| 研究実績の概要 |
This project proposes a category theory approach to the (co-)development of (symbol-like) higher cognition in relation to nonsymbolic (spatial) abilities. The classical view supposedly accounts for higher cognitive abilities as a Language of Thought (LoT), i.e. thought processes are language-like, and recent proponents of this view point to LoT-like properties in nonlinguistic (spatial) domains to claim that LoT is the “best game in town” for explanatory coverage. However, the LoT hypothesis does not explain how/why symbol-like cognitive representations connect to other (non-symbolic) formats. The main achievement for this financial year was to provide a category/topos theoretic approach as a formal framework for bringing together symbolic and nonsymbolic (spatial) cognition. This work was presented at the 45th Annual Meeting of the Cognitive Science Society, Sydney, Australia (Phillips, 2023) and published in the international academic journal, Frontiers in Psychology (Phillips, 2024).
|
| 現在までの達成度 (区分) |
現在までの達成度 (区分)
2: おおむね順調に進展している
理由
Topos theory brings together algebra, topology and logic, cf. language, perception and reasoning. This approach is supposed to bridge the gap between symbol-like and spatial cognition as a duality, in a category theory sense, between perception and conception. Various properties of spatial cognition that provide further support for LoTs (e.g., discrete constituents, role-filler bindings and logical connectives) follow from categorical universal constructions. Together, these constructions constitute a special kind of category, called a topos, and every topos has an interpretation in first-order logic, hence the apparent pervasiveness of logic across age-groups and species.
|
| 今後の研究の推進方策 |
The plan for the next financial year is to extend this approach to incorporate resources and goals (intentions) thereby also explaining why there are systematic failures to induce higher cognitive abilities, such as exhibited in the Relational Schema Induction paradigm. An early version of this work was submitted for presentation at the 46th Annual Meeting of the Cognitive Science Society, Rotterdam, The Netherlands.
|
