ABSTRACT
This chapter describes a preliminary version of the “single channel wavelet model,” and its application to cat auditory evoked potentials. Event-related potentials (ERPs) are the brain electrical potentials associated with sensory and cognitive processing. Analysis of ERP data requires separation of the recorded potentials into component waveforms and estimation of the effects of experimental conditions and disease states on the components. The chapter discusses a preliminary version of the “topographic wavelet model,” which generalizes the single channel wavelet model to multichannel data. There are two conventional approaches to decomposition and statistical analysis of ERPs. These are: frequency domain filtering followed by identification of peaks and statistical analysis of peak amplitudes, and principal component analysis (PCA) with varimax rotation followed by statistical analysis of the factor scores. The chapter explains a wavelet approach that allows components to overlap in time and frequency while avoiding the unrealistic mathematical constraints of PCA.
