Koester, D. 1, 2
1 Faculty of Psychology and Sport Science. Bielefeld University. Bielefeld, Germany
2 Center of Excellence - "Cognitive Interaction Technology." Bielefeld, Germany
A long-standing debate in morphology revolves around the storage format of compounds; whether and, if so, how the brain represents internal word structure or whether it stores compounds as unitary elements. Other issues complicate the answer to this question such as the different modes of information reception in the visual and acoustic domain (more holistic vs. serial) or the divergent direction of information transmission in comprehension and language production. Furthermore, it is of interest whether morphological processes are qualitatively comparable for native speakers and (late) second language learners. In this talk I will present recent evidence from event-related potentials (ERPs), functional magnetic resonance imaging (fMRI) and behavioural studies to support the (de)composition view on compound processing in the domain of comprehension and production. (Under specific circumstances, it appears that compounds can be stored holistically.) In particular, ERP data suggest that compounds are decomposed morphosyntactically in auditory comprehension and, consistently, that compound constituents are semantically integrated in an incremental fashion. In contrast to comprehension, language production is more difficult to investigate with ERPs due to movement artefacts. However, when investigating the processing of internal word structure during overt language production, we could provide direct neurophysiological evidence for the time course of morphological encoding according to the Indefrey & Levelt model of speaking. fMRI data obtained with a similar paradigm point to the left inferior frontal gyrus (BA 47) as the associated functional neural correlate. Finally, I will report behavioural evidence to suggest that there are qualitative differences during compound reading between native speakers and second language learners regarding the decomposition of compounds depending on the compound word length. These findings may help to sketch the neural basis of compounding, an important mechanism in human verbal communication.