EEG–fMRI in Animal Models

Abstract

Neuroscientists have long sought techniques for investigating the neuronal mechanisms of normal behaviour and disease. The uniquely enigmatic nature of the brain and the difficulties inherent to its study limited early physiological investigations of brain function. For example, although able to provide great insight into the localisation of brain function, lesion studies are inherently destructive, and thus reveal limited information about brain functioning in situ. Techniques for the noninvasive monitoring of brain function were needed. The advent of recording electrical brain activity via electroencephalography (EEG) opened up new avenues for the noninvasive study of brain activity (Berger 1929). For many years, neuroimaging lagged behind electrophysiological techniques. However, early studies of cerebral haemodynamic responses showed that brain function could be related to measurements of blood flow. Seizures occurring during neurosurgery have long been known to produce a focal blood flow increase in the cerebral cortex (Horsley 1892; Penfield 1933), and early measurements using intracarotid sensors likewise demonstrated increased cerebral blood flow (CBF) during seizures (Gibbs et al. 1934). Advancements in electrical recording and functional imaging technology in recent decades have now made it possible to noninvasively study the brain at sufficiently high temporal and spatial resolutions to reveal fundamental neuronal processes in great detail. EEG measures extracellular electrical field potentials generated by populations of cortical neurons, and can capture brain electrical activity with excellent temporal resolution. Although EEG provides high temporal resolution, it is limited in its spatial sampling and cannot completely characterise neuronal activity throughout the entire brain. The spatial resolution of EEG is not sufficient to reveal the contribution of individual brain regions to neuronal function. The electrical signal recorded in the EEG reflects a spatial summation of the underlying cortical electrical activity, and does not sample subcortical areas; thus, EEG with scalp electrodes may not detect deeply originating ­discharges (Gloor 1985).