Synchronization between Cortical Regions during Language Processing

This project is aimed at examining the functional role of EEG coherence within different frequency bands during language and memory processes. Knowledge about activities within certain frequency bands as well as information about cortical co-operation is absolutely necessary for an adequate characterisation of brain functioning. In studying functional relationships, coherence analyses yield important new aspects of brain activities which complement the data obtained by power- or amplitude analyses. Experiments are designed based on findings obtained in project P11572-MED. Essentially, this project will be a continuation of the previous studies. One series of experiments addresses the question whether linguistic categories have a biological reality and how these categories are functionally represented within the brain. In recent years several studies support the notion of a, different functional representation of lexical categories within the brain. The division into lexical categories like e.g. concrete and abstract nouns and even subclasses of concrete nouns. like common nouns and proper names was made introspectively by linguists and philosophers about 2000 years ago. Since then these classifications remained unchanged. By computation of EEG coherence cortical co-operation and the participation of different neural networks accompanying processing of certain categories shall be monitored. The second series of experiments deals with the role of coherence during processing congruous and incongruous German sentences. In close co-operation with Dr. M. Kutas (University of California, San Diego) and DDr. H.M. Müller (University of Bielefeld) English relative clauses should be analysed with spectralanalytical methods in order to obtain new results in the frequency domain in addition to existing time domain analyses earlier performed by Kutas and colleagues. Furthermore, MEG coherence and EEG coherence of simultaneously recorded signals during language processing should be compared. Recordings were already made within the frame of a co-operation with the Institute of Medical Statistics, Computer Science and Documentation and the Biomagnetic Centre Jena, both University of Jena, Germany. Main emphasis is laid on the analysis of time relations, i.e. the direction of information transfer, between different recording sites during language processing. In close co-operation with Dr. B. Schack (University of Jena, Germany) coherence and phase should be analysed using adaptive bivariate autoregressive moving average (ARMA) models. The results shall be compared with the results of conventional Fourier analyses.