How a single Purkinje cell could learn the adaptive timing of the classically conditioned eye-blink response

Abstract

Experimental evidence supports the view that the cerebellum is involved in the adaptive timing of the classically conditioned eyeblink response. Previous modelling studies have demonstrated that a group of cerebellar Purkinje cells can learn the adaptive timing of the eye-blink response if the cells in the group have predefined response latencies which cover the range of conditionable interstimulus intervals (ISIS). Here we show how the timing can be learnt by a single Purkinje cell. Phosphorylation of metabotropic glutamate recptors (mGluRs) in our model causes the time delay between parallel fibre input and voltage response to be adaptive and makes it unnecessary to specify a conditionable ISI for each cell in advance. The model is able to learn conditioned responses (CRs) for delay conditioned ISIs between 200 and 1000 msec. Modification of parts of the intracellular signalling network might represent a general mechanism for neurons to learn the timing between input and output