The Implications of Embodiment for Cognitive Theories

This paper discusses epistemological aspects of the new eld of embodied Artiicial Intelligence and its consequences for the study of cognition. It is argued that the new emphasis on bodily phenomena fundamentally changes the nature of cognitive theories and the way these theories are formed. This result is based on an analysis of conventional Cognitive Science methodology as it has been succinctly described by Herbert Simon. It is argued that the inherently dynamic and physical nature of embodied AI can serve to correct oversimplifying assumptions of Cognitive Science methodology. Cognitive Science is proud of its role as a prototypical interdisciplinary science. Its commentators and epistemologists regard it as a noteworthy example for overcoming the diversiication of science in order to study a problem that has puzzled (at least some) humans for more than 2500 years: cognition. The integration of areas as diverse as philosophy, psychology, computer science, artiicial intelligence, linguistics and medicine arose from the unifying power of computational models. Today, the study of the phenomena of cognition by means of computer models that are designed, developed, implemented and tested by people who are not computer scientists has become part of an accepted and widely propagated methodology. However, recent developments in Artiicial Intelligence, which is often regarded as Cognitive Science's more technical counterpart, suggest that this approach to modeling cognition may be constructed on weak grounds. What has nowadays become known as \embodied AI" challenges a sizeable number of the foundations on which the methods of Cognitive Science were built. In order to 1