Ina Bornkessel-Schlesewsky, 19 November 2013
Towards a neurobiologically and cross-linguistically plausible model of language processing
Institut für Germanistische Sprachwissenschaft, Philipps-Universität Marburg
This talk presents a new model of language processing that aspires towards neurobiological plausibility (Bornkessel-Schlesewsky & Schlesewsky, 2013). Based on earlier versions of the extended Argument Dependency Model (eADM), it combines insights on cross-linguistic diversity in language comprehension with design principles from neurobiology. Like other current models, the latest version of the eADM posits that auditory language processing proceeds along two distinct streams in the brain emanating from auditory cortex: the antero-ventral and postero-dorsal streams. In contrast to other models, however, the eADM assumes that both streams are organised hierarchically and information processing takes place in a cascaded fashion. Moreover, each stream has functionally unified computational properties congruent with its role in primate audition. While the dorsal stream performs time-dependent computations (i.e. sequence processing) in temporal receptive windows of increasing size, the ventral stream performs time-independent computations comprising the recognition of increasingly complex (i.e. feature-rich) auditory objects and their (commutative) combination. I will describe how these assumptions derive a range of existing neuroanatomical and neurophysiological findings and, in addition, allow for novel, testable predictions with regard to the neurobiology of language.
- Where and when:
15:45-17:00 Nov 19, 2013Donders Centre for Cognitive Neuroimaging (Colloquium Room), Kapittelweg 29, 6525 EN Nijmegen