The Cerebellum: Brain for an Implicit Self (FT Press Science)

The Cerebellum

This page intentionally left blank

The Cerebellum: Brain for an Implicit Self Masao Ito

Vice President, Publisher: Tim Moore Associate Publisher and Director of Marketing: Amy Neidlinger Acquisitions Editor: Russ Hall Editorial Assistant: Pamela Boland Senior Marketing Manager: Julie Phifer Assistant Marketing Manager: Megan Graue Cover Designer: Alan Clements Managing Editor: Kristy Hart Project Editor: Jovana San Nicolas-Shirley Copy Editor: Charles Hutchinson Proofreader: Kathy Ruiz Indexer: Angela Martin Senior Compositor: Gloria Schurick Manufacturing Buyer: Dan Uhrig © 2012 by Pearson Education, Inc. Publishing as FT Press Upper Saddle River, New Jersey 07458 FT Press offers excellent discounts on this book when ordered in quantity for bulk purchases or special sales. For more information, please contact U.S. Corporate and Government Sales, 1-800-382-3419, [email protected]. For sales outside the U.S., please contact International Sales at [email protected]. Company and product names mentioned herein are the trademarks or registered trademarks of their respective owners. All rights reserved. No part of this book may be reproduced, in any form or by any means, without permission in writing from the publisher. Printed in the United States of America First Printing August 2011 Pearson Education LTD. Pearson Education Australia PTY, Limited. Pearson Education Singapore, Pte. Ltd. Pearson Education Asia, Ltd. Pearson Education Canada, Ltd. Pearson Educación de Mexico, S.A. de C.V. Pearson Education—Japan Pearson Education Malaysia, Pte. Ltd. Library of Congress Cataloging-in-Publication Data Ito, Masao, 1928The cerebellum : brain for an implicit self / Masao Ito. p. ; cm. Includes bibliographical references. ISBN-13: 978-0-13-705068-0 (hardback : alk. paper) ISBN-10: 0-13-705068-2 (hardback : alk. paper) 1. Cerebellum—Physiology. 2. Neuroplasticity. I. Title. [DNLM: 1. Cerebellum—physiology. 2. Motor Neurons—physiology. 3. Neuronal Plasticity. 4. Synaptic Transmission—physiology. WL 320] QP379.I85 2012 612.8’27—dc22 2011010870 ISBN-10: 0-13-705068-2 ISBN-13: 978-0-13-705068-0

To my wife, Midori Ito

This page intentionally left blank

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .viii Chapter 1:

Neuronal Circuitry: The Key to Unlocking the Brain . . . . . . .1

Chapter 2:

Traditional Views of the Cerebellum . . . . . . . . . . . . . . . . .22

Chapter 3:

The Cerebellum as a Neuronal Machine . . . . . . . . . . . . . .29

Chapter 4:

Input and Output Pathways in the Cerebellar Cortex . . . . .44

Chapter 5:

Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51

Chapter 6:

Pre- and Post-Cerebellar Cortex Neurons . . . . . . . . . . . . .60

Chapter 7:

Conjunctive Long-Term Depression (LTD) . . . . . . . . . . . . . .69

Chapter 8:

Multiplicity and Persistency of Synaptic Plasticity . . . . . . .81

Chapter 9:

Network Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90

Chapter 10:

Ocular Reflexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .105

Chapter 11:

Somatic and Autonomic Reflexes . . . . . . . . . . . . . . . . . .121

Chapter 12:

Adaptive Control System Models . . . . . . . . . . . . . . . . . .139

Chapter 13:

Voluntary Motor Control . . . . . . . . . . . . . . . . . . . . . . . . .150

Chapter 14:

Voluntary Eye Movement . . . . . . . . . . . . . . . . . . . . . . . .159

Chapter 15:

Internal Models for Voluntary Motor Control . . . . . . . . . .167

Chapter 16:

Motor Actions and Tool Use . . . . . . . . . . . . . . . . . . . . . .181

Chapter 17:

Cognitive Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . .193

Chapter 18:

Concluding Thoughts . . . . . . . . . . . . . . . . . . . . . . . . . . .204 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .213 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .261

Preface The primary rationale for my writing this book is that I have been involved in research on the cerebellum for half a century and it seemed appropriate to share with younger generations of researchers how thrilling and dramatic this epoch has been, particularly since research on the cerebellum has advanced not only our understanding of this fascinating structure but also that of overall neuroscience. I also have another rationale, however, which is more implicit but no less compelling. This is the desire to know how and to what extent we might proceed toward the goal of completely understanding the human brain, not only the detail of its huge mass of neurons, but also the means by which it can generate human intelligence, which has evolved over millions of years. Current systems neurobiology addresses this issue to some extent, but available methodology and technology are limited and guiding hypotheses are still sparse. To this end, research on the cerebellum is on the forefront for asking the question “How does our brain accomplish its most complex and sophisticated actions? Forty-five years ago, I co-authored the monograph The Cerebellum as a Neuronal Machine with John Eccles and Janós Szentagothai. This book described several neuronal circuits of the cerebellum, using analytical techniques that had advanced greatly in the late 1950s and early 1960s. Seventeen years later, I wrote a monograph The Cerebellum and Neural Control. Its focus was on the role of long-term depression in the cerebellum and this structure’s control of the vestibulo-ocular reflex. Such work suggested to me that the cerebellum was capable of learning and thereby played an essential role in adaptive neural control. In that 1984 book, the cerebellum was viewed as an assembly of many modular units (microcomplexes), each of which constituted a neurocomputing machine embedded in a control system of the brainstem and/or spinal cord. The book also contained a germ of the idea that the cerebellum performed internal model-based controls that were delineated and formulated computationally a few years later (in 1987) by Mitsuo Kawato and his colleagues. This 2011 monograph discusses advances made since 1984 in the overall study of neuronal circuits and the adaptive and model-based control of movement. It also presents new developments concerning the involvement of the cerebellum in motor actions and cognitive functions. The subtitle of the book, “Brain for an Implicit Self,” reflects my current view of the cerebellum. Its role in the adaptive control of movement is performed unconsciously. Even though voluntary movements, such as those needed to ski, skate, or play a piano, and so on, are performed under conscious awareness (of at least some components of the movements), there is no such awareness when these movements become more refined due to their practice. A similar situation prevails for our

Preface

ix

thoughts. When we think about some topic repeatedly, the thought becomes more and more implicit; that is, it requires less and less conscious effort, as in intuition. This suggests that the cerebellum aids the self in both movement and thought, but covertly, by use of its internal models. The question of just how neuronal circuits of the cerebellum can accomplish such an all-encompassing role will be a major challenge in the coming decades. I wish to thank all the cerebellar researchers cited in this monograph, whether living or deceased. Their expertise embraced, or continues to embrace, both the traditional disciplines of anatomy, physiology, biochemistry, pharmacology, and pathology and the many newer subdisciplines of neuroscience that derive their merit from both the life and physical sciences. Over the years these disciplines have continued to generate both new experimental data and novel theories. I am grateful to those who kindly permitted me to reproduce their illustrations in this monograph. I wish also to thank the many colleagues who spent some time in my laboratory at the University of Tokyo before 1989 and at the RIKEN Brain Science Institute after 1990. I am also greatly indebted to the University of Tokyo for strongly supporting my earlier research activity, particularly through the difficult period of campus disruption, and the RIKEN Brain Science Institute for providing me with such an excellent research environment. In publishing this monograph, I am particularly thankful to Prof. Douglas G. Stuart, Regents’ Professor Emeritus of Physiology, University of Arizona, for his advice about my use of the English language and discussions on posture and locomotion mechanisms. I am also indebted to Drs. Soichi Nagao and Tadashi Yamazaki, RIKEN Brain Science Institute, for our many discussions about the content of this monograph. Finally, I wish fervently that research on the cerebellum in the coming decades will be fruitful, particularly in clarifying its neuronal mechanisms in processing information of both a motor and a cognitive nature. Such progress will be a major step in the unlocking of brain mechanisms that support the implicit self.

Acknowledgments Grateful thanks are due to the publishers and editors of the following journals for their generosity in giving permission for reproduction of figures: Annual Review of Neuroscience (Annual Reviews), Journal of Neurophysiology (American Physiological Society), Journal of Neuroscience (Society for Neuroscience), Journal of Physiology London (The Physiological Society, John Wiley & Sons), Nature Reviews Neuroscience (Nature Publishing Group), Proceedings of National Academy of Sciences USA (National Academy of Sciences USA), and Progress in Brain Research (Elsevier). Grateful thanks also goes to the Instituto de Neurobiología “Ramón y Cajal,” Madrid, Spain.

About the Author Masao Ito is professor emeritus and former dean of the medical faculty at the University of Tokyo, and the founding director of the RIKEN Brain Science Institute. He has served as president of many international scientific organizations, including the International Brain Research Organization, the International Union of Physiological Sciences, the Human Frontier Science Program, the Science Council of Japan, and the Japan Neuroscience Society. Dr. Ito won the 1996 Japan Prize and the 2006 Gruber Prize in Neuroscience.

1 Neuronal Circuitry: The Key to Unlocking the Brain

1-1 Introduction The central nervous system (CNS) of vertebrates contains an enormous number of neurons, each having elaborate electrical and chemical signaling mechanisms. These neurons are interconnected via synapses to form intricate neuronal circuits. While such a circuit is composed of molecules within cells, it also processes information and generates a multitude of functions. Much effort has been and continues to be devoted to bridging these two properties of neuronal circuits to explore still largely unknown mechanisms of the CNS. The circuits of the cerebellum have been on the forefront of this endeavor. This chapter addresses the methodologies and fundamental concepts that are currently being used in the study of generic complex neuronal circuits before focusing in succeeding chapters on the cerebellum.

1-2 Decomposition and Reconstruction At a far earlier time, René Descartes (1596–1650) discussed the search for complex mechanisms of the universe and life by using the clock as a metaphor. During his time, this machine was considered the most complex of all the world’s man-made structures. Following Descartes (1649), it can be argued, as is prevalent today, that if one can dismantle a clock into its pieces and then successfully reconstruct them into the same functional machine, the precise nature of the clock is revealed. This methodology is still widely applicable when examining an object of unknown nature. It is dissected into simpler pieces, which can be understood, and then an attempt is made to reconstruct a model composed of all the pieces. If this model exhibits all the properties of the original object, it is indeed understood.

1

2

the cerebellum

The CNS includes the brain (contained within the skull), which weighs 1.3–1.4 kg in humans, and also the spinal cord, which extends into the vertebral canal. On the basis of conventional anatomy, the brain is grossly divided into the brainstem, cerebellum, and cerebrum. The cerebrum is further divided into the basal ganglia, limbic system, and neocortex (Figure 1). The cortex of the cerebral hemisphere is further subdivided to 52 areas (Brodmann, 1909; Garey, 1994) (Figure 2). The cerebellar cortex is also subdivided into nearly a hundred areas (see below and Color Plate II). Currently, we know that each of these divisions is composed of characteristic neuronal circuits that consist of numerous neurons of diverse types interconnected with each other via synapses. Moreover, there are even more numerous glial cells and finely branch blood vessels that support and nourish the neurons. The neuronal circuits in each subdivision constitute local networks, which are further integrated to form global neural systems across subdivisions or divisions. Current neuroscience is based on the belief that these networks and systems operate through specific mechanisms and play specific functional roles in the living body.

Figure 1 A sketch of major divisions of the CNS.

How can we unveil such mechanisms and the functional roles of neuronal circuits? The initial approach was to dissect the brain into experimentally manageable parts. This was the strategy adopted a century ago by Sherrington (1857–1952) and his group. They severed a segment of the cat spinal cord from its upper (and sometimes lower) segments (Figure 3A). Freed from the effects of other structures, the

1 • Neuronal Circuitry: The Key to Unlocking the Brain

3

severed segments exhibited reflexes with stable, straightforward input-output relationships via the dorsal and ventral roots, which could be subjected to precise scientific investigation.

Figure 2 Brodmann’s cerebral cortical areas.

The original dotted map published by Brodmann (1909) is converted here to outlined areas. (The original color version was provided by courtesy of Mark Dubin: http://spot. colorado.edu/~dubin/talks/brodmann/brodmann.html.)

When a neuronal circuit is defined in terms of its gross structure and function, it can then be decomposed into its individual neurons and their dendrites, axons, and synapses, using the currently available technologies of neuroscience. Thereafter, one may try to reconstruct a model of the initial reflex circuit by using the

4

the cerebellum

properties of all its constituent parts. In the process of reconstruction, the mechanistic principle(s) operating in the neuronal circuit may well be revealed.

Figure 3 Sketch of some spinal reflex circuitry.

(A) Schematic of some spinal reflex pathways (modified from Eccles et al., 1954). (B) Spinal circuitry drawn by Eccles as based on his group’s intracellular recording data on the recurrent Renshaw cell pathway (Eccles, 1963). In this and subsequent figures, sketches of a single cell and fiber (axon) usually represent groups of such units. A includes muscle spindles and their Ia afferents, two spinal cord segments, spinal motoneurons, Ia inhibitory interneurons, and two opposing muscles. B includes motoneurons and their axons supplying parts of muscle fibers, recurrent motor axon collaterals, Renshaw cells, and other spinal inhibitory interneurons. Abbreviations: ACh, acetylcholine; AS, annulospiral endings; BST, biceps and semitendinosus muscles; E, excitatory synapses: I/IS, inhibitory synapses; L6-L7, lumbar spinal segments; Q, quadriceps muscle; QIa, spindle Ia afferents supplying Q spindles. Symbols: blackfilled neurons and their endings, inhibitory; open neurons and their endings, excitatory. This figure is dedicated to a 1963 Nobel Laureate, John Carew Eccles (1903–1997), who was my postdoctoral mentor in Canberra, Australia, from 1959 to 1962. (See Ito, 1997, 2000; Stuart and Pierce, 2006.)

Sherrington’s group assumed that peripheral stimuli induced excitatory and inhibitory “states” in the spinal centers for various reflex circuits. John Eccles (1903–1997) and his colleagues (e.g., Brock et al., 1952) later identified these as formed by the membrane depolarization and hyperpolarization of spinal motoneurons via excitatory and inhibitory synapses (Figure 3B). Hubel and Wiesel (1960) discovered the unique responsiveness of individual neurons to line stimuli in the

1 • Neuronal Circuitry: The Key to Unlocking the Brain

5

visual cortex. They proposed a model of a neuronal circuit to explain how the characteristic responsiveness of “simple” and “complex” cells are formed, using input from concentric receptive fields of the lateral geniculate neurons. These early discoveries marked the start of modern neuroscience. Neuroscience is now dominated by the effort to decompose neuronal circuits into their cellular and molecular components. Many would argue, however, that reconstructing models of such circuits is equally important in our attempt to comprehend their functional principles (e.g., van Hemmen and Sejnowski, 2006; Stuart, 2007; for biology as a whole, see Noble, 2006). In the reconstruction process, it is possible to uncover novel principles operating in the original neuronal circuits. Analogies to man-made systems such as computers, control devices, and communication networks have also been helpful, as emphasized in the field of cybernetics by Nobert Wiener (1894–1964). The circular approach through decomposition and reconstruction provides a general method of fundamental research that features close interactions between experiments and theory (Figure 4). Initially, a factual observation of a complex subject may suggest a crude conceptual model, which serves to generate a prediction for a more focused experimental observation. If the prediction turns out to be correct, it supports the crude model, which is then refined to a more accurate conceptual model. This, in turn, can be converted into a substantial computational model, which is reproducible on a computer. Such an advanced model enables us to make further predictions, which can again be tested in even more precise experiments. In this iterative, cyclic development using observation-inspired models, modelbased predictions, and experimental testing of the predictions, a model is continuously refined until it accurately simulates the complex subject. A well-known and unique difficulty in research on the CNS arises from its highly hierarchical structure. Comprehension of our current understanding of the brain requires knowledge integrated across several hierarchical levels including molecules, cells, circuits, systems, and behaviors. It seems that ever since organic molecules appeared on earth, these hierarchical levels gradually accumulated through evolution until the human CNS evolved. The above-mentioned decomposition-reconstruction approach can be applied to any two successive levels of the overall hierarchy. For example, a simple neuronal circuit set can be reduced to its component neurons having somata, axons, dendrites, and synapses (Figure 5). In turn, these component neurons can be combined to reconstruct a model of the circuit at its original hierarchical level. Next, the component neurons can be further reduced to the lower level of ion channels, receptors, first and second messengers, and various organelles, whose combined properties can provide models of

6

the cerebellum

electrical and chemical signaling processes in neurons. Ion channels, receptors, and messenger molecules can be further reduced to an even lower level of proteins and their genes. The latter’s properties can be incorporated into models of the original ion channels and signaling molecules. By this method, the initial simple neural circuit can be linked step by step (not by jumps) to the molecular mechanisms subserving neuronal functions.

Figure 4 A decomposition/reconstruction cycle.

Research on the CNS starts usually with the experimental dissection of a relatively complex system into its simpler elements. To this end, a system is defined as a CNS unit (e.g., a spinal segment, the pineal body, oculomotor system) while it is undertaking a specific operation. In some cases the system can include peripheral effectors (i.e., glands, muscles). The dissected elements are assorted to construct models of the original system by means of theories and simulations. This circular approach may be based on observation-inspired models, model-based predictions, or experimental testing of a prediction. The model is continuously refined until it accurately simulates the complex system, as symbolized by three trajectories, which represent the first cycle (outer trajectory), an intermediate (middle) cycle, and the most refined (inner) cycle.

