Electroencephalography and High-Density Electrophysiological Source Localization

In 1924, Hans Berger, a German psychiatrist, performed the first electroencephalographic (EEG) recording in humans (Berger, 1929), a discovery that was initially greeted with great skepticism by the scientific community. By recording from one electrode placed over the forehead and one over the occipital cortex, Berger discovered the existence of rhythmic activity oscillating at approximately 10 Hz, particularly during relaxed wakefulness and in the absence of sensory stimulation or mental activity. In this landmark discovery, Berger described for the first time what would become known as alpha waves. As a result, Berger was among the first to suggest that the periodic fluctuations of the human EEG may be associated with mental processes, including arousal, memory, and consciousness. Over the years, developments in data collection and analyses transformed EEG into one of the prime techniques for studying the human brain. Table 3.1 summarizes selected landmark discoveries and developments that have contributed shaping the EEG field throughout the century. The past two decades in particular have witnessed unparalleled progress in our ability to image human brain function noninvasively. Different imaging techniques are currently available to investigate brain function based on hemodynamic (functional magnetic resonance imaging, fMRI), metabolic (positron emission tomography, PET), or electromagnetic (electroencephalography, EEG; magnetoencephalography, MEG) measurements. In order to investigate spatiotemporal dynamics of brain activity, methods that directly assess neural activity are required. By measuring electrical activity of neuronal assemblies with millisecond temporal resolution, EEG and MEG, unlike hemodynamic techniques, offer the possibility of studying brain function in real time. Unfortunately, as will be discussed in this chapter, the spatial resolution afforded by EEG/MEG is constrained by several factors. The most important of these factors are the distorting effects of the head volume conductor,1 low signal-to-noise