EEG: Origin and Measurement

Abstract

The existence of the electrical activity of the brain (i.e. the electroencephalogram or EEG) was discovered more than a century ago by Caton. After the demonstration that the EEG could be recorded from the human scalp by Berger in the 1920s, it made a slow start before it became accepted as a method of analysis of brain functions in health and disease. It is interesting to note that this acceptance came only after the demonstration by Adrian and Mathews (J Physiol 81:440–471, 1934) that the EEG, namely, the alpha rhythm, was likely generated in the occipital lobes in man and was not artefactual. However, the neuronal sources of the alpha rhythm remained undefined until the 1970s, when we demonstrated, in dog, that the alpha rhythm is generated by a dipole layer centred in layers IV and V of the visual cortex (Lopes da Silva and Storm van Leeuwen, Neurosci Lett 6:237–241, 1977). It may be not surprising that the mechanisms of generation and the functional significance of the EEG remained controversial for a relatively long time considering the complexity of the underlying systems of neuronal generators, on the one hand, and the rather involved transfer of signals from the cortical surface to the scalp due to the topological and electrical properties of the volume conductor (brain, cerebrospinal fluid, skull, scalp) on the other. The EEG consists of the summed electrical activities of populations of neurons, with a modest contribution from glial cells. The neurons are excitable cells with characteristic intrinsic electrical properties, and their activity produces electrical and magnetic fields. These fields may be recorded by means of electrodes at a short distance from the sources (the local EEG or local field potentials, LFPs) or from the cortical surface (the electrocorticogram or ECoG), or at longer distances, even from the scalp (i.e. the EEG, in the most common sense). The associated MEG is usually recorded via sensors, which are placed at short distances around the scalp and are highly sensitive to changes of the very weak neuronal magnetic fields.