These processes can be considered as a long chain of decomposition-reconstruction events. By successively linking hierarchical levels, neuroscience research can trace the long pathway of evolution, from organic molecules to the cells of multicellular organisms, and eventually to the differentiated and diversified neurons that constitute simple neuronal circuits. In addition, evolutionary processes starting from simple neuronal circuits gradually led to the development of increasingly complex circuits and finally the human CNS. The fields of many subdisciplines of neuroscience are found at specific levels of the hierarchy. For one to understand the mechanisms and roles of neuronal circuits in the CNS, consistent and sustained effort is required to link coherently all levels of the hierarchy centering

1 • Neuronal Circuitry: The Key to Unlocking the Brain

7

around neuronal circuits that extend to cells and molecules on one hand, and to complex networks and systems on the other. In later chapters, we will see how far the cerebellum has been decomposed and reconstructed using these general methodologies.

Figure 5 The progression of decomposition/reconstruction cycles.

Shown are levels of analysis that extend from gene regulation (-2) to the cellular/molecular- (-1), neuronal- (0), simple circuit- (+1), and finally, complex circuit (+2) level of analysis. Major themes at levels -1 to +1 are also shown.

1-3 Neurons and Synapses The concept of the “neuron” was established over a century ago as the unitary component of neuronal circuits. Ramón y Cajal (1852–1934), hereafter shortened to “Cajal,” presented clear evidence for this in 1888, when referring to the relationship between Purkinje and basket cells in the cerebellum (see below and Color Plate IV) (Lopez-Munoz et al., 2006). Heinrich von Waldeyer-Hartz (1836–1921) formally proposed the neuron theory in 1891. Also, near the end of the nineteenth century, Sherrington and Michael Foster (1836–1907) coined the term “synapse” and spotlighted it as a key structure of the CNS. Since then, neurons and synapses have been the major targets of neuroscience investigations. All neurons commonly have somata extruding axons and dendrites (except for dorsal root ganglion cells, which have no dendrites). Dendrites not only expand the membrane area to accommodate many hundreds of synapses, but they also have finely

8

the cerebellum

compartmentalized functions (Hausser and Mel, 2003). On the other hand, different types of neurons are distinguished by their characteristic morphology, spike activities, synaptic actions (excitatory or inhibitory), and synaptic receptiveness (chemical or electrical). Subcellular structures such as postsynaptic density (PSD), cytoskeleton, endoplasmic reticulum, Golgi organ, and mitochondrion support these neuronal functions. Signal transduction involves various transmitters, modulators, receptors (ionotropic or metabotropic, or both), and second messengers. For these molecular mechanisms of neurons, numerous proteins, glycoproteins, and lipids, and their genes play essential roles.

1-4 Neural Networks Numerous neurons in the CNS assemble to form a structure called a “nucleus.” In certain areas of the brain and spinal cord (e.g., the superior colliculus, cerebellar cortex, hippocampal cortex, cerebral neocortex), different types of neurons regularly assemble to form a multilayered network. Donald Hebb (1904–1985) proposed the concept of “neuron assembly,” that is, a collation of neurons interconnected by synapses, in which the connectivity is modifiable according to experienced activities (Hebb, 1949). A famous proposal by Hebb is that the connection between two neurons firing synchronously is strengthened. Because of this “Hebbian” type of synaptic plasticity; a neuronal assembly can change its circuitry structure (corresponding to memory) and consequently modify its input-output relationships (corresponding to learning), as dependent on experienced activities. In an effort to verify Hebb’s concept of neuron assembly, Frank Rosenblatt (1928–1971) constructed a model network named a “simple perceptron.” It consisted of three layers of neurons connected in one direction, from the sensory cell layer to the association cell layer, to the response cell layer (Figure 6). Connections from the first to the second layer were fixed, whereas those from the second to the third layer were modifiable according to the instruction of an outside “teacher.” The teacher increased the weight of connection at all junctions transmitting signals from the second to the third layer when the simple perceptron responded correctly to sensory stimuli. The teacher decreased the weight at all second-to-third layer connections transmitting signals when the response was incorrect. When this training process was repeated, the simple perceptron improved its performance toward a success rate of 100%. This was the first man-made machine capable of learning. Ten years later, a counterpart of the simple perceptron was found in the cerebellum (see Chapters 3 and 9). The simple perceptron exemplified the success of the constructive approach (i.e., to understand by construction) for clarifying the operation of neuronal networks in the CNS.

1 • Neuronal Circuitry: The Key to Unlocking the Brain

9

Figure 6 The simple perceptron model of the cerebellum.

This figure is self-explanatory. See the text for further details on the operation of a simple perceptron. Abbreviation: CF, climbing fiber.

Twenty-four years after the construction of the simple perceptron, another form of multilayered neuronal assembly was proposed. It is usually called a “neurocomputer” (Rumelhart et al., 1986), in which errors were estimated by comparing the output of the third layer with a preset goal and were back propagated to the third-layer neurons. The errors acted on the junctions on the third-layer neurons formed with second-layer (hidden layer) neurons, and modified the efficacy of transmission from second-layer to third-layer neurons. The neurocomputer is often applied to model information processing in hippocampal and neocortical networks.

1-5 Systems Control Mechanisms in the CNS Local networks are interconnected globally throughout the CNS to form neural “systems.” A major type of such a system has the general form of a “control system,” which consists of a “controller (g)” acting on a “controlled object (G)” (Figure 7A). The controller receives input instruction that provides information about the nature of the required output (e.g., the goal, the trajectory of a movement). In turn, the controller generates command signals that drive the controlled object to respond appropriately. The controller may receive information about the performance of the controlled object (Figure 7A, feedback control), or it may operate without peripheral information (Figure 7B, feedforward control). The goal

10

the cerebellum

of a control system is to generate output responses identical to the input instruction. This can be achieved in a feedback control system if g is sufficiently larger than G, but in a feedforward control system, g needs to equal 1/G (Figure 7B). As emphasized by Baev (1999), this basic control system concept applies to various levels of organization within the CNS: in this monograph from reflexes to isolated voluntary movements and finally to coordinated motor actions. In addition, the concept is applied formalistically to cognitive functions.

Figure 7 The fundamental structures of a control system.

(A) A basic feedback control system. (B) A basic feedforward control system. (C) An adaptive control system equipped with an adaptive mechanism. This schematic applies to the cerebellar control of reflexes.

In recent years, modern control theory studies have opened the new fields of “adaptive control” and “model-based control.” In adaptive control, the controller is equipped with an adaptive mechanism to constitute an adaptive controller, which learns how to perform effectively in a given situation by altering its performance to match ever-changing environments. When a mechanism is attached to a feedforward controller, their overall input-output relationship f should be adjusted to 1/G (Figure 7C). On the other hand, model-based control was developed for robotic

1 • Neuronal Circuitry: The Key to Unlocking the Brain

11

arm control (An et al., 1988), and it has opened a new field of computational neuroscience for research on the cerebellum (Kawato et al., 1987). In the model-based control, a feedforward control system (Figure 7B) is attached with one of the two types of internal models, “forward” and “inverse” (Figure 8A, B). An internal forward model simulates the kinematics of a controlled object, whereas an internal inverse model simulates the dynamics or kinetics of them (for a definition, see Chapter 15, “Internal Models for Voluntary Motor Control”). Internal forward models support the controller by predicting the state of the system during actual actions. On the other hand, internal inverse models map the relationship between intended actions (or goals) and the motor command to bring about the action. An internal inverse model uses the desired position of the body as inputs to estimate the necessary motor commands, which would transform the current position into the desired one. An adaptive mechanism is involved to secure close simulation by these models. Such models may be formed in various parts of the CNS including, in particular, the elaborate neuronal networks of the cerebellar and cerebral cortices. Hereafter, models formed in the cerebellum and cerebral cortex will be called “cerebellar internal models” and “cerebral cortical models,” respectively.

Figure 8 General forms of model-based control systems.

(A) Internal forward model (G’) simulates the input-output relationship of the controlled object (G) and is inserted between the output and input of the controller (g). (B) Internal inverse model (1/G) simulates the output-input relationship of the controlled object (G) and is inserted between the input instruction and output response of the controller (g).

12

the cerebellum

1-6 Reflexes and Voluntary Movements The most fundamental control system structure in the CNS is an individual reflex operating via the spinal cord or brainstem. A reflex is performed by sequential activities in a neuronal circuit that connects serially sensory receptor cells, an afferent path, a reflex center, an efferent path, and an effector (a muscle, set of muscles, or a secretory gland or glands). This is typically exemplified by the stretch reflex, in which motoneurons maintain the length of a muscle constant by using feedback from muscle spindles (Chapter 11, “Somatic and Autonomic Reflexes”). In this case, a group of motoneurons and associated segmental interneurons constitute a controller, whereas the motor apparatus composed of the muscle(s) and a joint provides the controlled object. Numerous reflexes of various types operate in the spinal cord and brainstem to control simple somatic and visceral functions of a living body. The operation of reflexes is usually automatic—that is, it does not reach the level of conscious awareness—but in ever-changing environments it is indeed modifiable by use of adaptive mechanisms of the cerebellum (Chapters 10–12). We traditionally consider voluntary movements as a much higher order of movements than reflexes in the sense that they are controlled by “free will” and can be performed both automatically and at the level of conscious awareness, whereas reflexes are driven by peripheral stimuli and executed solely by automatic means. However, as our understanding advances for neuronal mechanisms underlying voluntary movements, distinctions between such movements and reflexes become blurred because many of the same neuronal circuits are employed for both types of movement (Prochazka et al., 2000; Hultborn, 2001). Practically speaking, however, we may still distinguish voluntary movements as initiated from the cerebral cortex, whereas reflexes operate largely within the spinal cord and brainstem. Typically, two cortical areas, the primary motor cortex and the frontal eye field, are involved in voluntary movements of the limbs and eyes, respectively (Chapters 13 and 14). In the systems control parlance emphasized in this volume, reflexes and voluntary movements may share neuronal circuits for their controller and controlled object structures, but they are separated from each other by the nature of the instruction signals that drive the controller. Instructions for reflexes arise from periphery, whereas voluntary movements are driven by “top down” instruction signals generated in higher centers of the cerebral cortex, including but not limited to the supplementary motor cortex and the anterior cingulate gyrus (see Chapter 13, “Voluntary Motor Control”). An interesting idea has been put forward to suggest that a central instruction causes a voluntary movement by an imitation or replacement of the peripheral stimulus that induces a reflex (the imitation hypothesis; Berkinblit et al., 1986). For instance, the CNS can voluntarily elicit a saccadic eye movement by means of the

1 • Neuronal Circuitry: The Key to Unlocking the Brain

13

imitation of the visual signals that could elicit the saccadic movement reflexively. In this sense, the central instruction may imply an “afference copy” of the peripheral stimulus. Such a capability of imitating a peripheral stimulus might emerge during evolution to develop a neuronal mechanism of voluntary motor control. Neuronal mehanisms underlying the postulated capability of imitation are unknown, but one may suppose that a group of neurons memorize those signals of peripheral stimuli that evoke a motor behavior reflexively and reproduce the same signals whenever a similar motor behavior is to be generated voluntarily. Here, one may recall the “mirror” neurons, which are present in certain cerebral cortical areas and are activated during both observed and performed hand actions, as discussed below (Section 8 and also in Chapter 16, “Motor Actions and Tool Use,” Section 5). These neurons appear to memorize perceptive signals representing certain successful motor actions performed by another individual and reproduce them as central instructions for their own body’s motor actions. Admittedly, however, the neuronal sites and mechanisms underlying free will in the high cerebral centers are still an enigma (Wegner, 2002).

1-7 Integration of reflexes One of the major ideas that Sherrington outlined in his 1906 book “The Integrative Action of the Nervous System” was that complex actions of the nervous system could be composed of a collation of reflexes, somewhat like building a house by piling up bricks. From the control systems perspective, there now appear to be at least eight ways to integrate reflexes into the overall control of movement. First, many that are driven by different sensory inputs may share the same controller and controlled object (Figure 9A). For example, three types of relatively slow ocular reflexes are driven individually by vestibular or visual stimuli, as will be seen later in Chapter 10, “Ocular Reflexes.” Nonetheless, they commonly share vestibular nuclear neurons as the controller, and eyeballs and the associated oculomotor system, as the controlled object. By this means, such a group of reflexes can achieve the common purpose of securing visual stability and acuity under natural behavioral conditions. In other words, these individual ocular reflexes are combined together to form a “multi-input” control system. Second, several individual reflexes may have different controllers (Figure 9B, Reflex 1, 2, 3 controllers), but they may share the same controlled object. For example, a slow ocular reflex can be integrated with a brisk saccade only in the form of half-fused control because these eye movements require controllers having substantially different properties for generating slow and brisk eye movements, respectively (Chapter 10). Third, reflexes may also be combined with a voluntary motor control system in a hybrid way (Figure

14

the cerebellum

9C) because of the similarity of control system structures for reflexes and voluntary movements (see Section 6). Design problems in such hybrid systems will be discussed later (Chapter 15).

Figure 9 Schematics of types of integrated reflex control.

(A) A multi-input reflex. (B) Half-fused system. (C) Hybrid control of a reflex and a voluntary movement. For further explanation, see the text. Abbreviations: I 1,2,3, three sensory inputs of different modality; VM, voluntary movement.

The spinal cord and brainstem contain a collation of reflexes to elaborate compound movements such as when assuming a posture, or when walking, swimming, and flying. Hence, a fourth way of integration is for some reflexes to be combined by mutual interaction (Figure 10A, Reflexes 1 and 2). For example, when one explores the visible world, a saccade and a head movement, the latter inducing the vestibuoocular reflex, occur in combination. This eye-head coordination involves an inhibitory cross talk between the independent eye and head controllers (Kardamakis and Moschovakis, 2009). A fifth way is for reflexes to be compounded when signals in a descending tract activate some combinations of reflexes to express behaviorally meaningful compound reflexes (Figure 10B, for review, see

1 • Neuronal Circuitry: The Key to Unlocking the Brain

15

Lemon, 2008). Anders Lundberg (1920–2009) and his colleagues found a good example in the cervical segments of the spinal cord. In the C3–C4 propriospinal system of the cat, interneurons were shown to receive extensive convergence from different primary sensory afferents and supraspinal centers (Lundberg, 1999; Alsternark et al., 2007). Through excitation or inhibition of relevant interneurons in this system, signals of each descending tract could produce compound reflexes to provide desired movement patterns, such as target reaching by the hand (Chapter 13). Operation of segmental spinal circuits sometimes involves a type of function generator (FG in Figure 10B, b). Locomotion is a good example of this type of compounding reflex. It involves flexion reflexes, crossed extension reflexes, interlimb coordination, and, in addition, a central pattern generator (CPG) mechanism for rhythm generation (Grillner et al., 1991, 2007; Grillner and Jessel, 2009) (see Chapter 11).

Figure 10 Schematics of the mutual interaction, compounding, and neuromodulation of reflexes.

Arrows denote synaptic actions, either excitatory or inhibitory. A shows how reflexes are combined by mutual interactions (Reflexes 1 and 2). In B, the compounding of reflexes 1, 3, and 4 is brought about by commands from descending tracts (a). Alternatively, reflexes 2, 3, and 4 can be coactivated by descending tracts (b) via FG, a function-generator. C shows how reflexes are modulated by aminergic and/or peptidergic innervation (represented by m and n) to exhibit a specific pattern of combination for behavior (m to reflexes 1, 2, and 4 and n to reflexes 1, 3, and 4). “Incentive” means a stimulus that leads to a specific behavior.

16

the cerebellum

The sixth way to integrate reflexes is by “neuromodulation.” As demonstrated in the crustacean stomatogastric nervous system (Marder et al., 1986; Selverston 1995), a small amount of a single peptide or amine may instantaneously rewire a neuronal circuit and switch behavior expression of the system. This mechanism may apply to the hypothalamus located in the most rostral and ventral part of the brainstem, which regulates innate behaviors including food intake, drinking, and reproduction; they are evoked by incentives such as the need for food, water, and reproductive activity, respectively. These behaviors involve a series of complex movements in order to approach and acquire the incentives. On the other hand, noxious stimuli such as drinking stale water or the figure of an enemy induce aversion, aggression, or defense reactions. An innate behavior involves a combination of reflexes and compound movements. For example, food intake involves locomotion to approach the food, rhythmic mastication, and the swallowing reflex. The hypothalamus contains a number of innate behavioral centers, each of which produces a specific pattern of behavior by secreting a neuromodulator substance through their widely distributed axons in the brainstem and spinal cord. The secreted neuromodulator substance may activate or inhibit a number of component reflexes and compounded movements (Figure 10C); hence, one circuit can be configured to perform a variety of different behaviors by activating neurons via certain types of neuromodulator receptors. The seventh way to use reflexes, and probably the most important in regard to the cerebellum, is by “nesting” (as in “matryoshka”), which has been used to explain perceptual organization (Leyton, 1987) and even the entire hierarchical organization of the CNS (Baev, 1999). The nesting idea is that a reflex composed of a controller and controlled object at the lowest level can be regarded as a controlled object at the next higher level. For example, a stretch reflex is a control system at the segmental level (Figure 11A), but at a brainstem level, it acts as the controlled object of the vestibulospinal descending tract neurons, which act as the controller (Figure 11B). In a similar vein, the primary motor cortex acts as a controller of the spinal segmental circuits, which are the controlled objects (Figure 12A, B). Furthermore, the entire corticospinal system constitutes a controlled object for the premotor cortex, which serves as its controller (Figure 12C). Through use of this nesting principle, collective reflexes integrated in the previous six ways constitute a controlled object for a higher-level controller, which can thereby exert control over many reflexes in various combinations.

1 • Neuronal Circuitry: The Key to Unlocking the Brain

17

Figure 11 The one-step nesting of a reflex within a supraspinal controller.

(A) The control system structure of a segmental reflex. (B) A being nested within the controlled object of a supraspinal controller. For further details, see the text.

Finally, the imitation hypothesis discussed in Section 6 provides the eighth strategy used by the CNS to evolve voluntary motor control systems utilizing the reflex control systems formed in the brain stem and spinal cord as a basement structure.

18

the cerebellum

Figure 12 The two-step nesting of a reflex within the premotor cortical control of a motor action.

