Functional neuroimaging of morphological processing in nonnative speakers of English

The nature of morphological processing has been a focus of research in the cognitive neurosciences of language for decades, primarily because the systems underlying simple, word-level linguistic processes may also contribute to fundamental human cognitive capacities and brain functions (e.g., categorization, functional neural organization, and memory). To date, neuroimaging research has yet to demonstrate whether nonnative speakers of a language sort out and process morphologies in the same way that native speakers do. This study, therefore, is intended to identify neural mechanisms that have so far eluded detection. Using functional magnetic resonance imaging, this study adopts an event-related design to investigate the neural responses during English regular and irregular past tense verb generation by Korean nonnative speakers of English. A whole-brain analysis reveals that the processing of irregular verbs evoked greater neural activation than for regular verb processing, and distinctive regional differences of neural responses were found. Specifically, neural activation in regions of the middle and the superior temporal gyri in the right hemisphere was found to be relevant to regular past tense processing, whereas neural recruitment in regions of the inferior frontal gyrus, the supramarginal gyrus, the caudate, and the thalamus in the left hemisphere was found to be significant in irregular past tense processing in nonnative speakers of English. In particular, the results support claims for an inhibitory role of the caudate in prepotent responses and for a thalamic function controlling retrieval of specific items in language and memory. In addition, as reported in earlier studies with native speakers, the results showed increased activity in the frontal cortex and the cingulate cortex bilaterally during both regular and irregular past tense processing. In this study, however, the neural involvement of the cortices in both hemispheres was viewed as evidence for a more general cognitive control function induced by the experimental task rather than by their essential role in morphological processing, since the selective attention required for the rapid past tense generation task would itself entail such cognitive control. Taken together, the results shed further light on the cortical and subcortical representation of language in the human mind and brain.