Cognitive Science

One definition of Cognitive Science is "the study of intelligence and intelligent systems, with particular reference to intelligent behavior as computation"[1] Much effort by Cognitive Science researchers today emphasize that intelligence must be explained, and perhaps only understood, from the prospective of evolution. Approaching the study of intelligence, researchers and developers are using evolutionary computation (EC) to model intelligent systems. EC generated entities can often result in non-intuitive functional decompositions, but nevertheless, can be quite adaptive to their specific environments. Intelligence emerges from the integrated entity-environment meta-system.

EC is a class of general-purpose problem solving algorithms and techniques inspired by biological evolution. Types of problems that have been solved include, but are certainly not limited to, Genetic Programming (GP) to discover Artificial Intelligent (AI) programs and Genetic Algorithms (GA) to evolve artificial Neural Networks (NN).

An example use of EC is as a neural biological research tool. Using EC algorithms, computational theories of the mind may be refined by evolving (searching) existing neural or functional models for best solutions. Moreover, new and novel functional models may be generated from "tabula rasa" neural network architectures (unspecialized networks). EC search spaces may be constrained by observational data obtained from neuroimaging techniques or experimental psychology. Resultant evolved models may be validated against observation. EC genetic encodings of neural organization may themselves be evolved to new genotype encodings such as neural development (ontogeny) and synaptic plasticity during learning.

[1] Attributed to H. A. Simon, and C. A. Kaplan C. A., "Foundations of cognitive science", in Posner, M.I. (ed.) 1989, Foundations of Cognitive Science, MIT Press, Cambridge MA. www.psych.ualberta.ca