(A) and (B) are similar to those in Figure 11. (C) shows a further nesting for control at a cortical level.

1-8 Motor Actions In the study of voluntary movements, the convention is to begin with analysis of a simple movement such as flexion of an elbow. However, voluntary movements that we perform daily are parts of “motor actions” that involve the participation of many body parts and even use of a tool to attain a purposeful goal. Moreover, motor actions also involve perceptual and conceptual activities, for example, in piano playing and dancing. It has been suggested that an “action schema” representing and coding motor actions is expressed in the posterior parietal and premotor (Broadmann’s area 6, see Figure 2) cortices (Jeannerod, 1994). For the present purposes, an action schema can be considered to be a cerebral cortical model. In primates, the premotor cortex expands rostrally from the primary motor cortex. It is generally assumed that the premotor cortex, in particular its dorsal part, plays a major role in computing and controlling complex motor actions (Wise et al., 1997). Moreover, the premotor cortex includes “mirror neurons,” which discharge similarly during a motor action performed by the self or by another

1 • Neuronal Circuitry: The Key to Unlocking the Brain

19

individual, as discovered by Rizzolatti and his colleagues (Rizzolatti and Craighero, 2004). Based on these and other lines of evidence presented in Chapter 16, we assume that the premotor cortex acts as the controller of motor actions. The premotor controllers act on controlled objects, which nest the primary motor cortex and lower motor systems (see Figure 12C). The postulated action schema is assumed to reside in the temporoparietal cortex and provide a cerebral cortical model to the premotor cortex. The same control system structure may apply to tool use if a tool is represented in an action schema together with body parts (Chapter 15). Action schema may include two related concepts (actually CNS properties) that are prominent in psychology: the “body schema,” which possesses a continually updating map of the self’s body shape and postures; and the “motor schema,” the self’s long-term memory structure capable of being retrieved as a whole, and then controlling the elaboration of motor skills composed of complex actions and movements. Both schemata are acquired or at least further refined by learning (Arbib, 2005; Stamenov, 2005). Along with the model-based control concepts discussed above (Section 5), these body/motor schemata can be considered as components of cerebral cortical models representing the forward and inverse models of the controlled object (Figure 8). These cerebral cortical models are presumably acquired during the initial learning of motor actions. As learning advances, the acquired body schema and motor schemata are transferred to cerebellar internal models (Chapter 15).

1-9 Cognitive Functions The prefrontal cortex is located rostral to the premotor cortex. It is implicated in working memory and acts as a key executive part of the CNS (Fuster, 1997). In the previous parlance on motor control systems, the prefrontal cortex receives instructions from the anterior cingulate gyrus and acts as an executive controller of a cognitive controlled object represented in the temporoparietal cortex. In Chapter 17, “Cognitive Functions,” this executive-cognitive system is considered to control highorder brain functions involving language, thought, evaluation, and decision making in humans, these being core components of human intelligence. During a thought process, the executive controller manipulates a cognitive concept encoded in the temporoparietal cortex, instead of a body part such as an arm or a leg. A crucial question is how to consider a cognitive concept as a controlled object. One may recall the term “mental model,” which Craik (1943) and Johnson-Laird (1983) defined as a psychological substrate for a mental representation of real or imaginary situations. It is a small-scale model of reality that the mind

20

the cerebellum

constructs and uses to reason, to underlie explanations, and to anticipate a future event. More concretely, mental models are representations of images, concepts, and ideas. One may also recall the term “schema,” which Jean Piaget (1896–1980) defined as being formed in a growing child learning to interpret and understand the world (Piaget, 1951). Piaget’s schema included both a category of knowledge and the process of obtaining that knowledge. Currently, these concepts are not mechanistic, and they lack a computational basis. In the presently used control system model for cognitive function, these schemas correspond to cerebral cortical models. We assume that learning proceeds further by incorporating them into cerebellar internal models (Chapter 17).

1-10 Beyond Movements Figure 8, which shows model-based control system designs, may be referred to for considering a mental model as a controlled object. For the present, computer simulation cannot reproduce this model because it lacks a computational basis. This difficulty is like the one that arose in the field of artificial intelligence. More than 50 years ago, a group of computer scientists proposed a study that would “... proceed on the basis of the conjecture that every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it.” These scientists were eager to make “... an attempt to find how to make computers that use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves” (McCarthy et al., 1955). This tempting approach in artificial intelligence, however, remains unsuccessful because it lacks the clarification provided by neural network mechanisms that can encode a concept or a specific piece of knowledge. Another profound question is how the operation of a neuronal circuit can be undertaken with conscious awareness. Sigmund Freud (1856–1939) and many more recent researchers have emphasized that only a few of the activities of the CNS are executed consciously. For example, one cannot bring to conscious awareness the thought processes involved in improving motor skills (e.g., skiing) by training (non-declarative memory). In contrast, one can readily recall cognitive experiences (declarative memory) (Squire, 2009). In other words, the neuronal circuits implicated in non-declarative memory are remote from the mechanisms of conscious awareness, whereas those involved in declarative memory are closely connected to conscious awareness. On the other hand, it has been shown that electric or transcranial magnetic stimulation (TMS) of the neocortex usually evokes vivid sensations or perceptions (Penfield and Perot, 1963; Coway and Welsh 2001), whereas stimulation of the subcortical tissues of the cerebellum (Riklan et al.,

1 • Neuronal Circuitry: The Key to Unlocking the Brain

21

1976; Koch et al., 2006) and basal ganglia (Chen et al., 2006) has no impact on conscious awareness. Conventionally, intelligence has been considered to require consciously activated cortical functions, but a substantial part of it is probably exerted subcortically and consequently unconsciously. In fact, intuitive thought is an important part of intelligence, but it is exerted unconsciously without obvious reasoning (Chapter 17). Neuroscience has reached a level of sophistication that is on the verge of addressing neural mechanisms underlying intelligence and conscious awareness. It seems likely that research on the cerebellum will be on the forefront of this endeavor.

1-11 Scope of This Monograph In the chapters that follow, the neuronal circuits of the cerebellum are decomposed and reconstructed as explained in this chapter. Early and recent historical material is presented in Chapters 2 and 3, and Chapters 4–9 update understanding of the cerebellum as an elaborate neuronal and molecular machine. Next, recent progress is presented about how this machine provides an advanced type of systems control for reflexes (Chapters 10–12) and voluntary movements (Chapters 13–15). The material covered in Chapters 10–15 reviews findings that came mainly after my 1984 book, The Cerebellum and Neural Control, and Barlow’s 2002 monograph, The Cerebellum and Adaptive Control. Chapters 16 and 17 examine the new possibility that the involvement of the cerebellum goes beyond movements to the higher-level functions of motor actions and cognition. The last Chapter 18, “Concluding Thoughts,” includes a summary of points made in preceding chapters about structural-functional relationships in neuronal circuit structures of the cerebellum as developed step by step in evolution.

1-12 Summary The decomposition-reconstruction method provides a logical and effective approach to studying the structure-function relationships of neuronal circuits. They are composed of local multilayered networks that interconnect globally to form neural control systems. Reflexes are the most fundamental units of neuronal circuits. Multi-input, half-fused, hybridized, mutually interacting, compounding, neuromodulating, nesting, and imitating are the eight ways to integrate reflexes into complex movements, voluntary movements, and innate behavior. A further integrated control is needed for both motor actions and, as yet to be determined, the mechanisms of cognitive thought.

2 Traditional Views of the Cerebellum

2-1 Introduction The cerebellum is a regular part of the CNS in vertebrate animals. It is recognized in lampreys, fish, amphibians, reptiles, birds, and mammals, up to humans. Among nonvertebrate animals, a cerebellum-like structure has been reported in octopus ganglia (Hochner et al., 2006), but its presence in other nonvertebrate species is unclear. The unique morphology of the cerebellum has been studied thoroughly for over a century. This led to the establishment of a map commonly applicable to various animals. Characteristically, the map initially involved the cerebellum’s connections with the vestibular nuclei in the medulla oblongata and with the spinal cord. The map then expanded in parallel with the further evolutionary development of the cerebral cortex. Distinctive involvement of the cerebellum in the acquisition of motor skills has also been uncovered on the basis of a large amount of data accumulated from lesion studies on animals and clinical investigations in humans. Furthermore, the cerebellum has been examined extensively by microscopy, revealing the presence of Purkinje cells and other cells of unique morphology. In this chapter, we trace the history of how these traditional views of the cerebellum have been formulated.

2-2 Morphological Map Erasistratus (304–250 BC) of Greece distinguished the cerebrum from the cerebellum (Malomo et al., 2006). Anatomical descriptions of the brain were almost complete during the Renaissance period, as we see in fine sketches of human brains drawn at that time. In the middle of the twentieth century, the characteristic morphology of the cerebellum was rigorously analyzed, and the data were compiled in two monumental volumes (Jansen and Brodal, 1954; Larsell, 1970). 22

2 • Traditional Views of the Cerebellum

23

When viewed from above, the mammalian cerebellum appears like a butterfly (Color Plate I). The middle part lying along the anterior-posterior axis is called the vermis because of its wormlike appearance. From the vermis, the hemispheres expand to the right and left like wings. Deep grooves divide the wings into three parts; the primary fissure separates the anterior and posterior lobes, and the posterolateral fissure separates the flocculonodular lobe from the posterior lobe. These divisions are further subdivided by transversely running grooves into individually named lobules (Color Plate II). Each lobule contains shallow folds, i.e., folia. The cerebellum is subject to a greater range of species variations than any other parts of the brain, but after much effort for over a century, anatomists reached a fundamental design of the cerebellum common to various vertebrate species (Larsell, 1970). It consists of ten divisions (lobules I–X) anteroposteriorly laid along the vermis and their transverse expansion into the hemispheres (lobules HI–HX). These divisions are shown schematically in an unfolded surface of the cerebellum (Color Plate II). Comparison of cerebella in various vertebrate species reveals the evolutionary origin of the lobal and lobular structures of the cerebellum. The flocculonodular lobe is phylogenetically the oldest (thus called the archicerebellum) and closely associated with the vestibular organ (thus called the vestibulocerebellum). The vermis is also old (paleocerebellum), and it is closely associated with the spinal cord (spinocerebellum). The cerebellar hemispheres are “new,” being expanded in mammals and primates in association with the development of the cerebral neocortex (neocerebellum or cerebrocerebellum). Particularly notable are the large paraflocculus in the porpoise and whale, the wide ansoparamedian lobule in the monkey, and the width of the entire hemisphere in the human. In addition to the right-left transversal lobular structure, the cerebellum is also divided into a number of longitudinal zones by various landmarks (Voogd, 1964; Groenewegen and Voogd, 1977; Groenewegen et al., 1979). First, the connections from the cerebellar cortex to the cerebellar nuclei by Purkinje cell axons are organized in three parts. The vermis is connected to the medial (fastigial) nucleus, the intermediate part of the hemispheres to the interpositus nucleus (in humans, emboliform and globose nuclei), and the lateral part of the hemispheres to the lateral nucleus (dentate nucleus in primates) (Color Plate III). In addition, a part of the vermis and the flocculonodular lobe are connected directly to the vestibular and other nuclei in the medulla oblongata. Second, in the afferent projection from the inferior olive (IO) to the cerebellar cortex, each small area of the IO projects to an anteroposteriorly extended longitudinal zone of the cerebellar cortex (Color Plate III). In this projection, seven major zones (A, B, C1, C2, C3, D1, and D2) are distinguished. The A and B zones are in the vermis, whereas the C1, C2, and C3

24

the cerebellum

zones cover the intermediate part of the hemispheres. The D1 and D2 zones are in the lateral part of the hemispheres. Third, a peptide, zebrin, distributes unevenly in the cerebellar cortex and marks the seven zones (Leclerc et al., 1992). A number of marker molecules are now available to label the longitudinal bands. Additional zones such as A2-, X-, Y-, and D0- zones have been defined (see Apps and Hawkes, 2005); they are not shown in Color Plates II and III. An interesting finding used intraventricular injection of adenovirus-carried markers, which labeled neuronal progenitor cells in a birth date-specific manner. Such injection in embryonic mice revealed that a cohort of Purkinje cells generated by mitosis on the same day formed a specific set of longitudinal bands, whereas Purkinje cells born one day earlier or later formed different sets of such bands (Hashimoto and Mikoshiba, 2004). The horizontal lobules and longitudinal bands form a latticed map that divides the cerebellar surface into nearly a hundred small areas (Color Plate II)). This lattice provides a guide map for exploration of the cerebellum. Earlier, Bolk (1906) noticed in the giraffe that the lobule simplex (lobule VI) was extraordinarily large. In view of the giraffe’s long neck, he pointed out the possibility that this cerebellar area was for the precise control of the long neck by powerful shoulder muscles (Glickstein and Voogd, 1955). Bolk’s idea of a single somatotopical map in the cerebellum does not hold for the entire cerebellum, but it pointed to the presence of functional localization in the cerebellar cortex (Manni and Petrosini, 2004). The elephant with a long trunk that is used like a human hand has an extraordinarily large cerebellum (18.6% of the total brain, as contrasted to 10% to 11% in humans), but no regional expansion has been described yet as specifically related to this animal’s nose (Shoshani et al., 2006). Another interesting and long-standing question concerns the prominent difference in the maps of the cerebellum of whales versus primates. In human and nonhuman primates, the cerebellum expands laterally, particularly in crus I and crus II (part of lobule HVII). The large cerebellum in whales (20% to 25% of the total brain), however, is due to an expansion of the paraflocculus, which occupies about three-fourths of the cerebellar surface (Oelschläger, 2008; Oelschläger and Oelschläger, 2009). It has been suggested that the whale’s large paraflocculus is a result of its adaptation to aquatic life, in which echolocation and acoustic communication are essential for survival and a meaningful social life (Oelschläger, 2000). In this regard, it is interesting that in the rat, the auditory cortex projects to the paraflocculus via the pontine nucleus (Azizi et al., 1985). In bats, which possess supersonic echolocation, the paraflocculus neurons respond to acoustic stimuli, specifically to their first harmonics (Horikawa and Suga, 1986). Hence, it is

2 • Traditional Views of the Cerebellum

25

probable that the paraflocculus is involved in echolocation and acoustic communication. The question of how the cerebellum contributes to such sonar systems will be discussed later in connection with its role in “sensory cancellation” (i.e., cancelling sensory perturbations evoked by self-initiated movements; Chapters 15 and 18). A large expansion of the ventral paraflocculus has been recognized in the endocast of a Cretaceous multituberculate, a mouse-sized mammal that appeared 130 million years ago and became extinct 90 million years later (KielanJaworowska, 1986). It is interesting to speculate that this mammalian species also developed a sonar system for communication in the dark with the aid of the paraflocculus. Another unique development is observed in the cerebellar valvula of mormyrid fish. This is basically a rostral protrusion of the cerebellum in the midbrain ventricle and much enlarged and folded over the brainstem and the telencephalon (Shi et al., 2008). The valvula is characterized by the prominent pattern of ridges on its dorsal surface. The valvula has been called a gigantocerebellum and intensively studied by neuroanatomists (Nieuwenhuys and Nicholson, 1967). It shares the general and basic organizational features with all other cerebella and cerebellar subdivisions but has in addition a number of unique features. For example, the efferent cells (corresponding to cerebellar nuclear neurons in Chapter 6, “Pre- and Post-Cerebellar Cortex Neurons”) are located close to Purkinje cells, and there are numerous deep stellate cells supplying specific inhibitory projections to efferent cells (Meek et al., 2008). This does not occur in usual cerebella (see Chapter 5, “Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex”). Because the valvula receives much of its input from the electrosensory system, its role in electrosensation is probable (Finger et al., 1981). The latticed areas have common homogenous microscopic structures, as will be examined later. In functional terms, they play diverse roles in connection with other divisions of the brain and spinal cord. It appears that numerous small “computers” of uniform structure and function are utilized individually for diverse purposes. The still remaining large blank in the mosaic map of the cerebellum implies that many more concrete roles of the cerebellum are yet to be identified.

