parameters), and as a result Bayesian cognitive scientists and neuroscientists can model any psychological or neuroscientific phenomenon. The first goal of this presentation is to vindicate this concern by examining the robustness or brittleness of influential models. The second goal is to identify some telltale properties that a given psychological or neural process must have to be correctly modeled by a Bayesian model. The last goal is to examine whether psychological or neural processes possess these telltale properties.
Department of History and Philosophy of Science
University of Pittsburgh