New horizons: Dynamical systems and situated cognition

Abstract

Overview Throughout this book we have been working through some of the basic consequences of a single principle. This is the principle that cognition is information processing. It is in many ways the most important framework assumption of cognitive science. The historical overview in Part I explored how researchers from a number of different disciplines converged on the information-processing model of cognition in the middle of the twentieth century. In Part III we looked at different ways of thinking about information processing – the physical symbol system hypothesis and the neural networks model. Despite their very significant differences, the physical symbol system and neural network approaches share a fundamental commitment to the idea that cognitive activity is essentially a matter of transforming representational states that carry information about the agent and about the environment. In this chapter we turn to some of the new horizons opened up by two different ways of modeling cognitive abilities. Sections 13.1 and 13.2 explore how some cognitive scientists have proposed using the mathematical and conceptual tools of dynamical systems theory to model cognitive skills and abilities. One of the particular strengths of dynamical systems theory is the time-sensitivity that it offers. Dynamical models can be used to plot how a system evolves over time as a function of changes in a small number of system variables.