Just how conservative is conservative predictive processing?
Just how conservative is conservative predictive processing?

Summary/Abstract: Predictive Processing (PP) framework construes perception and action (and perhaps other cognitive phenomena) as a matter of minimizing prediction error, i.e. the mismatch between the sensory input and sensory predictions generated by a hierarchically organized statistical model. There is a question of how PP fits into the debate between traditional, neurocentric and representation-heavy approaches in cognitive science and those approaches that see cognition as embodied, environmentally embedded, extended and(largely) representation-free. In the present paper, I aim to investigate and clarify the cognitivist or ‘conservative’ reading of PP. I argue that the conservative commitments of PP can be divided into three distinct categories: (1) representational, (2) inferentialism, and (3) internalism.I show how these commitments and their relations should be understoodand argue for an interpretation of each that is both non-trivial and largely ecumenical towards the 4E literature. Conservative PP is as progressive as conservatism gets.

• Details
• Contents