Distinctively human thinking in a massively modular mind

This talk will take up one of the main challenges to massively modular models of the human mind: to explain the distinctively flexible and creative character of human cognitive processes. The goal is to sketch how one might build a mind with such properties out of modular components (where these components are ‘modular’ in the sense to be explained in my Wednesday talk). The model proposed finds an important place for natural language in realizing distinctively-human thought processes (an idea which is consistent with ‘dual process’ models of human reasoning proposed by Stanovich and others). Language may be able to combine contents that wouldn’t otherwise get conjoined, and then cycles of linguistic activity in ‘inner speech’ may recruit a variety of other inferential systems, and other capacities, to generate novel contents, as well as to acquire novel inference-types (such ascanons of scientific method). Some of these capacities – like mental rehearsal of action – are likely to be shared with other animals, whereas others may be uniquely human. In the latter category might fall human meta-representational capacities and the human system for normative reasoning and normative motivation.