The hippocampus is one of the most seizure-prone regions of the central nervous system. Indeed, even in hippocampal slices, epileptiform activity can be easily induced by pharmacological blockade of GABAA receptors. The resulting activity consists of one or more synchronized population bursts that are similar to interictal events recorded from the intact epileptic brain. Detailed information about hippocampal pyramidal cell properties and their connectivities has made it possible to model this epileptiform behavior and thus provide insights into the special characteristics of hippocampus that render it so excitable. In this short review, we concentrate on the synchronization of individual neural elements in the CA3 region. We show how synchrony can be initiated by a process analogous to a chain reaction. A series of synchronized burstsanalagous to a brief seizure-arises when sustained excitatory dendritic currents (as can be mediated by NMDA receptors) interact with intrinsic membrane currents to produce a train of dendritic bursts. This account predicts that, in a large ensemble of neurons with spatially restricted connectivity (as in the longitudinal CA3 slice), later bursts should have different propagation properties than the initial burst. This prediction appears to be correct.
It is a truism of statistical physics that deep and complex phenomena can arise through simple interactions between sufficiently many simple objects. The brain exhibits complex phenomena involving non-trivial interactions between many rather complicated objects, the neurons. Such complex interactions are particularly characteristic of epileptic events, even in in vitro systems containing only thousands of cells. Because of this complexity, the computer has played an important role in linking observed properties of cell membranes and synaptic receptors to emergent properties in the population. We shall illustrate this role in two types of epileptic events induced by GABA, blockade. Simulation and experimental results are presented together; the two are conceptually linked and both are directed toward the same goal-understanding normal and abnormal brain function. Some of these results are discussed in greater depth elsewhere (Traub and Miles, 1991;.
Epilepsy was believed by Hughlings Jackson (1890) to depend, at least in part, upon abnormally increased synchrony in populations of neurons.