Klinische Neurophysiologie 2010; 41 - ID101
DOI: 10.1055/s-0030-1250930

Analysis of EEG data on complex human thinking

K Klawitter 1, M Öllinger 2, A Faber 2, 3, T Filk 2, 4, J Timmer 1, 4, 5, 6, B Schelter 1, 4, 5
  • 1Universität Freiburg, Freiburger Zentrum für Datenanalyse und Modellbildung (FDM), Freiburg, Deutschland
  • 2Parmenides Center for the Study of Thinking, Munich, Deutschland
  • 3LMU München, Graduate School of Systemic Neurosciences, München, Deutschland
  • 4Universität Freiburg, Department of Physics, Freiburg, Deutschland
  • 5Universität Freiburg, Bernstein Center for Computational Neuroscience, Freiburg, Deutschland
  • 6Universität Freiburg, Freiburg Institute for Advanced Studies, Freiburg, Deutschland

Introduction: Studying higher cognitive functions of he human brain poses a challenge not only to neurosciences, but also to the methods of statistical data processing and analysis. As an example for the investigation of higher cognitive functions this study investigates methods of analysis of electroencephalographic (EEG) data that was collected in experiments on the visual perception of causal events and on a better understanding of cognitive remodeling. The enhanced analytic methods are expected to help identifying the participating cognitive processes and to support their modelling.

Methods: Several techniques are used to analyse EEG-data which was recorded using a high-resolution EEG system of 64 channels at a sampling frequency of 1000Hz. First of all, normal event related potentials (ERP) were averaged to identify active parts of the brain. In this context, taking a look at the current source density can produce important findings. Mean phase coherence, which can detect even weak coupling of activity between different parts of the brain, was determined as a measure of interaction. Moreover, time-frequency analyses were used to detect sudden changes in the EEG signal during re-structuring tasks. This may help to overcome the difficulties of identifying a precise event, which would be essential for the calculation of ERPs.

Results: Brain areas that are involved in the perception of perceptual causality showed lateralized activity. New insights are gained using network analyses. Thereby, interactions of involved structures can be identified. In combination with biomedical and insight psychological interpretations an enhanced understanding of the processes of human thinking results. Challenges in analysing EEG data during cognitive restructuring are demonstrated and possible further options of analysis are discussed.

Conclusion: This project presents methods of analysis that aim to create neuronal network maps. This provides an opportunity to gain new insights into the processing of complex operations in human brain.