The coupling-constitution fallacy revisited

The hypothesis of extended cognition maintains that cognitive processes sometimes span the brain, body, and world. One of the most prominent types of arguments for this hypothesis begins with observations of the role of certain sorts of bodily and environmental influences on cognitive processes, then concludes that these bodily and environmental influences are themselves cognitive processes. This paper will argue that, while it is widely appreciated that the simplest versions of these “coupling arguments” are fallacious, the advocates of extended cognition still seem to underestimate how prevalent even the simplest forms are in the literature, why they are so pervasive, and how these fallacious arguments might be avoided by greater attention to a plausible theory of what distinguishes cognitive processes from non-cognitive processes.