ABSTRACT
One of the most pervasive enigmas regarding brain function is how infonnation becomes transmitted from one location to another. In view of the fact that nerve impulses are per se more or less all or none in character, attention has been focused on the pattern of interspike, i.e. inter impulse, intervals as the carriers of information. A good deal of evidence has accrued to the effect that in an anesthetized unstimulated brain, the interspike intervals recorded from single neurons in a variety of locations are essentially random in their distribution. Models of the interspike process have therefore been constructed upon the assumption that this randomness reflects a stochastic process. Stochastic resonance models [1, 2, 3] and stochastic resonance with noise models [4] have been especially fruitful in simulating actual data obtained from spike trains. However recording randomness does not in itself insure that a process is stochastic. Recently a surge of interest has developed for the possibility that the behavior of spike trains, though random, could be generated by a detenninistic nonlinear process which results in chaos. The current project sets out to test whether deterministically chaotic or stochastic processes best characterizes the patterns
of interspike intervals recorded from hippocampus and from somatosensory cortex in the lightly anesthetized rat.
