Space-time transformation in the frog cerebellum through an intrinsic tapped delay-line.

Abstract

THE cerebellum is interesting for the arrangement of its neurones in a symmetrical three-dimensional lattice, with only two classes of input fibres and one of output fibres1. This arrangement led Braitenberg2 to postulate a set of logic operations that the cerebellar cortex might perform. One postulate was that, by virtue of the distribution of the “output” neurones (Purkinje cells) at regularly spaced intervals along beams of parallel fibres (Fig. 1A), the cerebellum might function as a type of clock, transforming an event that occurred at some peripheral spatial location into a temporal event. A central assumption of this hypothesis is that the time of firing of a Purkinje cell is determined by the conduction time of afferent impulses travelling in parallel fibres from some reference point on the beam of parallel fibres on which it lies to the Purkinje cell. If this assumption is correct the parallel fibre–Purkinje cell system would constitute a tapped delay-line2–4. If the reference point is the position of another Purkinje cell along the beam, then a corollary is that the time between the firing of the two cells must be equal to the conduction time of the segment of the beam connecting them.