Precise spike timing of tactile-evoked cerebellar golgi cell responses: a reflection of combined mossy fiber and parallel fiber activation?

Abstract

Introduction In recent years a number of papers emerged with tales of ventures in what J.I. Simpson (this Volume) named terra incognita in the cerebellar cortex: the gray zone (granular layer) between input (mossy fibers) and output (Purkinje cells). This gray zone holds feedforward and feedback circuits which pre-process afferent information before it reaches the Purkinje cells. Some of these recent publications specifically focused on the Golgi cell, the primary inhibitory interneuron of the granular layer circuitry (Dieu- donn6 1998a, b; Maex and De Schutter, 1998a, b; Watanabe et al., 1998; Vos et al., 1999a, b). They unequivocally concluded that Golgi cells do more than control the gain of parallel fiber input to Purkinje cells (see also De Schutter et al., Chapter 6, this Volume), which was suggested to be their only function in the classic cerebellar theories of Man (1969), Albus (1971) and Ito (1984). Figure 1 shows a schematic drawing of the anatomical connections of Golgi cells. Golgi cells receive direct excitatory input from mossy fibers (Hfimori and Szent~igothai, 1966). They are also contacted by parallel fibers (Fox et al., 1967), which excite Golgi cells after activation of granule *Corresponding author. Fax: +32-3-8202669; e-mail: erik@bbf.uia.ac.be cells by mossy fibers. In addition, Golgi cells do not contact each other (nor do granule cells) (Palay and Chan-Palay, 1974). Being inhibitory, Golgi cells exert both a feedback (Eccles et al., 1966b) and a feed-forward (Precht and Llin~is, 1969) inhibition of granule cell activity. The classic view of Golgi cells as local controllers of granule cell