EEG Models and Analysis

Abstract

Continuous recording of neural tissue registers extracellular potentials known as electroencephalogram or EEG. This type of electrical activity accounts for spatiotemporal synchronization of neuronal assemblies in the vicinity of the recording electrode. EEG has been linked to complex processes in the brain, such as sleep, epilepsy, Parkinson's disease, memory, spatial navigation, and cognition, to name a few. Moreover, the number of applications involving neuronal oscillations has increased exponentially due to improved hardware, well-established recording paradigms, and better understanding of neural mechanisms. It is then necessary to analyze EEG in a principled framework with the proper rigor of statistical methods. The first part of this chapter details the neurophysiological basis of EEG-generating mechanisms at the cellular level, different types of recorded extracellular potentials (categorized by invasive level), and the well-established EEG rhythms alongside their main behavioral correlates. The second part of the chapter describes the statistical properties of EEG and how different methods address these constraints. The third section thoroughly details the mathematical models and analysis tools most commonly used to represent the spectral-temporal-spatial structure of EEG. Lastly, the chapter ends with a brief summary, discussion, and an outline of future directions. Overall, this chapter is intended as a literature review of the statistical treatment commonly applied to this signature brain activity. PU - SPRINGER-VERLAG SINGAPORE PTE LTD PI - SINGAPORE PA - 152 BEACH ROAD, #21-01/04 GATEWAY EAST, SINGAPORE, 189721, SINGAPORE