2-3 Motor Skills In addition to the comparative anatomy mentioned previously, valuable strategy used widely in neuroscience research is to place a lesion in a brain and test for subsequent dysfunctions. Cutting, ablating, coagulating, and injecting certain toxic amino acids into brain tissues have been used to make lesions. Clinical cases with discrete cerebellar lesions also provide similarly useful data. Dow and Moruzzi

26

the cerebellum

(1958) compiled such data from lesion studies collected up to the middle of the twentieth century. Transient blockade of functions by applying various pharmacological inhibitors or antagonists is a further development, and most recently, genetic manipulation has become a powerful tool to form a lesion in a specific element of neuronal circuits. Elaborate test paradigms for detecting lesion-induced behavioral disorders have also been developed recently. It is interesting that in phrenology, as initiated by Franz Joseph Gall (1758–1828), the occipital part of the cranium that overlies the cerebellum was assigned as the center of love. This assertion was groundless, but it apparently stimulated pioneers of the nineteenth century. Luigi Rolando (1773–1831) removed the cerebellum in various animals and found subsequent motor disturbances. Jean Pierre Flourens (1794–1867) ablated the cock cerebellum, but seeing the cock still seeming to be attracted to a hen, he discarded the phrenological hypothesis (Glickstein et al., 2009). Flourens (1822) also observed that animals with a lesioned cerebellum still moved spontaneously, but only clumsily. He concluded that the cerebellum was responsible for movement coordination, in contrast to the cerebrum, which initiated movements via the spinal cord. This was amazing insight for a nineteenth century scientist! In the early twentieth century, clinical neurology revealed that cerebellar dysfunction in humans was characterized by the loss of smooth, precise movements. Babinsky (1857–1932) defined dysmetria as a characteristic symptom of cerebellar dysfunction. A simple clinical test for dysmetria is a patient’s failure, with the eyes closed, to quickly and accurately touch the nose with an index finger (Chapter 15, “Internal Models for Voluntary Motor Control”). During World War I, Gordon Holmes (1876–1965) examined soldiers with a discrete gunshot wound in the cerebellum. He found that they exhibited a significantly slower onset of arm retraction on the damaged side. The cerebellum thus appeared to be required for a movement that was too quick to be influenced by sensory feedback. Intention tremor is another symptom frequently observed in cerebellar patients. It is characterized by coarse trembling of a forelimb, which is accentuated by the execution of purposeful voluntary movements such as reaching by the hand. It may expand to involve the head, eyes, and the upper half of the body. Intention tremor has been reproduced in monkeys by cooling the dentate and interpositus nuclei (Flament and Hore, 1988). It thus appears that the cerebellum normally prevents intrinsic potential oscillations to ensure smooth movements. Early on, such subtle control of movements suggested the need for a form of motor learning in the cerebellum. In fact, there were classic observations that suggested the learning capabilities of the cerebellum. Flourens (1842) removed the

2 • Traditional Views of the Cerebellum

27

superior half of the cerebellum in a young cock and observed 15 days later that equilibrium was totally reestablished. However, when he removed the entire cerebellum from a hen, it did not recover its equilibrium even four months after the operation. Luciani (1891) reported a notable observation in 1891. He made a partial lesion in the cerebellum of a dog. After the animal’s full recovery, he placed a second lesion adjacent to the original one. A severe motor disturbance ensued as if to suggest that the first and second lesions were made at the same time. This was interpreted to mean that the second lesioned area was involved in recovery from the first lesion. Moruzzi (1910–1986) and his group postulated that cerebellar circuits either have a learning capability in and of themselves, or they are required for recalling a memory stored somewhere else in the CNS (Batini et al., 1976). They apparently considered that the cerebral cortex was a possible memory site, because of the observation that a later cerebral lesion cancelled motor recovery after cerebellar ablation. The capability of cerebellar circuits to recover from and compensate for a functional deficit, as demonstrated in animals, argues against a perplexing question we sometimes face: why does a human who lacks a cerebellum, as is sometimes reported, exhibit no obvious dysfunction? Glickstein (1994) examined a number of such individuals and reported that the cerebellum was still present, albeit severely atrophied. He also emphasized that individuals with such an atrophied cerebellum definitely exhibited certain abnormal motor behavior. It seems likely that the viable portion of the cerebellum may compensate for deficits produced by its damaged portions. Possibly, the cerebral cortex provides additional compensation. Experimental lesion studies and clinical observations have certainly highlighted the involvement of the cerebellum in the control of movement. As a result, the popular idea that the cerebellum is solely a motor center has tended to prevail. However, Moruzzi (1940) recognized several decades ago that the cerebellum contributed to both cardiovascular and respiratory regulation. The involvement of the cerebellum in mental activities has been suggested, albeit only occasionally and mainly on the basis of clinical observations that focused until recently on disturbances in expression using spoken words and gestures (Chapter 17, “Cognitive Functions”).

2-4 Microscopic Features In 1837, Jan E. Purkinje (1787–1869) observed cerebellar tissues under a microscope and found oval objects. These were the first individual neurons observed in the literature, and they are now called “Purkinje cells.” In the early twentieth century, using the amazing silver staining method, Cajal observed and drew intricate

28

the cerebellum

neuronal network structures that he found throughout the brain (Ramón y Cajal, 1911; Sotelo, 2003). In the cerebellum, he depicted the characteristic morphology of Purkinje cells, Golgi cells, basket cells, stellate cells, and granule cells (Color Plate IV). He also identified mossy fiber and climbing fiber terminals. His drawings even indicated the possible directions of neuronal signals conducted and transmitted from one cell to another. These elements are arranged in three layers (molecular layer, Purkinje cell layer, and granular layer). The structure is homogenous throughout the cerebellar cortex except for some regional differences. The tradition of Cajal’s fine microscopic anatomy was continued by a number of distinguished anatomists, and in the 1960s it was further advanced by the widespread contributions of electron microscopists. As a harbinger of the coming breakthrough in the 1960s, some important findings were reported in the 1950s. Ragner Granit (1900–1991) and Charles Phillips (1916–1994) used microelectrode recording to demonstrate the antidromic spikes of Purkinje cells and the so-called D potentials. The latter corresponded to intracellularly recorded climbing fiber responses (Granit and Phillips, 1956). Janos Szentágothai (1912–1994) and coworkers identified the origin of climbing fibers in the IO. They surmised from the pattern of connections among granule cells, Purkinje cells, and basket cells that basket cells were inhibitory neurons (Szentágothai and Palkovits, 1959; Szentágothai, 1963).

2-5 Summary Morphological studies revealed evolution-related structural-functional maps of the cerebellum, and lesion studies established the specific involvement of the cerebellum in the learning of precise movements. These studies laid a firm foundation for the modern approach to the cerebellum, which focuses on neuronal circuits that are discussed in later chapters.

3 The Cerebellum as a Neuronal Machine

3-1 Introduction In the 1960s, the cerebellum was considered to be an elaborate neuronal machine composed of intricate neuronal circuits with geometrical refinement. It was thought to process information that was critical for the acquisition of motor skills. During the five subsequent decades, research on the cerebellum has been devoted largely to addressing questions about how its neuronal circuits are constructed and function, and what specific roles they play. The principles of modern systems control, particularly adaptive and model-based control, have been introduced. Furthermore, the role of the cerebellum in the manifestation of intelligence is now under consideration. This progress is summarized in the following sections, including some of my personal experiences and impressions throughout this 50-year period.

3-2 The 1960s In the 1960s, Professor John Eccles, who had discovered inhibitory synapses in the spinal cord (Brock et al., 1952), turned to the study of the cerebellum with his talented colleagues in Canberra, Australia. It was a remarkable time when electrophysiology with glass microelectrodes enabled intracellular recording in individual neurons. Using this technology, Eccles distinguished two types of neurons, excitatory and inhibitory, in the cat spinal cord (Figure 3A, B). Excitatory neurons were shown to supply solely excitatory synapses and induce excitatory postsynaptic potentials (EPSPs) or currents (EPSCs) in their target neurons. In contrast, inhibitory neurons were shown to supply inhibitory synapses that induced inhibitory postsynaptic potentials (IPSPs) or currents (IPSCs) in their targets. Using the same technology and taking advantage of the geometrical arrangement in cerebellar circuits, Eccles and his associates quickly identified basket cells, stellate cells, and Golgi cells as inhibitory neurons, and granule cells, mossy fibers, and climbing fibers as excitatory elements (Chapters 4 and 5). 29

30

the cerebellum

Here, let me recall my first experience with the cerebellum. In 1962, I returned to Tokyo from Eccles’ laboratory where I had studied spinal motoneurons for three years. On my return I worked with several colleagues on two types of giant neurons in the brainstem. These were Otto Deiters’ (1834–1863) giant neurons (Deiters, 1865; see Mazzarello, 1999) and magnocellular red nucleus neurons. I was familiar with these neurons from the earlier anatomy lectures of Professor Teizo Ogawa (1901–1984), which I had heard while a medical student. We equipped one laboratory exclusively with hand-made electronic instruments. It was shared by two subgroups: the late Nakakira Tsukahara (1933–1985) and Kesiuke Toyama for the red nucleus and the late Mitsuo Yoshida (1933–1998) and me for Deiters neurons. In November 1963, we were successful in recording intracellularly in a Deiters neuron of an anesthetized cat. When we applied an electric shock to needle electrodes inserted into the cerebellum, it caused a large swing of green spots on the screen of a cathode ray oscilloscope. This was an IPSP induced via long axons of Purkinje cells (Figure 13) (Ito and Yoshida, 1964). We then recorded from cerebellar nuclear neurons and confirmed the consistent occurrence of inhibition, thereby enabling our conclusion that Purkinje cells were uniformly inhibitory neurons (Ito et al., 1964). Moreover, when Kunihiko Obata joined us a short time later, we found that iontophoretic application of gamma-amino-butyric acid (GABA) to Deiters neurons induced a membrane hyperpolarization like IPSPs (Obata et al., 1967). This evidence showed that Purkinje cells were GABAreleasing inhibitory neurons. At that time, there were the beliefs that (1) large neurons with long axons were excitatory, whereas small neurons with short axons were inhibitory; and (2) excitatory neurons were major “players” in the brain, whereas inhibitory neurons acted as “local commutators.” Indeed, the inhibitory neurons identified by Eccles and his colleagues in the spinal cord, hippocampus, and cerebellar cortex were all short-axoned, relatively small neurons. Our Purkinje cell finding was also at variance with the then-conventional thought that the cerebellum was involved in both excitatory and inhibitory functions because its stimulation induced either contraction or relaxation of limb muscles, as dependent on the stimulation conditions. We showed, however, that target neurons for Purkinje cell inhibition receive excitation via axon collaterals of mossy fiber and climbing fiber afferents (Figure 13) (Ito et al., 1969). Morphological details of such axon collaterals were revealed later (Shinoda et al., 1992; Sugihara et al., 1996). We found also that stimulation of the cerebellum often facilitated Deiters neurons via inhibition of Purkinje cell inhibition—that is, disinhibition (Ito et al., 1968). When these controversies subsided, Eccles generously offered me the opportunity to co-write with Szentágothai and him the 1967 monograph, The Cerebellum as a Neuronal Machine.

3 • The Cerebellum as a Neuronal Machine

31

Figure 13 Schematic neuronal circuit showing how electrical stimulation of the cerebellar cortex induces three major effects in Deiters neurons.

When recorded intracellularly in a Deiters giant neuron (DE), electrical stimulation of Purkinje cells induces inhibitory postsynaptic potentials (IPSPs)(a). However, excitatory postsynaptic potentials (EPSPs) are also induced by activation of mossy fibers and climbing fibers via their axon collaterals (b). If basket cells are stimulated, they inhibit Purkinje cells so that Deiters neurons are disinhibited and generate slow disibhibitory depolarization (c). Part of the neuronal circuit located in the cerebellar cortex is enboxed by broken lines. Abbreviations: CA, cells of origin of mossy fibers (MF); IO, inferior olive that issues climbing fibers (CF); ME, microelectrode; PC, Purkinje cells. Note that both CF and MF project collaterals to the Deiters cells. (Based on the data of Ito and Yoshida, 1966, Ito et al., 1968, 1969.)

We wrote this book at a time when computers were beginning to be used widely in neuroscience and artificial intelligence seemed of particular promise. After Wiener popularized cybernetics in his 1948 book, modern control theories had appeared to be a promising approach for advancing understanding of the mechanisms of the CNS. For example, Arbib (1971) applied cybernetic concepts to brain theories. Our 1967 monograph emphasized wiring diagrams of the cerebellum, and we encouraged computational scientists to collaborate with biological researchers to determine their significance. At the end of the book, we stated confidently that the enlightened discourse between such theorists on the one hand and neurobiologists on the other will lead to the development of revolutionary hypotheses of the way in which the cerebellum functions as a neuronal

32

the cerebellum

machine and predicted that these hypotheses will lead to revolutionary developments of experimental investigation (Eccles et al., 1967). Several international symposia were held with a focus on this theme. The most impressive one for me was held in 1967 at Salishan Lodge near Gleneden Beach on the Oregon coastline, USA, as organized by Francis Schmitt (1903–1995) and Eccles. Donald MacKay (1922–1987) led discussions among theorists, computer experts, and bioengineers. There was no immediate outcome from this and other such meetings, however. Despite the impressive beauty of its wiring diagrams (Color Plate V), the “neuronal machine” concept of the cerebellum remained vaguely defined as “a relatively simple machine devoted to some essential information processing.” I was frustrated enough at the Salishan meeting to ask what else experimentalists would need to uncover before we would be able to understand the meaning of these wiring diagrams. Someone equally frustrated replied that the available diagrams were too simple to construct even a primitive radio, so more information was urgently needed before any meaningful model could be conceived. However, an important clue had already been with us for a long time—that is, the presence of climbing fibers in the cerebellum, as described by Cajal (1911) (Figure 14). The contrasting connectivity of each Purkinje cell with only one

Figure 14 Convergence of climbing and parallel fibers onto Purkinje cells.

A part of Figure 104 of Cajal (1911) is shown with the right, left axis reversed to match Color Plate V. A, mossy fiber; B, Purkinje cell axon; a, granule cell; b, parallel fiber; c, Purkinje cell; d, climbing fiber. Arrows indicate the supposed directions of signal flow.

3 • The Cerebellum as a Neuronal Machine

33

climbing fiber and numerous parallel fibers had been interpreted only as unique cases of convergence and divergence. Characteristic electrical events induced in a Purkinje cell by impulses of parallel fibers and climbing fibers (as exemplified in Figure 15) had been revealed by Eccles et al (1966a, b) and Thach (1967), but no one thought of its implication for synaptic plasticity except Brindley (1964) who pointed out this possibility.

Figure 15 Bioelectric potentials of Purkinje cells.

Both intracellular recordings in slices (A–D) and extracellular recordings in vivo (E–G) are shown. (A) Simple spikes induced by stimulation of parallel fibers (PFs) on the pial surface of a cerebellar folium. (B) Complex spikes evoked by stimulation of climbing fibers in the white matter. The complex potentials so evoked are composed of an EPSP and Na+ and Ca2+ spikes. (C) An AMPA-EPSP evoked by stimulation of parallel fibers. (D) mGluR-EPSPs evoked by repetitive stimulation of parallel fibers in the presence of an AMPA antagonist. (E) Spontaneous discharge from a Purkinje cell. (F) A simple spike in an expanded time scale. (G) A complex spike similarly shown. In A–D, five consecutive sweeps repeated at 0.2 Hz were averaged. (From unpublished data of Le and Ito.)

3-3 The Marr-Albus Model In the aforementioned climate, David Marr (1945–1980), James Albus, and a few other theorists proposed theoretical models of the cerebellar neuronal machine (e.g., Marr, 1969; Albus, 1971). This was an eagerly awaited breakthrough for computational neuroscience. I remember its great impact on me after reading Marr’s 1969 article. I felt that his theory converted our wiring diagrams of the cerebellum into a meaningful blueprint.

34

the cerebellum

The crucial assumption adopted in Marr’s theory was the use of synaptic plasticity as a memory element in neuronal circuits. At that time this was but a theoretical possibility and totally lacking in supportive experimental evidence. As mentioned in Chapter 1, “Neuronal Circuitry: The Key to Unlocking the Brain,” Hebb (1949) had already proposed the concept of Hebbian synapses, whose transmission efficacy increased when the presynaptic and postsynaptic neurons fired in synchrony. Brindley (1964) pointed out the possibility that the convergence of parallel fibers and climbing fibers onto Purkinje cells implied the presence of Hebbian synapses, since climbing fiber signals are so powerful that these inevitably excited Purkinje cells. Thus, if both parallel fibers and climbing fibers were activated synchronously, parallel fiber-Purkinje cell synapses were activated both presynaptically and postsynaptically, that is, the type of condition that induced a Hebbian form of plasticity. In Marr’s (1969) model, as based on Brindley’s suggestion, learning actions were considered to occur as follows. Each climbing fiber conveyed a cerebral instruction for an elemental movement, and the receiving Purkinje cell was also exposed via the mossy fiber input to information about the context in which the climbing fiber fired. During rehearsal of an action, each Purkinje cell could learn to recognize such contexts, and later, after the action had been learned, the occurrence of the context alone was enough to fire the Purkinje cell, which then caused the next elemental movement. Albus’ model (1971) was a close analogy to the simple perceptron, assuming that climbing fibers played the role of the outside teacher as a supervisor (recall Figure 6). When a successful performance of the cerebellum was recognized, relevant climbing fibers sent signals that potentiated concurrently activated parallel-fiber synapses on Purkinje cells (i.e., potentiation of the synapses that brought about success). On the other hand, when the performance was unsuccessful, relevant climbing fibers sent signals to depress concurrently activated, parallel-fiber synapses on Purkinje cells (i.e., depression of the synapses involved in failure). However, it is impossible to use the same climbing fiber for both potentiation and depression in real synapses. This meant that one of them had to be chosen. Albus (1971) selected depression for several technical reasons, whereas Marr used potentiation after success. Theoretically speaking, learning was possible using either model. Thus, these models raised alternative possibilities to be selected on an experimental basis. It is to be noted that the simple perceptron is primarily designed for discrimination of spatial patterns and has no capability of discriminating temporal patterns. A decade after Marr’s and Albus’ models, Fujita (1982a) proposed an adaptive filter model of the cerebellum able to discriminate temporal patterns by assuming that the neuronal circuit involving mossy fibers, granule cells, parallel fibers, and Golgi cells constitutes a phase converter, which generates a set of multiphase

3 • The Cerebellum as a Neuronal Machine

35

versions of a mossy fiber input. Figure 16 shows schematically the early idea of the operation of Fujita’s adaptive filter model of the cerebellum when the input signal is sinusoidal. Fujita (1982b) incorporated successfully this phase converter concept into a model of VOR adaptation and reproduced successfully the adaptation of the VOR (Chapter 10, “Ocular Reflexes”). The importance of the granule cellGolgi-cell-granule cell pathway as a clock in the cerebellum has now been well recognized (Chapter 9, “Network Models”).

Figure 16 Adaptive filter model of the cerebellum.

This model explains how the cerebellar network recognizes temporally encoded signals. It is assumed that a phase-converter consisting of the mossy fiber (MF)-granule cell (Gr open circle)-Golgi cell (Go filled circle) circuit generates a set of multiphase versions of mossy fiber (MF) input (represented by sinusoidal discharge). When Purkinje cells (PC) use conjunctive LTD in their learning, a certain phase-shifted version of the input, which is out of phase to the climbing fiber error signals, is selected by Purkinje cells. On the other hand, granule cell to Purkinje cell transmission in phase with the climbing fiber (CF) input (indicated by a left-directed arrow) will be depressed. (Explanation based on Fujita’s [1982a] model; see also Dean et al., 2010 for another explanation of the model.)

