ABSTRACT
Stochastic models are serve a very useful purpose, but they are difficult to test physiologically, since they do not generate predictions about neural dynamics, that is, about the spatio-temporally organized patterns of activity recorded in brains, which evidently underlie signal transmission. To serve this purpose, theories are required which emphasize the temporal aspects rather than the purely combinatorial aspects of the electrical activity of cells and populations. At the cellular level, characteristic discharge patterns are enormously variable, especially in unanesthetized animals. Large stellate cells in the thalamus are supposed to excite pyramidal cells in the neocortex. To construct a model which will do justice to such data authors start with a simplified electrical model of the cell membrane which goes back almost to the beginning of the century. But of course such a model neglects many of the details of cellular morphology and physiology, but it does provide some insight into the nature of neural variability.
