Toward a neurolexicology

Abstract

This article analyzes the organization of the mental lexicon based on neurophysiological data. The neuroscience literature has devoted many studies to the semantic processing of words. However, the research remains specific to certain categories, studied separately, and does not address the lexicon as a system. In order to provide further insight into the neuronal organization of the lexicon, we conducted an EEG-based semantic decision experiment using words from eight categories (four living and four nonliving categories) as the material. A data-analysis method (correspondence analysis or CA) commonly used in computational linguistics was applied to the electrophysiological signals. The results revealed a two-factor structure: an ontological organization separating the living from the nonliving, and an organization with a human referential structured by proximity to the person. A comparison of the ERP-CA and the linguistic-CA data revealed organizational analogies. Lastly, a tomography software (Loreta®) was used to estimate the cerebral sources associated with the ERP signals.