[PS-2.48] Neural correlates of auditory artificial grammar learning

Goranskaya, D. 1 , Kreitewolf, J. 1 , Mueller, J. 2 , Friederici, A. 1 & Hartwigsen, G. 1

1 Max Planck Institute for Human Cognitive and Brain Sciences, Department of Neuropsychology, Leipzig, Germany
2 Osnabrueck University, Institute of Cognitive Science, Osnabrueck, Germany

Neural sensitivity to regularities plays a crucial role in the acquisition of various linguistic aspects during auditory sequence processing. We investigated the neural correlates of learning regularities in auditorily presented syllable sequences using an artificial grammar (AG) learning paradigm. While previous neuroimaging studies selectively compared response differences in processing correct vs. incorrect AG sequences as the outcome of learning, we focused on the actual learning process. To explore the neural bases of learning and application of AG rules we employed an fMRI paradigm with alternating learning and test phases in two groups of adults: successful learners and non-learners. We expected that successful learning would modulate neural activity in premotor and parietal areas, associated with speech-motor and verbal working memory processes. Relative to non-learners, successful learners showed increased neural activity in a fronto-parietal network encompassing regions in the left lateral premotor cortex and bilateral superior-inferior parietal areas during both learning and rule application. Based on these activation patterns and the reported learning strategies, we conclude that successful learning and processing of AG rules is mediated via a fronto-parietal network for phonological processes. Specifically, phonological segmentation, phoneme matching, inner rehearsal, and verbal working memory are core processes for successful AG learning.