Analysis of event related potentials using PCA and matching pursuit on the time-frequency plane

Joint time-frequency representations offer a rich representation of event related potentials (ERPs) that cannot be obtained through individual time or frequency domain analysis. This rich representation, however, comes at the expense of increased data volume and the difficulty of interpreting the resulting representations. Therefore, methods that can reduce the large amount of time-frequency data to physiologically meaningful components are essential. The method presented in this paper extends principal component analysis to the time-frequency plane to reduce a large set of ERPs to a small number of significant components. These components are then characterized using a Gabor dictionary to offer a succinct parametrization of the ERP data. The results show that the principal component analysis is successful at extracting components that can be described as the superposition of a small number of Gabor logons, and that the resulting set of logons succinctly represent physiologically meaningful ERP events.