Classification of the EEG feature components

In this paper we propose a novel method for the EEG signal processing based on the classification of independent components of the signal features. ICA algorithm has been successfully applied to the area of EEG artefact detection, however this algorithm can be applied to identify independent components of signal features. We decomposed the EEG signal to the descriptive features, calculated independent components for specific features, linked them to the appropriate electrodes and classified these feature components by several algorithms. This method was applied to the data from a psychological experiment focused on the adoption of a specific frame of reference within the spatial navigation. The results were compared with the widely adopted method of signal feature classification. The feature components method revealed the brain structures involved in the spatial navigation similar to the results of recent EEG and fMRI studies.