Controversies in Neuroscience IV: Motor learning and synaptic plasticity in the cerebellum: Introduction

The histological simplicity and organization of the cerebellar cortex have fascinated neuroscientists for more than 150 years. These structural features made it possible for early anatomists to establish the basic connectivity among the different cellular elements of the cerebellum; the same features facilitated the physiological work of J. Eccles, M. Ito, R. Llinas and others in the 1960s that established the polarity and other aspects of the synaptic connections. The cerebellum thus became the first central nervous system structure in a vertebrate for which a wiring diagram could be drawn showing the morphology of the different elements, their connectivity, and their physiological interactions. This knowledge generated a great deal of excitement in the late 1960s and convinced many neuroscientists that a fundmental understanding of a major central nervous system structure was near. It seemed that only a few years' work would be necessary to establish "what the cerebellum does and how it does it" in the phrase of the tinie. The excitement and promise was reflected in the title of the 1967 book by J. Eccles, M. Ito, and J. Szentagothai that summarized the anatomical and physiological findings, "The Cerebellum a s a Neuronal Machine." The circuitry of the cerebellum and its promise still fascinate many neuroscientists, but a good functional understanding continues to elude us. Fascination with the cerebellum was heightened by the addition of a second theme to that of circuitry in our conceptual Pproach to cerebellar function. This is the theme of the cerebellar cortex as a site of learning and, in particular, of motor learning. This theme began with theoretical work in the late 1960s and early 1970s by D. Marr and J. Albus, which hypothesized that a temporal association between climbing fiber and parallel fiber inputs to PHdnje cells could alter the synaptic efficacy of the parallel fiber apse and that this could serve as the substrate for motor 'earning. Interest in this theme was strengthened by the discovery °' estibulo-ocular plasticity, a type of motor learning, and the rk of M. Ito and others suggesting that such plasticity might be explained by a Marr-Albus type of learning process in the cerebellar cortex. A possible role for the cerebellum in motor learning was ^ther supported by the finding that the temporal association between parallel fiber and climbing fiber inputs does indeed lead to a change, in fact a depression, in the synaptic effect of the paired Parallel fiber input. This phenomenon of long term depression (LTD) was first identified in vivo by Tongroach, Sakurai, and Ito d later in the in vitro preparations of Sakurai, Crepel, Hirano, linden, and their colleagues. The phenomenon of LTD is now well established, but the connection between this and other types of synaptic plasticity in the cerebellum and motor learning remains unresolved and controversial. In addition, although most would agree that the cere"um plays some role in motor learning, the nature of that role is also quite unresolved and controversial. These controversies concerning the role of the cerebellum in motor learning and the connection between motor learning and synaptic plasticity were " l e subjects of the symposium "Controversies in Neuroscience IV: Motor learning and synaptic plasticity in the cerebellum" upon which this BBS issue is based. The symposium brought together entists working on the cerebellum at the cellular level and entists working at the systems and behavioral levels. The goal was to build bridges across these separate levels of analysis and to 'P clarify the controversies surrounding the role of the cerebellum in motor learning. *nis symposium, held in Portland, Oregon on 24-26 August l»93, was the fourth in a series of five with the theme Controversies j n Neuroscience. These symposia were organized by the Robert S. Dow Neurological Sciences Institute and were supported by NIH, NSF and the Good Samaritan Foundation. Earlier symposia in the series focused on movement control, neural transplantation, and g-proteins in the nervous system. In August 1994, the last symposium of the series focused on controversial issues in persistent pain, and the papers from that symposium will be published in a future issue of BBS. The papers on plasticity at the cellular level are presented first, followed by the papers on learning at the systems or behavioral levels. The first paper ("Cerebellar long-term depression as investigated in a cell culture preparation"), by D. Linden, and the second ("Cellular mechanisms of long-term depression in the cerebellum"), by F. Crepel, N. Hemart, D. Jaillard, and H. Daniel, are concerned with the phenomenology of long-term depression (LTD) and the contributions of various cellular processes to the generation of LTD. Linden's work was done in cultured Purkinje cells whereas Crepel et al.'s was done in the in vitro slice preparation. Both papers show that LTD is dependent on calcium influx through voltage-gated calcium channels and on activation of protein kinase C via metabotyropic glutamate receptors. The two sets of studies obtained opposite results, however, with regard to the role of nitric oxide (NO). Blockade of NO synthesis had no effect on LTD in the culture preparation but had a clear effect in slices. The third paper ("Long-lasting potentiation of GABAergic inhibitory synaptic transmission in cerebellar Purkinje cells: Its properties and possible mechanisms"), by M. Kano, is concerned with plasticity at the synapses between inhibitory interneurons and Purkinje cells. Plasticity at inhibitory synapses has received much less attention than plasticity at excitatory synapses yet appears to be clearly present in the cerebellum and must have an important role in the plastic phenomena that take place there. The fourth paper ("Nitric oxide and synaptic plasticity: NO news from the cerebellum"), by S. Vincent, discusses the biochemical pathways responsible for the synthesis and action of nitric oxide (NO) and argues against an essential role for NO in LTD. The fifth paper ("Models of the cerebellum and motor learning"), by J. Houk, J. Buckingham and A. Barto, is the first of four papers concerned with the systems or behavioral level of analysis. It addresses the question, "How can modeling studies help us understand the role of the cerebellum in motor learning?" and reviews several important models of the cerebellum, including the authors' own model, which includes certain cellular properties of Purkinje cells as well as properties of the cerebellum as a whole. Houk et al. provide a bridge between studies at the cellular and systems levels. The sixth paper ("On climbing fiber signals and their consequence(s)"), by J. Simpson, D. VVylie, and C. De Zeeuw, reviews what is known about the messages conveyed by climbing fibers and about the effects of climbing fibers on Purkinje cell activity as well as some of the theories about the role of the climbing fiber in cerebellar function. There is much controversy regarding this role. This is not surprising, however, given the similar controversy about cerebellar function and the likelihood that the climbing fiber is a key, which once understood, might unlock the problem of cerebellar function. The seventh paper, by A. Smith, ("Does the cerebellum learn strategies for the optimal time-varying control of joint stiffness?"), is directly concerned with motor learning and argues that the cerebellar stores and shapes the time-varying patterns of muscle activation that control posture and movement. The author further agues that the teloceptive and proprioceptive sensory stimuli could serve as learned cues for associated patterns of muscle activation. The eighth and final paper, by W. Thach ("On the specific role of the cerebellum in