Our computational nature: comment on Barrett et al.

I argue that Barrett et al. (2014) have misinterpreted evolutionary psychologists’ notion of computation. Barrett et al. seemingly presume that the notion of computation deployed by evolutionary psychologists (e.g., Pinker, 1997; Tooby and Cosmides, 2005) is tantamount to positing a physical architecture whose form of computation proceeds via the syntactic-like transformations of spatially discrete representational symbols or sentence-like structures—i.e., in the manner of a Turing machine. But this is simply not the notion of computation that evolutionary psychologists advocate (in spite of the fact that a Turing machine architecture, for instance, is nonetheless compatible with it). The notion of computation is philosophically complex, with many different meanings and amultifaceted history (Piccinini, 2012). By “computation,” evolutionary psychologists fundamentally mean to say that the brain evolved to compute in the generic sense of the term. This more generic notion of computation, and its relation to a physical substrate, is outlined by Pinker (2005):