3-4 Long-Term Depression In the late 1960s and early 1970s, many laboratories apparently tried to reveal such synaptic plasticity, but in vain. It is widely known that Eccles invited Marr to sit in front of a cathode ray oscilloscope with him while they tested the effects of conjunctive stimulation of climbing fibers and parallel fibers using stimulus parameters chosen by Marr. No sign of synaptic plasticity was then observed, however. At that time, experiments were conducted in vivo such that stable intracellular

36

the cerebellum

recording was not possible for a period sufficiently long to detect synaptic plasticity. Accordingly, transmission across parallel fiber-Purkinje cell synapses was examined only by extracellular recording of field potentials. However, as compared to field potentials recorded in the hippocampus to reveal long-term potentiation (Bliss and Lomo, 1973), those in the cerebellar cortex were ten times smaller. This meant that before the availability of high-performance electronic averagers, any potential long-term modification of synaptic transmission could not be detected. The observation was further impeded because several factors in in vivo experiments were later shown to interfere with the occurrence of the synaptic plasticity: postsynaptic inhibition caused by basket/stellate cells (Ekerot and Kano, 1985), local bleeding resulting in the release of hemoglobin that absorbs nitric oxide (Nagao and Ito, 1991), and general anesthesia (Vigot et al., 2002). Despite the preceding evidence to the contrary, I agreed with the two theorists, Marr and Albus, because, as shown below, the flocculus hypothesis of the vestibuloocular reflex (VOR) that I was proposing at that time matched very well with their models. In 1979, I visited Professor David Hubel at Harvard Medical School to present a seminar. When it ended, Marr approached me, this being our first and only interaction. He mentioned his interest in my flocculus hypothesis for the VOR and asked me to send him any related publications. He also said that he would soon leave for the U.K. for leukemia treatment but would possibly visit Japan the following year to receive a prize from an artificial intelligence group. I told him that I had been waiting to meet him for ten years and that I was continuing my research on synaptic plasticity. Upon returning home, I received a letter from Marr, in which he mentioned gracefully that he, too, had been waiting ten years to meet me. Sad to say, Marr did not come to Japan, and I regretted that I could not tell him person to person about the new positive evidence of synaptic plasticity, which I reported at the XXVII Congress of the International Physiological Union (IUPS), which was held in Budapest in June 1980 (Ito et al., 1981). Béla Julesz (1928–2003) consoled me to some extent, however, when he informed me that he had written to Marr, who was by then quite ill in bed in Cambridge, Massachusetts, to tell him about my Budapest report. Sadly, Marr died in late 1980. The new evidence presented in Budapest (Ito et al., 1981) was a result of my change in strategy from using field potentials to test for parallel fiber-Purkinje cell transmission to measuring the rate of Purkinje cell discharge in response to halfmaximum parallel fiber stimulation (“firing index”). While recording from a Purkinje cell in the flocculus, Masaki Sakurai, Pavich Tongroach, and I witnessed that conjunctive stimulation of vestibular mossy fibers and climbing fibers decreased unfailingly the firing index (Ito et al., 1982). Even though we were stimulating vestibular mossy fibers, field potentials in the vestibular nuclei and flocculus

3 • The Cerebellum as a Neuronal Machine

37

granule layer were confirmed not to reveal any related changes. We also recorded from putative basket cells, in which conjunction induced no depression like that observed in Purkinje cells. Because Purkinje cells and basket cells share the mossy fiber-parallel fiber pathway, we reasoned that the depression specific to Purkinje cells must have taken place in the Purkinje cells, themselves. Moreover, we demonstrated that the sensitivity of Purkinje cells to iontophoretically-applied glutamate (the transmitter released from parallel fibers), but not to aspartate or N-methyl aspartate (not a transmitter for parallel fibers), was depressed for a considerable duration after combining climbing fiber stimulation and glutamate application. Shortly thereafter, we received a grant to purchase a high-performance electronic averaging instrument. Its use enabled Masanobu Kano and me to record the field potentials representing monosynaptic activation of Purkinje cells by parallel fiber impulses and to demonstrate that conjunction induced long-lasting depression of these potentials, this being definite evidence of the manifestation of LTD (Ito and Kano, 1982). Next in my laboratory, Karl-Frederic Ekerot and Kano used direct stimulation of parallel fibers combined with Purkinje cell firing indices to reveal the occurrence of LTD (Ekerot and Kano, 1985). Later, the successful recording of LTD in cerebellar slices (Sakurai, 1987) prompted many more studies of LTD, which were undertaken worldwide. By 1990, LTD was established as a unique type of synaptic plasticity (Ito, 1989). Nowadays, conjunctive LTD can be observed routinely in tissue cultured Purkinje cell preparations developed by Linden’s group and in the cerebellar slice preparations used in other laboratories, including my own (Figure 17). In the 1990s, signal transduction processes underlying LTD became a subject of extensive investigation in many laboratories (see Daniel et al., 1998). I recall that when I moved to RIKEN (Institute of Physical and Chemical Research) in 1990, little was known about this subject. Now, however, a complex flow chart is available. It shows chemical signals involving more than 30 different molecules (for review, see Ito, 2001, 2002). While I was concentrating on the mechanism of signal transduction for LTD, there were notable research developments in several directions on the nature of cerebellar synaptic plasticity. Postsynaptic LTP as the counterpart of conjunctive LTD had long been missing, but Lev-Ram et al. (2002, 2003) finally found it. The involvement of cerebellar/vestibular nuclear neurons in learning, in addition to LTD in the cerebellar cortex, was suggested early on (Miles and Lisberger, 1981; Lisberger and Sejnowski, 1992; Raymond et al., 1996). It has now been shown quite clearly (Kassardjian et al., 2005; Shutoh et al., 2006; McElvain et al., 2010). Moreover, a wide variety of synapses in the cerebellar cortex have been shown to be activity-dependent and subject to plastic modification (see Hansel et al., 2001). These advances are reviewed in later chapters.

38

the cerebellum

Figure 17 Induction of LTD in a slice of the cerebellum.

(A) Intracellular recording from a Purkinje cell in a slice of the mouse cerebellum. 1, EPSPs evoked by double shock stimulation of parallel fibers (2PF). 2, Five Ca2+ -spikes induced by application of a membrane depolarizing current pulse (md). 3, Similar to 2, but initial two Ca2+ -spikes were driven by 2PF (at upward arrows) superimposed on an md. 4, An averaged single shock-evoked parallel fiber EPSP recorded before (x) and after (y) conjunction of PF stimulation and an md at 1 pulse/second for 5 minutes to bring on LTD (x minus y). 5, A record similar to that in 4 except for the stimulation being restricted to before and after 2PF. 6, A record similar to 5 except for the stimulation being restricted to before and after an md. (B) Time course of LTD in two mouse strains (C57BL WT and C3H WT). Abscissa, time in minutes (min) relative to onset of stimulation. Ordinate, relative rising rate of Purkinje cell EPSP responses to PF stimulation. For both mouse strains, plots are shown for control 2PF and md stimulation versus conjunction of these stimuli. x and y, time for recording of the traces x and y in A4. In brackets, number of tested cells for each stimulating condition. (From Le and Ito, unpublished material.)

3-5 Adaptive Control In the 1970s, functional roles of the cerebellum were under intense discussion. Even though the role of the cerebellum in the control of body equilibrium, the finger-nose test, and arm retraction had been proposed in classic studies, mechanisms underlying these roles seemed too complex to analyze experimentally. Jun Fukuda,

3 • The Cerebellum as a Neuronal Machine

39

Stephen Highstein, and I searched for a simple system to analyze. We found that the vestibuloocular reflex (VOR) was an appropriate experimental model. In response to a head movement sensed by the vestibular organ, the VOR produces an eye movement to maintain stable retinal images during head movement. We first showed that the VOR was inhibited directly by Purkinje cells located in the flocculus, that is, in the phylogenetically oldest part of the cerebellum (Fukuda et al., 1972; Kawaguchi, 1985). This finding was the beginning of our flocculus hypothesis for VOR adaptation (Ito, 1972, 1974, 1982) and continuing debate about its mechanism, as will be introduced in Chapters 10 and 12. Importantly, VOR is a feedforward control system that has no feedback from output to input; that is, there is no way to inform the vestibular system directly about eye movement (Ito, 1974). In engineering systems, feedforward control alone is undesirable because without feedback, the control cannot be precise. However, in biological systems, feedback may not always be available. In such cases, another CNS pathway (or pathways) is needed to replace the traditional feedback loop. We reasoned that this could be the flocculus. Our assertion was supported by the observation that the vestibulospinal reflex held head position constant by using direct feedback from the neck’s position to the vestibular organ but using no cerebellar inhibition. For the VOR to obtain precise compensatory eye movements without feedback, there had to be a visual pathway to the flocculus that informed about errors in the operation of the reflex (Ito, 1970). To test this prediction, the late Kyoji Maekawa (1929–1990) and John Simpson (Maekawa and Simpson, 1973) indeed discovered in my laboratory a powerful climbing fiber projection from the retina to the flocculus. Input to the flocculus from the vestibular organ via mossy fibers had already been shown in the cat (Brodal, 1972). In summary, our VOR model incorporated a set of three elements of the Marr-Albus model: mossy fiber-parallel fiber input, climbing fiber input, and Purkinje cell output. Also, the VOR was testable in a behaving animal! The XXVth Congress of the IUPS was held in Munich in 1971. It was an unforgettable experience for me. I reported about the direct inhibition of VOR relay neurons by flocculus Purkinje cells and proposed that the flocculus plays a key role in the feedforward control of the VOR. To my great surprise in this same session, Geoffrey Melvill Jones reported that when a human subject wore Dove-prism goggles, which reversed the right-left relationship in the visual field for one month, the result was a clear-cut depression and final reversal of the VOR (Gonshor and Melvill Jones, 1974). David Robinson, a world-renowned oculomotor physiologist/bioengineer, was leading discussions in this session. After the Congress, he attached Doveprism goggles to a kitten and showed that the VOR was substantially depressed

40

the cerebellum

(Robinson, 1976). Moreover, he showed that the depression did not occur when the flocculus had been lesioned bilaterally. I learned the “double rotation” technique (use of vestibular and visual stimuli in various combinations) from an otologist, who was working on vestibular functions in race-car drivers, and applied it to rabbits. We oscillated them sinusoidally on a horizontal turntable and also moved a surrounding screen horizontally (Ito et al., 1974). When the screen was rotated in the direction opposite to the turntable rotation, the VOR was gradually enhanced (i.e., to catch up with the increased relative movement of the screen). Likewise, while the screen was rotated in the same direction as the turntable, the VOR was gradually depressed. These adaptive changes in the VOR were abolished when the flocculus was ablated bilaterally (Batini et al., 1979) or when climbing fibers were lesioned bilaterally (Ito and Miyashita, 1975). We then proceeded to record from flocculus Purkinje cells during VOR adaptation (Ghelarducci et al., 1975; Dufosse et al., 1978). Since that time, numerous such studies have been carried out in many laboratories, but nonetheless, VOR adaptation remains a valuable system for investigating mechanisms of cerebellar motor control and such work still generates new issues in cerebellar research (Chapter 10). A frequently discussed question in the late 1960s and early 1970s was what signals climbing fibers conveyed as a set of unique afferents to the cerebellum. Marr (1969) assumed that they provided instruction signals from the cerebral cortex, whereas Albus (1971) thought that climbing fiber input implied errors in the simple perceptron-like operation of a cerebellar network. Miller and Oscarsson (1970) proposed that the inferior olive acted as a comparator between command signals from higher centers and the activity these signals evoked at lower levels. I proposed that climbing fibers monitored “control errors” for the VOR (Ito, 1970). Amat (1983) observed in the frog cerebellum that climbing fibers responded to a shift in the position of a forelimb and suggested that these responses represented a deviation of the forelimbs from a predetermined position. Since then, the signal contents of climbing fiber discharges have been investigated extensively. It seems to be a general principle that climbing fiber signals encode errors of some sort; not always an error occurring as a consequence of a movement, but also an error generated intrinsically within a neuronal circuit (Chapter 13, “Voluntary Motor Control”). When climbing fibers convey error signals, LTD would be induced in those parallel fiber-Purkinje cell synapses that are involved in erroneous performance. Learning would then occur to reduce such incorrect behavior (i.e., “error learning”). This notion has been expanded to motor learning in general, and it is sometimes called Marr-Albus-Ito hypothesis.

3 • The Cerebellum as a Neuronal Machine

41

3-6 Cerebellar Internal Models Also about 40 years ago, I speculated about the function of a then well-known anatomical structure of the cerebellum, the cerebrocerebellar loop that links the primary motor cortex and the intermediate part of the cerebellar hemisphere (Figure 18). Initially, I failed to relate this loop to the circuit structure I had proposed for VOR adaptation. Therefore, I then introduced the idea that the cerebellum provided an internal model that helped the cortical controllers. The idea was as follows. In performing unskilled voluntary movements, the initial instruction arising from an association area of the cerebral cortex would be transferred to the primary motor cortex and then through the pyramidal tract down to the spinal motor centers. The final outcome would be checked through sensory pathways by the association cortex, there being a large negative feedback loop formed through the external world. In this case, the cerebral cortex had to be continuously aware of what was being performed and had to be available for adjusting the performance from time to time. As experience was gained in the performance of these movements, they would become refined to the level of being skilled voluntary movements. As the learning process progressed, it was suggested that the large loop through the external world would be effectively replaced by an internal loop passing through the cerebellum, such that it would serve as a model simulating the combination of the spinal control system, the external world, and the sensory pathways. In this gestalt, the original negative feedback system would be converted by learning into a feedforward system that needed no straightforward negative feedback from the output to the input. I submitted an invited manuscript on this idea of an internal model to the 4th Symposium of the Fulton Society on the Cerebellum, which was held in New York City in 1969. Unfortunately, an illness prevented me from attending the meeting, but the manuscript was nonetheless circulated among the participants and eventually published in a journal that collected publications concerning that meeting (Ito, 1970). I also presented the idea in my 1984 monograph, The Cerebellum and Neural Control. In the 1980s, movements of multijoint robotic fingers, arms, and hands became a challenging control task because such movements have a large number of degrees of freedom (Chapter 13). Hollerbach (1982) and An et al. (1988) introduced a clever way of controlling a robot’s arm using feedforward control via an inverse model of the arm. In 1987, Mitsuo Kawato and his colleagues proposed an ingenious two-degrees-of-freedom control, in which the feedback control by the primary motor cortex was combined with feedforward control by the cerebellum (Figure 8B). If the cerebellum represented the output-input relationship of the controlled object, this inverse model could play the role of a feedforward controller. For this system, Kawato et al. (1987) incorporated an ingenious way of

42

the cerebellum

learning, that is, “feedback error learning” that derived errors from the primary motor cortex performing its feedback control. The Kawato model seemed to be an effective way to explain the learning process in voluntary motor control. Initially, the primary motor cortex would exert feedback control to perform accurate movements. Meanwhile, the cerebellar inverse model would gradually be modified by error signals to provide precise feedforward control. Then the feedback control by the primary motor cortex would be replaced by feedforward control from the cerebellum unless the latter happened to be inaccurate. It could be reasoned that the initial feedback control was performed consciously, whereas the later feedforward control by the cerebellum is performed unconsciously, this idea being in good general agreement with our daily experiences. Moreover, a combination of forward and inverse models was applied successfully to the creation of a robot that was able to learn movement skills (Wolpert and Kawato, 1989). A major advantage of Kawato’s control system model was its computational expression, such that it could be installed in a robot that was capable of learning complex movements. The biological validity of the forward and inverse models is now being tested in an ever-increasing number of experimental studies on Purkinje cell discharges during various movement paradigms (Chapter 15, “Internal Models for Voluntary Motor Control”).

Figure 18 The cerebrocerebellar loops.

This figure schematizes the loop connections between the cerebral cortex and the cerebellum. Abbreviations: CC. cerebellar cortex; CF, climbing fiber; CN, cerebellar nucleus; IO, inferior olive; MF, mossy fiber; NRTP, nucleus reticularis tegmenti pontis; PC, Purkinje cell; PN, pontine nucleus; Py, pyramidal cell in the cerebral cortex; RNp. parvocellulr red nucleus; VL, ventrolatertal thalamic nucleus.

3 • The Cerebellum as a Neuronal Machine

43

3-7 Cognitive Functions of the Cerebellum A major new development in cerebellar research began in the early 1990s. A role for it was postulated in cognitive functions, such as language acquisition, on the basis of the considerable expansion of the most lateral part of the cerebellar hemispheres in humans (Leiner et al., 1993; for the general line of thought at that time, see Leiner, 2010). To support this hypothesis, an anatomical study revealed that area 46 in the cerebral prefrontal cortex formed a cerebrocerebellar communication loop with the D2 zone (Kelly and Strick, 2003). On the other hand, area 46 had been proposed to perform an executive role in initiating a thought and to be associated with the will to think (see Fuster, 1997). Clinical observations had also shown that lesions in the cerebellum often accompanied mental and affective disorders showing characteristic symptoms, which Schmahmann (1991) called mental dysmetria. Even more recently, a wealth of brain imaging data has demonstrated activation of the cerebellum during the execution of various cognitive tasks (Chapter 17, “Cognitive Functions”). Almost two decades ago, the late Robert Dow (1908–1995) asked me to write an article on the cerebellum for a special issue of Trends in Neuroscience. I reflected in my article on how cerebellar neuronal circuits might process information not only for body movements in the physical domain but also for conceptual functions in the cognitive domain (Ito, 1993b). I was inspired with an idea that in the control systems’ perspective, the movement of a body part was analogous to manipulation of a mental model, like those proposed by Craik (1943) and JohnsonLaird (1983), and even Piaget’s (1951) schema (Chapter 1), which could be represented in the temporoparietal cortex. It seemed to me that the cerebellum might form an internal model of such a mental model or Piaget’s schema, just as the cerebellum formed an internal model of limb movement. My hypothesis was supported at that time by very little experimental evidence, but it explained a number of phenomena, which were otherwise quite puzzling (Chapter 17).

