That the central nervous system is capable of accurate encoding of time is obvious, as numerous human and animal experiments can demonstrate. But it is not at all clear how the phenomena of storage and processing of temporal information are achieved; it is in fact rather difficult to see how neurons can operate accurately over time scales of seconds or minutes, which must be required for controlling everyday behaviours. In this chapter I want to first describe what I consider to be the main difficulties in the neural encoding of time, then list some of the ways in which neurons, as we currently understand them, might actually be used to encode temporal information. Two models will then be described in more detail, that could be used for timing. Both are 'network' models in which a large population of neurons combines to encode a temporal interval. I will also briefly and selectively review some artificial neural network models that deal will time. The ideas presented here have been tested mainly through computer simulations, and so it remains to discover which, if any, of these methods are used in biological systems.
ASJC Scopus subject areas