(B. Schack, Jena)
 

The aim of the present proposal is the development and the application of multivariate spectral analysis methods. The field of application is the investigation of high-dimensional electroencephalograms and magnetoencephalograms (EEG/MEG) of cognitive processes. Due to the importance of interactions between different areas of the brain the proposal is focussed on the further development of methods for detecting and describing synchronisation processes. Because of the high dynamic of cognitive processes the stationarity of the signals cannot be presumed. The methods to be developed  have to yield in time-variant estimations. From the methodical point of view there are three main focuses.

• Development of estimation methods for adaptive matrix coherences for the quantification of complex synchronisation processes with more than two components.
• Construction of adaptive methods for the characterisation of causal relationships on the basis of local-stationary AR-modelling.
• Development of dynamic parameters for the description of complex phase synchronisation on the basis of signal expansion.

• B. Schack, W. Klimesch, Frequency characteristics of evoked and oscillatory electroencephalic activity in human memory scanning task, Neuroscience Letters 331, 107-110 (2002).
• B. Schack, W. Klimesch, The use of time-variant EEG Granger-Causality for inspecting directed interdependencies of neural assemblies, to appear in Journal of Neuroscience Methods ( text and figures).
• W. Hesse, E. Moeller, M. Arnold, H. Witte, B. Schack, Untersuchung kurzzeitiger kausaler Beziehungen im EEG auf der Basis der adaptiven Granger Kausalitaet, Biomedizinische Technik 47, Ergaenzungsband 1 (Teil 2), 510-513 (2002).