3-8 Summary The history of research on the cerebellum is certainly replete with excitement and thrilling experimental possibilities. Over the past five decades, in particular, basic concepts of synaptic plasticity, error learning, adaptive control, and model-based control have been formulated and substantiated experimentally. This has changed the once widely held belief that the function of the cerebellum was strictly for a relatively simple form of motor control to the current idea that it is an elaborate neuronal machine equipped with learning capabilities and devoted to far-moreadvanced forms of systems control for posture and movement and probably also for participation in the control of complex motor actions and cognitive functions.

4 Input and Output Pathways in the Cerebellar Cortex

4-1 Introduction We are now ready to begin decomposing neuronal circuits in the cerebellar cortex into component neurons and examining them one by one. This exercise provides the basis for considering principles operating in these circuits and also for testing the validity of thus-far-derived hypotheses, which are discussed in later chapters. The focus in this chapter is on the neuronal elements that comprise the input and output interactions with the cerebellar cortex via mossy fibers, granule cells, and Purkinje cells. Unipolar brush cells and beaded fibers are also considered.

4-2 Mossy Fibers Mossy fibers are the most numerous afferent fibers that reach the cerebellar cortex through the white matter and terminate in the cerebellum’s granular layer, forming a mosslike structure (Figure 14). Some mossy fibers originate as sensory peripheral nerves, but most mossy fibers arise from neurons located within the spinal cord and brainstem. They also arise from unipolar brush cells in the granular layer (see below) and from cerebellar nuclei (Chapter 6, “Pre- and Post-Cerebellar Cortex Neurons”). Mossy fibers terminate within a glomerulus, which forms a characteristic rosette structure. Within a glomerulus, granule cell dendrites receive α-amino3-hydroxy-5-methyl-4-isoxazolone propionic acid (AMPA)-mediated excitatory synapses from a mossy fiber terminal. Granule cells receive also inhibitory synapses supplied by a Golgi cell axon terminal. Descending dendrites of mostly deep Golgi cells also receive excitatory synapses directly from a mossy fiber terminal. Most mossy fibers release glutamate as a transmitter, but some in the vestibulocerebellum release acetylcholine (Barmack et al., 1992a,b; Jaarsma et al., 1997). 44

4 • Input and Output Pathways in the Cerebellar Cortex

45

Both AMPA and N-methyl-D-aspartate (NMDA) receptors mediate mossy fibergranule cell synapses in glomeruli (Traynelis et al., 1993). Within a glomerulus, NMDA receptor subunits (NR1, NR2A, and NR2C) are co-located between the centrally positioned rosette structure and the peripherally positioned, tiny Golgi cell axon terminals at the postsynaptic junction with granule cell dendrites (Yamada et al., 2001). A marked “spillover” phenomenon has been reported to occur in both glutamate released from a mossy fiber terminal and gamma-amino-butyric acid (GABA) released from Golgi cells. Single AMPA-receptor-mediated excitatory postsynaptic currents (EPSCs) or potentials (EPSPs) at the mossy fiber-granule cell connection are mediated by both the direct release of glutamate and the rapid diffusion of glutamate from neighboring synapses. Spillover currents contribute about one-half of the synaptic charge and improve transmission efficacy by increasing both the amplitude and duration of EPSPs (DiGregorio et al., 2002). Fluctuation analysis indicates that these indirect release sites are at least fourfold more numerous than those directly connected to the postsynaptic cell. As a result, spillover is predicted to improve the reliability and reduce the variability of transmission at this glomerular synapse. The unique firing behavior of granule cells may also be relevant; a single impulse in a mossy fiber tends to induce bursting spikes in a granule cell (Chadderton et al., 2004).

4-3 Granule Cells Granule cells are individually the smallest (soma diameter, 5–8 micrometers (μm)) and the most numerous neurons in the CNS (Braitenberg and Atwood, 1958; Zagon et al., 1977). A large divergence and a small convergence characterize the mossy fiber-granule cell pathway. Each granule cell receives mossy fiber terminals via only four to five excitatory synapses (Eccles et al., 1967; Chadderton et al., 2004). The functional significance of this small convergence number will be considered later in Chapter 8, “Multiplicity and Persistency of Synaptic Plasticity.” In contrast, one mossy fiber supplies excitatory synapses to 400–600 granule cells in a folium and probably more when the branches of a mossy fiber reach two or more folia. The efficacy of synaptic transmission from a mossy fiber to granule cells may vary probabilistically from glomerulus to glomerulus. Such relative efficacy may also be affected by the following three factors: activity-dependent induction of long-term potentiation (LTP) (Chapter 8), enhancement of intrinsic excitability (Armano et al., 2000), and Golgi cell inhibition (Chadderton et al., 2004; see also below). Parallel fiber axons of granule cells run along the folia of the cerebellar surface after ascending vertically from the granular layer to the molecular layer and then

46

the cerebellum

bifurcating into two parallel fiber collaterals. The formation of parallel fibers is controlled genetically. This is known because in Pax6 mutant rats, granule cells in the external germinal layer fail to form parallel fiber axons (Yamasaki et al., 2001). In normal animals, the length of a parallel fiber from terminal to terminal across its T-junctions has been reestimated to be as long as 4–6 mm (Mugnaini, 1983; Harvey and Napper, 1988; Pichitpornchal et al., 1994). Optical recording in mice shows that the local stimulation of a parallel fiber bundle excites Purkinje cells along the bundle over a distance of more than 3 mm (Coutinho et al., 2004). This extent of excitation was also observed by optical recording in neonatal rats on postnatal day 5, although it reduced to 1.5–2 mm at postnatal days 6–7 (Arata and Ito, 2004). Ascending segments of granule cell axons form synapses with spines, which are located exclusively on the smallest diameter distal regions of Purkinje cell dendrites (Gundappa-Sulur et al., 1999; Lu et al., 2009). This contrasts to parallel fibers, which form synapses on the intermediate or large diameter regions of spiny branchlets, as well as the smallest diameter distal regions. The ascending segments form about 20% of the granule cell-Purkinje cell synapses. A differential stimulation of parallel fibers and ascending segments of granule cell axons in cerebellar slices revealed substantial differences in the properties of EPSCs generated in Purkinje cell dendrites (Sims and Hartell, 2005). Ascending segment synapses release a transmitter with a higher mean release probability and larger mean quantal amplitude than parallel fiber synapses, and they do not exhibit LTD (Chapter 7, “Conjunctive Long-Term Depression (LTD)”). These different properties of parallel fiber versus ascending segment synapses suggest that they have different roles in Purkinje cell function.

4-4 Unipolar Brush Cells Unipolar brush cells of unique morphology are located primarily in the granular layer of the vestibulocerebellum. This portion of the cerebellum, which corresponds roughly to the flocculonodular lobe, receives primary vestibular afferents in the form of mossy fibers. These cells receive excitatory synapses on their dendritic “brush” from a single mossy fiber terminal (Color Plate VI). This connection has the form of a giant glutamate-mediated synapse (Diño et al., 1999). The unipolar brush cell’s axon forms branches within the granular layer, which give rise to large terminals that synapse with both granule cell and unipolar brush cell dendrites. This arrangement is within glomeruli that resemble those formed by extrinsic mossy fibers. Hence, unipolar brush cells are an intracortical source of mossy fibers. Unipolar brush cells receive inputs from glutamate-mediated primary vestibular fibers and choline-acetyltransferase-positive mossy fibers. Some of the latter

4 • Input and Output Pathways in the Cerebellar Cortex

47

originate from the medial and descending vestibular nuclei (Diño et al., 2001). An excitatory effect of muscarine, but not nicotine, was detected in ~15% of granule cells tested in the vestibulocerebellum (Takayasu et al., 2003). Evidence suggests that this effect is caused by the inhibition of an intrinsic outward K+ current via the activation of muscarinic M3 receptors. Two subtypes of unipolar brush cells have been distinguished: one expresses calretinin, and the other expresses metabotropic glutamate receptor type 1a (mGluR1a) (Nunzi et al., 2002). Both subtypes express glutamate receptor subunit 2 (GluR2) (Sekerková et al., 2004). Tbr2/Eomes, a T-domain transcription factor (Tbr2), has been considered to be a specific marker of both subtypes of unipolar brush cells in the adult and developing cerebellum (England et al., 2006) (Color Plate VII). Unipolar brush cells express NMDA, kainite, and AMPA receptors in the synaptic membrane. They also express metabotropic glutamate receptors (mGluR1 and mGluR2/3) on the perisynaptic and extrasynaptic parts of the spiny appendages of dendrites (Jaarsma et al., 1995, 1998; Billups et al., 2002). Mossy fiber impulses induce an AMPA-mediated fast EPSC and a predominantly NMDAmediated slow EPSC in unipolar brush cells (Rossi et al., 1995). It has been suggested that unipolar brush cells may amplify mossy fiber inputs in the vestibulocerebellum (Kalinichenko and Okhotin, 2005; Barmack and Yakhnitsa, 2008).

4-5 Purkinje Cells Purkinje cells are the largest neurons in the cerebellum, extending magnificent planar dendrites to receive numerous synaptic inputs (Color Plate VII). Purkinje cells mediate the sole outputs of the cerebellar cortex, which are exclusively inhibitory in action upon their target neurons. Parallel fibers form excitatory synapses on dendritic spines of Purkinje cells. The synaptic membrane, lined with postsynaptic density (PSD), is located on the side (but not top) of a spine head and is therefore located at an optimal distance from the endoplasmic reticulum that protrudes to the spine head (Launey et al., 2004) (Color Plate VIII). A large divergence and an enormous convergence characterize the parallel fiber-Purkinje cell connection. While a single parallel fiber extends for ~3 mm (i.e., ~1.5 mm on each side of the T-junction), it passes through the dendrites of ~450 Purkinje cells and thereby forms synaptic contacts with the dendritic spines of at least 300 Purkinje cells (Eccles et al., 1967). On the other hand, the number of parallel fibers making synaptic contacts with the dendritic arborization of a Purkinje cell can be as large as 180,000 in the human (Fox and Bernard, 1957) or either ~60,000–80,000 (Palay and Chan-Palay, 1974) or ~175,000 (Napper and Harvey, 1988a,b) in the rat. Note that parallel fibers form synaptic contacts with only ~54% of the Purkinje cells through whose dendritic arborization they pass. Simultaneous whole-cell recording

48

the cerebellum

from synaptically connected granule and Purkinje cells in cerebellar slices revealed that an impulse from a single granule cell evoked a fast EPSC of 2–60 pA in a Purkinje cell (Barbour, 1993). This suggests that ~50 simultaneously active granule cells are sufficient to excite a single Purkinje cell. Parallel fiber impulses release glutamate as a transmitter, which evokes two pharmacologically distinct types of synaptic potential in Purkinje cells. One is mediated by AMPA receptors, and the other by mGluR1. AMPA-mediated EPSPs are fast and evoked individually by each granule cell’s impulses (Figure 15C), whereas mGluR1-EPSPs are slow and observed only after a brief tetanus of parallel fibers (8 pulses at 50 Hz) in the presence of an AMPA receptor antagonist (Batchelor and Garthwaite, 1993) (Figure 15D). Slow EPSPs are accompanied by an increase in intradendritic sodium concentration, but the mechanism underlying this excitation remains unclear. The above two types of EPSP have different frequency characteristics. For example, following a single-shock stimulation of parallel fibers, fast EPSPs predominate, whereas in a burst stimulation of parallel fibers, slow EPSPs are facilitated. When a parallel fiber bundle is repetitively stimulated with 10 pulses at 100 Hz, the mGluR1- and AMPA receptor-mediated activations of Purkinje cells are equally potent (Coutinho et al., 2004). Metabotropic GABAB receptors are expressed in the extra-postsynaptic sites of parallel fiber-Purkinje cell synapses. The activation of GABAB receptors leads to the augmentation of mGluR1-mediated parallel fiber-Purkinje cell transmission (Hirono et al., 2001). This is an interesting case of interaction between two types of metabotropic receptor. Purkinje cell outputs from the cerebellar cortex inhibit their target neurons with GABA as the transmitter. Because Purkinje cells provide ~73% of the total synapses of cerebellar nuclear neurons, including almost all of the somatic synapses of cerebellar nuclear neurons (Palkovits et al., 1977; De Zeeuw and Berrebi, 1995), the question arises as to how such inhibitory inputs accurately control spiking in the latter neurons. To answer this question, Gauck and Jaeger (2000) applied the dynamic clamp method, in which they injected a conductance waveform that simulated the synaptic input of several hundred GABA A-type inputs to a cerebellar nuclear neuron in in vitro slices. They found that the time of inducing individual spikes was controlled precisely by brief decreases in inhibitory conductance, these being the consequence of the synchronization of many inputs. They also showed that spike rate was controlled linearly by the discharge rate of inhibitory inputs. Purkinje cells project recurrent axon collaterals and thereby inhibit each other. These collaterals extend to neighboring Purkinje cells within ~300 micrometers of the parent cell (Hawkes and Leclerc, 1989; O’Donoghue and Bishop, 1990). Axon

4 • Input and Output Pathways in the Cerebellar Cortex

49

collaterals of Purkinje cells also inhibit basket cells, which, in turn, inhibit Purkinje cells. Therefore, Purkinje cells may be involved in a mixed reciprocally inhibitory network containing both Purkinje cells and basket cells.

4-6 Climbing Fibers Climbing fibers are a unique structure of the cerebellum with no homolog elsewhere in the CNS (Color Plate IX A–B). The major transmitter of climbing fibers is glutamate. Each Purkinje cell is innervated by one climbing fiber. This is a consequence of the postnatal elimination of multiple innervation, which, after birth in rats and mice, attains its maximum in one week and fades out in two weeks via its interaction with developing parallel fiber-Purkinje cell synapses (Mariani and Changeux, 1981; Hashimoto and Kano, 2003; Scelfo and Strata, 2005; Hashimoto et al., 2009). Each climbing fiber forms numerous synaptic contacts with the dendrites of a single Purkinje cell [~1,300 in proximal dendrites of rat Purkinje cells (Strata, 2002), but a much larger number,~26,000, is derived from the density ratio of climbing fiber to parallel fiber synapses (Nieto-Bona et al., 1997)]. The above arrangements for climbing fibers result in a particularly large EPSP in Purkinje cells superimposed with Ca2+ spikes (Llinas and Sugimori, 1980a,b). Extracellular recording has revealed that Purkinje cells spontaneously generate two different types of spike: simple spikes (Figure 15E, F) and complex spikes (E, G). In intracellular recording, stimulation of parallel fibers cells elicits simple spikes (Figure 15A), whereas climbing fiber stimulation evokes complex spikes (Figure 15B). Simple spikes are actually Na2+ spikes generated in the somatic region that spread passively into the dendrites, whereas complex spikes involve Ca2+ spikes generated in dendrites. In in vivo conditions, simple spikes discharge spontaneously at a rate of 50–100 Hz, whereas complex spikes discharge at an irregular, low rate of ~1 Hz (Thach, 1967). The unique role of climbing fibers in inducing synaptic plasticity in Purkinje cells is dealt with in Chapter 7. Because of the powerful depolarizing action accompanying Ca2+ entry, it has been suggested that climbing fiber responses also play a critical role in cellular function. Indeed, in rat cerebellar slices, climbing fiber discharges occurring at physiological frequencies (0.4–10 Hz) substantially modified the frequency and pattern of simple spike discharges (McKay et al., 2007). Repetitive climbing fiber discharges converted a spontaneous pattern of simple spike discharges into a more natural nonbursting pattern that consisted of simple spike trains interrupted by short climbing fiber-evoked pauses or longer pauses associated with state transitions. These effects were reproduced by injecting currents simulating complex spike depolarizations in the presence of synaptic blockers.

50

the cerebellum

Hence, these appeared to occur intrinsically—for example, by activation of Ca2+dependent K+ channels. In regard to the function of climbing fibers in cerebellar circuits, recent studies have revealed unexpectedly that climbing fibers also excite interneurons in the cerebellar cortex via atypical transmission mechanisms, as explained in Chapter 5, “Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex.” In brief, such transmission might be mediated by a spillover of glutamate released from climbing fiber terminals (Szapiro and Barbour, 2007), which may then spread to interneurons via volume transmission (Agnati et al., 1995). An alternative mechanism would be for climbing fibers to activate synaptically NG2+ glial cells, which could, in turn, excite interneurons (Lin et al., 2005).

4-7 Beaded Fibers The cerebellar cortex receives not only mossy fibers and climbing fibers, but also beaded fibers, which contain various amines, such as serotonin, norepinephrine, or histamine, or neuropeptides, such as angiotensin II or orexin (Haines and Dietrichs, 1984; Haines et al., 1984; Airaksinen and Panula, 1988; King et al., 1992; Onat and Cavdar, 2003; Zhu et al., 2006; Ito, 2009). The beaded fibers extend fine varicose axonal fibers sparsely throughout the granular and molecular layers to form direct contacts with Purkinje cells and other cerebellar neurons. These axonal fibers are often called the third type of cerebellar afferent. On the basis of their diffuse extensions, it is considered that this third type of afferent does not convey specific information to the cerebellar cortex. Rather, its role could be modulatory. Akin to stomatogastric ganglia (Marder et al., 1986), such neuromodulation would set the activity level or switch the operational mode of a cerebellar microcomplex (Chapter 9, “Network Models”) to match a behavioral demand (Schweighofer et al., 2004) (for further description, see Chapter 6).

