The Role of Sensory and Motor Information in Semantic Representation: A Review

Publisher Summary Embodied theories of cognition propose that simulation is the basis for cognitive representation. Simulation is assumed to use the same sensory–motor systems that are engaged during real experience; when this principle is applied to the representation of linguistic meaning (semantics) theories propose that semantic content is achieved by recreating, usually in weaker form, the sensory and motor information produced when the referent of a word or sentence is actually experienced. Simulations are content-specific; for example, words referring to motion, such as rise and fall, are thought to recruit sensory systems involved in perceiving motion, and words referring to motor actions, such as kick and walk, are thought to recruit the motor systems used for those actions. Therefore, embodied theories of semantic representation focus on semantic content, rather than the structure of the semantic system as a whole, how words are related to one another, or how categories are represented. At a basic level, embodiment extends the non-controversial idea that we learn from experience, so semantics must be grounded in our sensations and actions; however, it appears to be a departure from the commonly accepted view that semantic content is amodal and thus not dependent on sensory–motor information. This chapter reviews different theories of semantic representation placing them on a continuum as regards to their proposals about the role of sensory and motor information. A brief review of behavioral and neuroscientific evidence is then presented and the chapter ends with a discussion of what the implications are for semantic representation.