Response to Glenberg: Conceptual content does not constrain the representational format of concepts.

Replies to comments by A. M. Glenberg (see record 2015-22897-006) on the author's original article (see record 2015-22897-004) on the embodied cognition debate. The core issue at stake in the debate is whether the format of thought is amodal or modality-specific. In his paper, Glenberg argues that "embodied systems do just fine accounting for perception, action, concrete cognition, and abstract cognition" and goes on to suggest that "Now the onus is on traditional cognitive scientists, those who wish to maintain a Cartesian distinction between human thought and action, a cherished and seemingly obvious belief, but ultimately, a type of flat-world hypothesis." But, is it really a "type of flat-world hypothesis" to maintain that cognitive processes can be amodal in their representational format? Mahon does not believe that it is. There is a reasonable theoretical alternative to the embodied cognition hypothesis that can account for the extant data, and which does not require adopting the view that "cognition is sensorimotor processing." That theoretical alternative (a) maintains a strict representational distinction between amodal concepts and sensorimotor systems and (b) argues that sensorimotor activation during conceptual processing reflects the structure and dynamics of connectivity between amodal representations and sensorimotor systems. The types of findings that Glenberg (2015) cites as support for embodied cognition do not distinguish between the embodied cognition hypothesis and this theoretical alternative. Furthermore, neuropsychological data indicate that sensorimotor impairments can occur without concomitant conceptual level deficits.