Abstract

Current research into the neurobiological foundations of human language suggests that language function in the modern human is mediated by a coalition of two overlapping systems. A distributed bihemispheric system, largely shared with our primate relatives, provides a social and interpretative framework for language comprehension, as well as basic mechanisms for mapping sounds onto lexical meanings. A specialized left hemisphere system, possibly unique to humans, supports core combinatorial functions underpinning morphosyntax. In recent neuroimaging research we have begun to investigate how different types of morphological process (broadly defined as inflectional and derivational) interface with these two systems, and whether this differs across languages.

Working in English, Polish, and Arabic, we see a common pattern whereby concatenative inflectional complexity selectively activates the left hemisphere system, but where these languages diverge for derivational complexity. For both Polish and English, derived forms activate the bihemispheric system but not the left hemisphere system. For Arabic, where the non-concatenative word pattern morpheme arguably plays both derivational and inflectional roles, we see selective left hemisphere engagement. This suggests that left hemisphere systems are specifically engaged when language inputs are decompositionally analysed and represented, and that this cuts across conventional distinctions between inflectional and derivational linguistic functions.