4-8 Summary The mossy fiber-granule cell-Purkinje cell pathway provides the core of cerebellar cortical neuronal circuits. Unipolar brush cells appear to amplify the mossy fiberto-granule cell transmission, but their special need in the vestibulocerebellum is unclear. This pathway, together with climbing fiber and beaded fiber afferents, forms the skeleton of the cerebellar neuronal circuits. Other types of neurons and glial cells put flesh on this skeleton to achieve the elaborate functional mechanisms of the cerebellum.

5 Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex

5-1 Introduction The cerebellar cortex contains two types of inhibitory neurons in the molecular layer; basket and stellate cells. In physiological experiments, basket cells are not always distinguishable from stellate cells on the basis of their responses during recording; hence, they are often lumped together as “basket/stellate cells” or collectively called inhibitory interneurons in the molecular layer. The granular layer, on the other hand, contains large and small inhibitory neurons. Large neurons are the Golgi and Lugaro cells, whereas small neurons are “small Golgi” cells, “smallfusiform Lugaro” cells, and “small globular” neurons. The latter three types of neuron have recently been identified on the basis of their characteristic morphology and location. In this chapter we consider recent knowledge about these neurons. We also consider Bergmann glial cells and NG2+ glial cells as important elements of cerebellar neuronal circuits.

5-2 Basket Cells and Stellate Cells Basket cells are middle-sized neurons located deep in the molecular layer close to the Purkinje cell layer, whereas stellate cells are smaller cells dispersed through the molecular layer (Color Plate IV). Basket cells supply inhibitory synapses to the “bottleneck” of a Purkinje cell soma where they form a unique complex structure called a “pinceau.” It can be labeled specifically by monoclonal antibodies raised using Xenopus oocytes as immunological vectors (Tigyi et al., 1990). On the other hand, stellate cells supply inhibitory synapses to Purkinje cell dendrites. A bundle of parallel fibers forms a synaptic contact with not only the dendrites of Purkinje

51

52

the cerebellum

cells but also those of basket and stellate cells, which in turn supply GABA-mediated inhibitory synapses to Purkinje cells. The somata of Purkinje cells and stellate cells are immunopositive for GABA but not for glycine (Reichenberger et al., 1993). Basket and stellate cells mediate the feedforward inhibition of Purkinje cells supplemental to the direct parallel fiber-Purkinje cell pathway. In the C3 forelimb zone, it has been shown that basket and stellate cells share the same receptive field with Purkinje cells located in the same microzone (see Chapter 9, “Network Models”) (Ekerot et al., 1995). Parallel fiber-basket/stellate cell synapses are mediated by both AMPA and NMDA receptors, which produce prolonged (for hundreds of milliseconds) EPSCs in response to the burst stimulation of parallel fibers, the latter being presumably induced by the spillover of the transmitter, glutamate (Carter and Regehr, 2000). Parallel fiber-basket/stellate cell synapses are also mediated by mGluR1a (Karakossian and Otis, 2004). As in Purkinje cells, the activation of mGluR1s induces slow EPSCs. During low-frequency transmission, also as in Purkinje cells, basket/stellate cells are predominantly activated via AMPA receptors, whereas mGluR1s are recruited during high-frequency transmission. Basket cells extend axons perpendicular to parallel fibers and cover an area containing ~10 × 7 rows of Purkinje cells, with a probable divergence number of ~50 (Eccles et al., 1967). Twenty to thirty basket cell axons may converge onto one Purkinje cell, although this figure is not particularly accurate. The middle band of parallel fiber-excited Purkinje cells flanked by side bands of basket/stellate cellinhibited Purkinje cells constitutes a spatial pattern of lateral inhibition. This can be visualized in the cerebellar cortex following electrical stimulation of a selected number of parallel fiber bundles (Coutinho et al., 2004). However, because such a lateral inhibition pattern has not been observed to appear spontaneously, its significance under natural conditions remains uncertain. Activation of basket/stellate cells induces powerful IPSPs in Purkinje cells (Eccles et al., 1967). Basket cells receive collaterals of climbing fibers and also those of Purkinje cell axons (Palay and Chan-Palay, 1974; Jeager et al., 1988). Indeed, climbing fiber responses evoked in Purkinje cells are accompanied by EPSPs in basket cells at about the same latency. These EPSPs in basket cells are followed by IPSPs with a delay of ~1 millisecond and constitute EPSP-IPSP responses, which display oddly the all-or-none property at their threshold stimulus intensity (O’Donoghue et al., 1989). The possible source of the so-observed unitary IPSP is a Purkinje cell and its recurrent axonal collaterals because it has been confirmed electronmicroscopically that each basket cell receives somatic inputs from only one Purkinje cell. On the other hand, EPSPs were shown to occur in basket cells at about the same latency as the climbing fiber responses evoked in Purkinje

5 • Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex

53

cells, presumably being mediated by climbing fiber collaterals. However, no direct synaptic contact between climbing fibers and basket/stellate cells has been confirmed. Instead, it is now apparent that climbing fibers form synaptic contact with NG2+ glial cells via Ca2+-permeable AMPA receptors (Lin et al., 2005; see below). It is also apparent that climbing fiber signals activate interneurons in the molecular layer by a spillover of glutamate from their terminals, this being quite unlike typical synaptic transmission (Szapiro and Barbour, 2007). Purkinje cells in vivo discharge spontaneously at a highly irregular rate (Eccles et al., 1967). This discharge is not caused by excitatory synaptic drives but, unexpectedly, by tonic influences from inhibitory interneurons (Hausser and Clark, 1997). Indeed, the pharmacological blockade of ionotropic and metabotropic glutamate receptors does not affect spontaneous Purkinje cell discharges, whereas the blockade of GABAA receptors increases their rate and regularity. It thus appears that the tonic discharges of inhibitory interneurons modulate the spike discharges of Purkinje cells. Similar irregular discharges, and their dependence on tonic inhibition, have also been observed in inhibitory interneurons that receive tonic inhibition from themselves. Simulation studies using realistic models of Purkinje cell properties reproduce these irregular discharges and suggest that they are caused by endogenous tonic inhibitory current rather than excitatory current (De Schutter and Bower, 1994; De Schutter, 1999; Jaeger and Bower, 1999). The comprehensive functional meaning of this irregular spontaneous discharge in Purkinje cells and inhibitory interneurons is still unclear. A modeling study suggested, however, that the irregularity helps Purkinje cells and inhibitory interneurons to respond rapidly, sensitively, and linearly to external inputs (Van Vreeswijk and Sompolinsky, 1996). Inhibitory interneurons in the molecular layer are reciprocally connected via inhibitory synapses (Kondo and Marty, 1998). They are also linked with each other through electrical synapses (Mann-Metzer and Yarom, 1999). A computer simulation suggested that reciprocal inhibition causes a 100–250 Hz oscillation in the activity of basket/stellate cells (Maex and de Schutter, 2005). Isope et al. (2002) drew attention to oscillations with characteristic frequencies between 150 Hz and 270 Hz. Interestingly, these were recorded far earlier on the cerebellar cortical surface by Edgar Adrian (1889–1977) (1935).

5-3 Golgi Cells Golgi cells are large neurons that extend their dendrites like a bouquet into the molecular layer and also extend their descending dendrites to the granular layer. Golgi cells have broadly branching axons in the granular layer (Color Plate X A). It

54

the cerebellum

is now apparent that Golgi cells exert extensive control of spatio-temporal signal organization and information storage in the granular layer (see D’Angelo, 2008). It is also apparent that they play unique roles in computational aspects of cerebellar function (Chapter 9). The molecular layer dendrites of each Golgi cell receive ~4,800 excitatory inputs from parallel fibers, and the same cell’s descending dendrites in the granular layer receive ~230 excitatory mossy fiber terminals (Pellionisz and Szentágothai, 1973). Through such inputs, each Golgi cell recorded in the C3 zone has a localized receptive field in the forelimb skin (Ekerot and Jörntell, 2001a,b). The major excitatory inputs from parallel fibers to Golgi cells are mediated by both AMPA and NMDA receptors (Dieudonné, 1998) and mGluR2 (Watanabe et al., 2003). Interestingly, parallel fiber-induced mGluR2 activation hyperpolarizes Golgi cells via G-proteincoupled inward K+ channels, as contrasted to the depolarizing action mediated by mGluR1 on Purkinje cells. mGluR2 is expressed also in the somata and axons of Golgi cells whose axon terminals can be activated by glutamate released from mossy fiber terminals (Ohishi et al., 1994). Golgi cells also receive inhibitory synapses from Lugaro cells (see below) but not from Purkinje cells or other Golgi cells. Morphological evidence suggests that climbing fibers might make contact with Golgi cells (Palay and Chan-Palay, 1974; Sugihara et al., 1999) but its functional significance is unclear (see Chapter 8, “Multiplicity and Persistency of Synaptic Plasticity”). A Golgi cell projects, in turn, a broadly branching axon and supplies GABAmediated inhibitory synapses (Chadderton et al., 2004) to ~5,700 granule cells (in cats, Palkovits et al., 1971). Characteristically, Golgi cell-granule cell synapses produce significant cross-talk with non-postsynaptic cells. This is caused by the spillover of the inhibitory transmitter, GABA. This effect is manifested as a slowly rising and decaying, small-amplitude IPSC. GABA spillover within the mossy fiber glomerulus may be promoted both by the anatomical specialization of this glomerular synapse and by the presence of the high-affinity α6-subunit-containing GABAA receptor in granule cells. GABA spillover may also play a role in regulating the number of active granule cells in the cerebellar cortex (Rossi and Hamann, 1998). A study has shown that a major proportion (~70%) of Golgi cell somata are immunopositive to both glycine and GABA, and these two immunoreactivities are co-localized in the same glomeruli and even in the same Golgi cell terminals, as confirmed by electron microscopy (Ottersen et al., 1988). Hence, Golgi cells may release not only GABA but also glycine as an inhibitory transmitter. The selective ablation of Golgi cells by an immunotoxin-mediated cell targeting technique was shown to impair motor function (Watanabe et al., 1998). This effect is presumably attributable to attenuated GABA/glycine-mediated inhibition, as well as to the

5 • Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex

55

attenuated action of NMDA receptors on granule cells. A recent finding was that Golgi cells caused pure GABAergic inhibition of granule cells in contrast to pure glycinergic inhibition of unipolar brush cells (Dugué et al., 2005). This specialization resulted from the differential expression of GABAA and glycine receptors by target cells and not by a segregation of GABA and glycine in presynaptic terminals. These results exemplify the postsynaptic selection of co-released transmitters, this being a mechanism that enables target-specific signaling in mixed inhibitory networks. In the in vivo cerebellar cortex, Golgi cells discharge tonically at ~5 Hz to decrease the firing rate of granule cells via GABA synapses (Chadderton et al., 2004). This should help stabilize parallel fiber discharges despite large changes in mossy fiber discharge (Albus, 1971). Golgi cell inhibition decreases noise in parallel fiber discharges in the presence of noisy mossy fiber inputs (Philipona and Coenen, 2004). A computer simulation suggested that feedback inhibition from Golgi to granule cells induced 10–50 Hz oscillations in spike discharges from the latter (Maex and de Schutter, 2005). This may account for the large-amplitude oscillations recorded in the granular layer of freely moving rats (Hartman and Bower, 1998) and monkeys (Pellerin and Lamarre, 1997; Courtemanche et al., 2002, 2005). In rat cerebellar slices, Golgi cells display spontaneous firing at 1–10 Hz (at room temperature) or 2–20 Hz (at a bath temperature of 35°C–37°C). Because this firing persisted in the presence of various pharmacological blockers, Forti et al. (2006) proposed that Golgi cells behave under given conditions as pacemaker neurons with multiple ionic mechanisms. Moreover, the dendrites of Golgi cells are electrically coupled via gap junctions involving connexin-36. These gap junctions may explain how Golgi cells synchronize their firing in the absence of coincident synaptic input and how sparse, coincident mossy fiber input to Golgi cells triggers a mixture of excitation and inhibition in their discharge and its spike desynchronization (Vervaeke et al., 2010). The gap junctions may either promote network synchronization or trigger rapid network desynchronization, depending on the nature of the synaptic input.

5-4 Lugaro Cells The presence of Lugaro cells has long been known, but they have been characterized relatively recently as a unique type of inhibitory neuron located in the granular layer. Typically, their fusiform cell somata are located in or slightly below the Purkinje cell layer (Aoki et al., 1986; Sahin and Hockfield, 1990) (Color Plate X B). The categorization of Lugaro cells has recently been expanded to include cells with similar properties that are located throughout the granular layer (Sahin and Hockfield, 1990; Laine and Axelrad, 2002). Axons of Lugaro cells form a parasagittal

56

the cerebellum

plexus, but they also extend transversely for ~2 mm, forming contacts with the apical dendrites of Golgi cells. It is approximated that the cerebellar cortex contains one Lugaro cell and one Golgi cell for every 15 Purkinje cells. Moreover, the axons of more than 10 Lugaro cells converge onto only one Golgi cell, whereas the axon of one Lugaro cell diverges onto ~150 Golgi cells (Dieudonne and Dumoulin, 2000). Lugaro cells can be distinguished immunohistochemically from Golgi cells; that is, the latter express mGluR2 and somatostatin but not calretinin, whereas Lugaro cells express calretinin but not mGluR2 or somatostatin (Sahin and Hockfield, 1990; Geurts et al., 2001). A unique physiological feature of Lugaro cells is that in cerebellar slices, they are normally silent. They are, however, highly sensitive to serotonin, this being unlike basket cells, stellate cells, Purkinje cells, and Golgi cells. In the presence of serotonin, Lugaro cells discharge regularly at 5–15 Hz and induce IPSCs in Golgi cells (Dieudonne and Dumoulin, 2000). GABA-mediated inhibitory synapses between Lugaro cells and Golgi cells are unusually resistant to bicuculline, a commonly used GABAA antagonist, probably caused by an unusually long GABA dwell time and/or a high GABA concentration in synaptic clefts (Dean et al., 2003). Double immunocytochemical staining in the rat cerebellum has demonstrated that Lugaro cell axonal varicosities co-express GABAergic and glycinergic markers. Indeed, serotoninevoked activation of Lugaro cells induced IPSCs in Golgi cells, which were identified pharmacologically as mediated by both GABA and glycine. Golgi cells also exhibited spontaneously occurring IPSCs, which were purely GABAergic and were likely to arise via basket cell-Golgi cell synapses (Dumoulin et al., 2001). Because of the large divergence from Lugaro cells to Golgi cells (1:150), it appears that when activated by serotonin, Lugaro cells influence numerous Golgi cells. An interesting possibility is that Lugaro cells play a role in synchronizing activity among Golgi cells situated along a parallel fiber beam, as observed in anesthetized rats (Vos et al., 2000). Lugaro cells may switch the operation of Golgi cells from the individual rhythmic mode to the synchronous mode of discharge.

5-5 Small Inhibitory Neurons in the Granular Layer Small globular cells with globular somata are located at various depths in the granular layer (Laine and Axelrad, 2002). They extend three to four long-radiating dendrites that course through the three layers of the cerebellar cortex. Their axons project into the molecular layer and expand a local plexus, with a pattern similar to that of Lugaro cells. The axons of several small globular neurons project an axoncollateral that courses for a long distance in the transverse direction, immediately

5 • Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex

57

above Purkinje cell somata and parallel to parallel fibers. Cytochemical observations using mice expressing GFP-labeled glycinergic and GABAergic neurons have shown that small globular neurons are glycinergic/GABAergic inhibitory neurons lacking mGluR2 and neurogranin (Simat et al., 2007). Small globular neurons are characterized by their strong GABAergic inhibition received via axon collaterals of Purkinje cells and also by their activation to elicit spike discharges during the perfusion of monoamines, serotonin, or noradrenaline (Hirono et al., 2010). Small inhibitory neurons other than small globular neurons have not been examined well except for the findings that (1) they exhibit smaller and fewer spontaneous IPSCs and that (2) small fusiform Lugaro cells are only slightly responsive to noradrenaline.

5-6 Bergmann Glia Bergmann glial cells are astrocytes of unique morphology. Their cell somata are located among Purkinje cell somata, and their palisades (Bergmann fibers) extend to the molecular layer between Purkinje cell dendrites. A neuron-specific transmembrane protein, Delta/Notch-like EGF-related receptor (DNER), has been shown to play a role in the morphogenesis of Bergmann glia in the mouse cerebellum (Eiraku et al., 2005). In the developing cerebellum, DNER is highly expressed in Purkinje cell dendrites, which are closely associated with radial fibers of Bergmann glia expressing Notch. DNER deficiency retards the formation of radial fibers and results in an abnormal arrangement of Bergmann glia. Glutamate released from parallel fibers onto spines of Purkinje cell dendrites enters the glutamate-glutamine cycle. First, it is transported to Bergmann fibers that enclose the synapses of Purkinje cells and interneurons in the molecular layer, where it is converted to glutamine by a glial enzyme, glutamine synthetase. Glutamine itself has no action on glutamate or other receptors. It is transferred back to Purkinje cells where it is converted to glutamate by phosphate-activated glutaminase. Two major types of glutamate transporter, GLT1 and GLAST, are located in the plasma membrane of Bergmann fibers. Their highest density is in glial processes surrounding synapses between parallel fibers and Purkinje cell spines (Chaudhry et al., 1995). A Purkinje-cell-specific glutamate transporter, EAAT4, is expressed at a high density in the membrane of postsynaptic spines surrounding a synaptic cleft. EAAT4 is responsible for the vast majority of glutamate uptake by Purkinje cells. It regulates mGluR1 transmission and LTD (Auger and Attwell, 2000; Wadiche and Jahr, 2005). EAAT4 activity can be detected by recording glutamate transporter current from Purkinje cells. The tetanic stimulation of climbing

58

the cerebellum

fibers evokes a long-lasting potentiation of glutamate transporter current, probably implying an antiexcitotoxic adaptive response (Shin and Linden, 2005). A rather unexpected recent finding was that Bergmann glial somata express a relatively high density of AMPA receptors (in rat, 17-fold more than Purkinje cell somata), which are distributed uniformly throughout the Bergmann glial membrane. Quantal events were also observed in Bergmann glial cells. Their kinetics were as fast as those of neurons (Matsui and Jahr, 2003; Matsui et al., 2005). Bergmann glial cells were characterized by their tightly encapsulating excitatory synapses on Purkinje cells (Palay and Chan-Palay, 1974; Spacek, 1985). This arrangement makes it possible to locate glutamate transporters near their sites of release and to show an increase in the diffusion distance between adjacent synapses and thereby keep synapses functionally isolated from each other (Barbour and Hausser, 1997; Huang and Bergles, 2004). Moreover, recent findings suggest that AMPA receptors in perisynaptic Bergmann glial processes are activated by ectopic fusion of synaptic vesicles outside synaptic active zones (Matsui and Jahr, 2003; Matsui and Jahr, 2004) and also by spill out of glutamate from the synaptic cleft (Bergles et al., 1997; Dzubay and Jahr, 1999). Other studies have shown that stimulation of parallel fibers induces three types of currents in Bergmann glial cells: a fast AMPA receptor current, a slow glutamate uptake current that peaks in 5–10 milliseconds, and a slow G-protein-coupled current (Clark and Barbour, 1997; Bellamy and Ogden, 2005). The fast AMPA receptor current signaled the ectopic release of glutamate from parallel fibers. This implies a direct and rapid mechanism of neuron-to-glia communication that does not rely on transmitter spillover from synaptic clefts (Matsui and Jahr, 2004).

