Miller, G.F., and Todd, P.M. (1990). Exploring adaptive agency I: Theory and methods for simulating the evolution of learning. In D.S. Touretzky, J.L. Elman, T.J. Sejnowski, and G.E. Hinton (Eds.), Proceedings of the 1990 Connectionist Models Summer School (pp. 65-80). San Mateo, CA: Morgan Kaufmann.

Abstract

Psychology construed as the scientific study of adaptive agency can include not only modelling of specific psychological adaptations in particular species, but general exploration of the adaptive processes (including evolution, learning, and computation) that build, modify, and instantiate those adaptations. Connectionist theory has concentrated on understanding the adaptive processes of learning and computation, and has assumed general-purpose learning principles as the prime constructors of psychological adaptations. But connectionism has thereby ignored the central lesson of a century of learning theory in psychology: learning mechanisms must be understood in terms of their specific adaptive functions, just like other psychological adaptations. This paper introduces the notion of psychology as the study of adaptive agency, outlines a hierarchy of adaptive processes underlying adaptive agency, and reviews the history of learning theory and the emergence of ecological and evolutionary approaches to learning. We then develop a taxonomy of adaptive functions that learning mechanisms might serve, and outline a general simulation framework for exploring those adaptive functions. Finally, we present empirical results concerning the simulated evolution of associative learning.

See my other publications on Evolution of learning   Animal behavior   Evolutionary and adaptive simulation   



   Back to main page.
   View publications categorized by topic.
   View publication abstracts searchable by keywords