The Bodily Formats Approach to Embodied Cognition

Chapter Overview In the past few decades, many practitioners of cognitive science and philosophy of mind have staked out programs and positions under the label of " embodied cognition " (EC). They have widely differing views, however, of what embodiment consists in and why a program of embodied cognition might be an improvement over classical cognitivism. Frederique de Vignemont and I have recently proposed a general characterization of embodied cognition (Goldman & Vignemont, 2009), and I have expanded on the ramifications of this conception when one adds to it evolutionary considerations and a certain attractive conception of neuroarchitecture (Goldman, 2012). On this occasion, I review the definitional proposal in the context of other conceptions of embodiment and show how an assortment of empirical evidence lends strength to our proposal. Finally, I reply to Lawrence Shapiro (this volume, Chapter 3), who raises a series of challenges for our conception. Existing proposals for EC can be divided into two general categories: proposals predominantly derived from computer science, artificial intelligence (AI), and robotics; and proposals predominantly derived from cognitive psychology and cognitive neuroscience. In the first section, I look at samples of both kinds of proposals. The second section then reviews the definition of embodiment we have proposed and how extensive the range of embodiment is likely to be given the empirical findings that have already emerged. Finally, in the third section, I reply to Shapiro's comments.