A distributed framework for semi-automatically developing architectures of brain and mind

Abstract

Developing comprehensive theories of low-level neuronal brain processes and high-level cognitive behaviours, as well as integrating them, is an ambitious challenge that requires new conceptual, computational, and empirical tools. Given the complexities of these theories, they will almost certainly be expressed as computational systems. Here, we propose to use recent developments in grid technology to develop a system of evolutionary scientific discovery, which will (a) enable empirical researchers to make their data widely available for use in developing and testing theories, and (b) enable theorists to semi-automatically develop computational theories. We illustrate these ideas with a case study taken from the domain of categorisation.