A recent breakthrough in understanding brain-language mechanisms is the discovery of local motor cortex activations that index specific meaning features of words, phrases and sentences. The words "talk" and "walk" activate different parts of the motor cortex, reflecting the body part relationship of actions the linguistic items are typically used to speak about. It has been suggested that such semantic motor mapping can be explained by behaviorist theories, based on conditioning mechanisms also effective in Pavlov's dog when it salivates to bell sounds. In contrast, a neurobiological approach to language predicts modulation of semantic activation by grammatical, including inflectional-morphological, information. Here, we test these competing predictions by putting action words into different phrasal contexts invoking morphosyntactic and morphophonological processes and demonstrate that semantic motor mappings are modulated by grammatical sentence properties, especially the presence of overtly realized inflectional affixes on nouns or verbs embedded in grammatical phrases. Mechanistic neuroscience theories taking into account both meaning and grammar, including morphology and syntax, are required to explain these observations. A direct comparison between phrases containing nouns and verbs revealed a tendency towards greater activation to noun phrases in left-inferior premotor cortex and posterior Broca's region (BA 44), thus questioning previous suggestions that left inferiorfrontal areas might be dedicated to verb processing per se.
Copyright © 2011. Published by Elsevier Inc.