Interpersonal action semantics Comment on 'Action semantics: A unifying conceptual framework for the selective use of multimodal and modality-specific object knowledge' by van Elk, van Schie, & Bekkering

摘要

Van Elk et al. [1] propose a model of action semantics that covers a broad swath of tool-using behavior including developmental and neurophysiological perspectives. Missing, however, is a discussion of learning: How do people learn to use tools, and, how does that learning shape the learner's body schema? By body schema, we mean the neurophysiological representation of the limbs, their locations, and their abilities. Maravita and Iriki [2] (see also [3]) review the fascinating literature on these topics, and suggest a radical hypothesis. Primates are skilled tool users because of the remarkable ability to shape their body schemas to incorporate tools. Neurons in the macaque parietal cortex have bi-modal receptive fields: the same neuron responds to somatosensory stimuli at the hand and visual stimuli near the hand. After using the tool, the visual receptive field migrates toward the effector-end of the tool (e.g., the tines of a rake). Thus, the neuron provides a mechanism for meshing the feel and sight of the tool for accurate wielding. In effect, the animal has incorporated the tool into its body schema in that looking at the tool excites a neuron with a somatosensory receptive field.