5-7 NG2+ Cells The molecular layer of the cerebellum contains a group of glial progenitors termed “oligodendrocyte precursor cells” (Palay and Chan-Palay, 1974; Levine and Card, 1987). They express chondroitin-sulfate proteoglycan (NG2). Such NG2+ cells are found throughout the developing and mature CNS, and they are responsible for generating oligodendrocytes and myelin throughout life. NG2+ cells in the molecular layer of the cerebellum are small (soma diameter, 10–15 micrometers), and they extend numerous processes among the dendrites of Purkinje cells. Recently, Lin et al. (2005) found that climbing fiber stimulation induced AMPA receptormediated synaptic currents in NG2+ glial cells located in the molecular layer. A quantal analysis suggested that a climbing fiber formed up to 70 discrete junctions with one NG2+ cell and that each NG2+ cell could be innervated by more than one climbing fiber. Electron microscopic analysis of physiologically identified NG2+ cells revealed that anatomically defined climbing fibers formed direct

5 • Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex

59

synapses with the processes of NG2+ cells in the molecular layer. These recent findings require revision of our conventional concepts of neuronal circuits to include glial cells not only as supportive elements but also as important components for information processing.

5-8 Summary In Chapters 4 and 5, more than ten unique cell types in the cerebellar cortex are defined. They interact in complex neuronal circuits in the cerebellar cortex. Among the inhibitory neurons in the molecular layer, basket/stellate cells add to operation of the mossy fiber-granule cell-Purkinje cell pathway. Among the inhibitory interneurons in the granular layer, Golgi cells are suggested to provide a clock circuit (Chapter 9), but functional roles of the other cell types still remain to be clarified. In addition to these neurons, two types of glial cells, Bergmann and NG2+, appear to be involved importantly in the neuronal circuit function of the cerebellar cortex.

6 Pre- and Post-Cerebellar Cortex Neurons

6-1 Introduction In this chapter we decompose the neuronal circuits that connect the cerebellar cortex with other CNS structures. Some of the component cells are those of origin of afferents to the cerebellum that are located in the spinal cord and brainstem. These cells issue afferents known as mossy fibers, climbing fibers, and beaded fibers. Other component cells are the recipients of inhibitory signals from Purkinje cells, and they are located in cerebellar nuclei and selected brainstem nuclei. These precerebellar and postcerebellar cortex neurons, together with cerebellar cortical networks, constitute “global” neural systems that incorporate the cerebellum.

6-2 Cells of Origin of Mossy Fibers The vestibular nerve provides a unique case in which primary afferents project directly as mossy fibers to the vestibulocerebellum. These projections are prominent in the nodulus and uvula (lobules X and IX) and less represented in the flocculus (see Chapter 10, “Ocular Reflexes,” Section 3). The dorsal and ventral spinocerebellar tracts are examples of secondary afferents that are mossy fibers. These tracts convey proprioceptive and exteroceptive information from the body to the cerebellum. The dorsal spinocerebellar tract (DSCT) originates from neurons largely in Clarke’s column, which is located in the C8-L2/3 spinal segments. DSCT neurons project their axons through the ipsilateral region of the posteriolateral funiculus and then into the granular layers of cerebellar lobules I, II, III, and VIII. The ventral spinocerebellar tract (VSCT) originates from neurons located caudal to Clarke’s column with axons that ascend bilaterally through the ventral border of the lateral funiculi toward the cerebellum. They double cross the midline, first to the opposite side of the body and then back to their original side in the cerebellum. 60

6 • Pre- and Post-Cerebellar Cortex Neurons

61

The DSCT is usually considered to be purely sensory in transmission of proprioceptive information from the ipsilateral hindlimb. However, the presence of corticospinal input to DSCT neurons has long pointed to an integration of sensory input and motor command signals to these neurons (Hongo et al., 1967; Hongo and Okada, 1976). A recent study showed that DSCT neurons encode, in addition, a global representation of hindlimb mechanics during passive movement of a hindlimb. The discharge of these neurons was shown to be qualitatively but not quantitatively similar for active stepping versus passive, manually imposed steps (Bosco et al., 2006). The difference was suggested to indicate a contribution of DSCT to the spinal cord’s contribution to the control of locomotion (Chapter 11, “Somatic and Autonomic Reflexes”). In a very recent study by Hantman and Jessell (2010), proprioceptive afferents, corticospinal tract axons, and GABAergic and glycinergic synapses were labeled differentially by genetic/molecular markers to reveal their direct synaptic contact onto DSCT neurons. Moreover, patch clamp recording from DSCT neurons revealed that monosynaptic EPSPs were evoked by stimulation of both the dorsal roots and the dorsal column, the latter involving corticospinal axons. These observations indicated converging excitatory inputs to DSCT neurons from a proprioceptive pathway and the corticospinal tract. These neurons also have GABAergic and glycinergic synapses, possibly from the numerous GABAergic and glycinergic interneurons that surround Clarke’s column and receive dense innervation from corticospinal terminals. Consistent with these anatomical findings, dorsal column stimulation evoked long-lasting IPSPs in DSCT neurons. There was also indication of presynaptic inhibition, which occurred in corticospinal tract synapses on DSCT neurons. These synaptic connections point to the possibility that corticospinal descending signals for voluntary movements act on spinal segmental circuits not only for executing a voluntary movement, but also for adjusting signals ascending to supraspinal centers and the cerebellum via DSCT. This adjustment might be required for adapting posture and locomotion mechanisms to an intervening voluntary movement (see Chapter 11). In contrast, the activity of VSCT neurons seems to be determined mainly by a central mechanism rather than sensory input. These neurons discharge rhythmically at their own consistent phase of the step cycle when the scratch reflex or locomotion is elicited in cats (Arshavsky et al., 1978, 1988). During fictive scratching and locomotion in a paralyzed animal, the firing pattern of these neurons is similar to that during actual scratching. Therefore, the rhythmical burst firing of VSCT neurons must be determined centrally. Such a central mechanism may operate via segmental interneurons, which seemingly provide the VSCT with feedback information about their actions on motoneurons.

62

the cerebellum

The “rostral spinocerebellar tract” is the forelimb equivalent of the hindlimb’s VSCT. It transmits, among probably other information, Ib afferent discharge to the cerebellum that arises from Golgi tendon organs of the cranial half of the body. The connectivity organizations of these two tracts are largely similar but with certain differences (Oscarsson et al., 1965). The “cuneocerebellar tract” in the cat consists of a proprioceptive component that arises from the external cuneate nucleus and is activated by group I muscle afferents. A second exteroceptive component arises from cells in the rostral part of the main cuneate nucleus and is activated by cutaneous afferents (Cooke et al., 1971). The cuneocerebellar tract terminates in the intermediate part of lobule V of the anterior lobe of the cerebellum and in the rostral four folia of the paramedian lobule. The lateral reticular nucleus has three subnuclei: the parvocellular, magnocellular, and subtrigeminal subnuclei. The major part of the lateral reticular nucleus (mLRN) involves the parvocellular portion and the immediately adjacent magnocellular portion. These portions receive monosynaptic excitation from fibers ascending in the ventral part of the ipsilateral lateral funiculus of the spinal cord. In turn, mLRN projects almost exclusively to the classical spinocerebellum (Clendenin et al., 1974), involving bilaterally the intermediate part of the anterior lobe. Some fibers from the mLRN terminate also in the rostral part of lobule VI but hardly any in other parts of the cerebellar cortex. The rest of the magnocellular subnucleus receives inputs from higher brain structures and projects to the cerebellar hemispheres. The subtrigeminal nucleus sends its projections to the flocculonodular lobe. Another prominent source of mossy fibers is the pontine nuclei in the brainstem. The vast majority of pontine nuclear neurons receive monosynaptic glutamatergic inputs from corticopontine fibers and, in turn, send their axons to the cerebellar cortex. Large and functionally diverse areas of the cerebral cortex find their “own” territories in the pontine nucleus. Similarly, each part of the pontine nucleus projects to specific areas of the cerebellar cortex. This topographic organization of the pontine nuclei may ensure that information from various functionally diverse parts of the cerebral cortex and subcortical nuclei is brought together and integrated in the cerebellar cortex (Brodal and Bjaale, 1992). In the vermis and flocculonodular lobe, the major excitatory inputs to the target neurons of Purkinje cells are provided by collaterals of mossy fiber afferents (Figure 13). However, the presence of collaterals of pontocerebellar mossy fibers to cerebellar nuclei has been controversial. Shinoda et al. (1992) used intra-axonal horseradish peroxidase staining in cats to trace single axons of pontine nuclear neurons that received cerebral inputs. Forty such axons were shown to terminate as

6 • Pre- and Post-Cerebellar Cortex Neurons

63

typical mossy fiber rosettes in the granular layer of the cerebellar cortex. Of these, 22 axons projected collaterals to the lateral cerebellar nucleus. Virtually all of the axon branches observed in the lateral cerebellar nucleus were collaterals of mossy fibers from pontine nuclear neurons to the cerebellar cortex. One to three collaterals were projected from their parent axons to the lateral cerebellar nucleus. They were very thin (mean diameter, 0.6 micrometers) compared to the large parent axons (2.1 micrometers). Some collaterals were very simple with only one terminal branch with or without short branchlets. In other cases, single collaterals ramified before forming a terminal arborization. Hence, in contrast to the large divergence and convergence of the corticopontine-cerebellar projections, pontine nuclear neurons appear to form rather specific connections with lateral cerebellar nucleus neurons. Another source of mossy fiber afferents in the brainstem is the nucleus reticularis tegmenti pontis (NRTP). It contains neurons that relay visual signals to the flocculus (Kano et al., 1991) and other cells that are sensitive to head rotation (Taillanter and Lannou, 1988) (Chapter 10). The NRTP in monkeys also receives major inputs from the primary motor cortex and less input from the premotor region (area 6), the somatosensory cortex (areas 3, 1, and 2), and area 5 (Figure 2). These cortical afferents terminate throughout the NRTP except for its dorsomedial part (Brodal, 1980). The pontine nuclei and the NRTP project to the lateral cerebellar nuclei. This was shown by use of biotinylated dextran labeling in rats. The pontine nuclei also project to the posterior interpositus nucleus. In contrast, NRTP projects to the lateral cerebellar nucleus, the nucleus interpositus, and the caudal part of the nucleus medialis. Cerebellar nuclear neurons are innervated mainly by projections from contralateral pontine nuclei. They are also innervated bilaterally, albeit to a lesser degree by NRTP. These anatomical differences imply different functional roles for these two sources of mossy fibers (Parenti et al., 2002).

6-3 Inferior Olive and Preolivary Nuclei Climbing fibers originate solely from the inferior olive (IO) in the opposite side of the medulla. In rodents, this complex consists of three primary subdivisions: the principal olive (PO), medial accessory olive (MAO), and the dorsal accessory olive (DAO) (Color Plate III). In cats, the total number of IO neurons is ten times greater than that of Purkinje cells, such that one climbing fiber appears to project axonal branches to innervate ten Purkinje cells. This observation was verified in a study using the retrograde fluorescent double-labeling method in rats. The lateral parts of the IO project more rostrally within a longitudinal zone, and the medial

64

the cerebellum

parts project more caudally in the same zone. Double-labeled olivary neurons with axons branching rostrally and caudally within a single zone were found to lie in an intermediate position between the two groups of single-labeled neurons (Wharton and Payne, 1985). The IO contains two morphologically distinct types of neurons. One group is “curly” cells, with relatively small cell somata and curly dendrites as described by Scheibel and Scheibel (1975). The other type is “straight” cells with relatively large cell somata and long, relatively straight dendrites (Devor and Yarom, 2002) (Color Plate XI A, B). A characteristic feature of IO neurons is that they communicate with each other via gap junction-mediated electrical synapses (Llinás et al., 1974; Sotelo et al., 1974) as demonstrated in Color Plate XI. Electrical synapses provide reciprocal pathways for ionic current and small organic molecules, and they are maintained by the protein connexin36 (Cx36) (Condorelli et al., 2001). Neurobiotin spreads through gap junctions. When injected into an IO neuron, it was shown to produce indirect labeling of 1–11 nearby cells (average, 3.8 cells). All indirectly labeled cells were found within 70 micrometers of the injected cell, that is, within or immediately adjacent to the dendritic field of the stained cell (Color Plate XI). Dye passage through gap junctions from a straight type IO neuron stained only straight type neurons. Likewise, dye passage from a curly type IO neuron stained only curly type neurons. Each IO neuron was calculated to be directly coupled with about 50 other IO neurons. The coupling coefficient was defined as the ratio between voltage responses of the post- and prejunctional cell. As measured by simultaneous double patch recordings from IO neurons in rat brain slice preparations, the coupling coefficient varied between 0.002 and 0.17; that is, most of the pairs were weakly coupled (Devor and Yarom, 2002). IO neurons receive GABAergic synapses, most of which arise from cerebellar nuclear neurons (De Zeeuw et al., 1988). In rats, an antibody against a GABA-synthesizing enzyme (GAD) labeled axon terminals impinging onto dendrites (94% of the sample) or with somata (6%). These terminals were intermingled with dendrodendritic gap junctions within the complex structure of glomeruli (Sotelo et al., 2004). The functional association between IO neurons and axon terminals of cerebellar nuclear neurons was indicated by the systematic presence of GAD-labeled cerebellar nuclear neurons’ axon terminals directly apposed to the IO cells’ dendritic appendages linked by gap junctions, and by the frequent presence of type II synapses (allegedly inhibitory) straddling both elements. Gap junctions were located mostly in glomeruli at the distal dendrites of IO neurons (Sotelo et al., 1974; De Zeeuw et al., 1990a,b). One glomerulus contained a core of five to six

6 • Pre- and Post-Cerebellar Cortex Neurons

65

dendritic and axonal spiny appendages, derived from different IO neurons, coupled by gap junctions, and surrounded by both excitatory and inhibitory synaptic terminals of extrinsic origin (De Zeeuw et al., 1990a). These arrangements appear to serve for the modulation of electrical synapses by the release of GABA, which increases nonjunctional membrane conductance and thereby shunts the electric coupling between inferior olive neurons. Under certain conditions, the electrical synapses mediate regular sinusoidal oscillations of the membrane potential of IO neurons. About 10% of these neurons tested in guinea pig brainstem slices exhibited spontaneous subthreshold oscillations in their membrane potential at a frequency of 4–6 Hz and an amplitude of 5–10 mV. This oscillation was synchronized in all cells tested within the slice, persisted in the presence of tetrodotoxin, and was blocked by Ca2+ conductance blockers or by removal of Ca2+ from the bathing solution. The oscillation was not affected by the intracellular activation of any given neuron. Therefore, the oscillation seemed to reflect the ensemble properties of a large number of IO neurons coupled in a network. A similar ensemble oscillation was induced by adding harmaline and serotonin to the bathing solution and it was blocked by adding noradrenaline to the bathing solution (Llinás and Yarom, 1986). Given that IO neurons discharge with a highly regular rhythm under selected conditions, a hypothesis has been proposed that climbing fibers provide a periodic clock for coordinating movements or motor timing (Kazantsev et al., 2004). In alert monkeys, however, climbing fiber signals have been reported to occur randomly at a characteristically low rate (~1 Hz) (Keating et al., 1995). It appears that IO neurons recode the high-frequency information carried by its synaptic inputs into stochastic, low-rate discharge in their climbing fiber outputs. This possibility has been simulated by a computational model (Schweighofer et al., 2004). The characteristic multiple spike discharges induced in Purkinje cells by IO stimulation partly reflect (1) Ca2+ spikes in the Purkinje cell membrane (see Figure 15B) and (2) the long-lasting after-depolarizing potential (ADP) that follows a single spike in IO neurons. The number of spikes in the burst discharge of IO axons was shown to depend on the phase of their subthreshold oscillations, which suggested that the burst stage encoded the state of the IO network (Mathy et al., 2009). Preolivary neurons projecting ascending pathways from the spinal cord to the IO complex can be labeled by the anterograde transport of WGA-HRP from spinal segments (Matsushita et al., 1992). It was thus shown that C1-T2 and also L6-S1 segments project to the caudal half of the MAO, C1-C4 segments to the most

66

the cerebellum

medial part of the DAO, and C5-T1 segments more laterally. Even more lateral projections to the DAO were shown to arise from T2-L5 and L6-S1 spinal segments. These findings imply distinct somatotopic projections in the mediolateral order to the caudal MAO and the DAO. On the other hand, certain midbrain neurons project descending axons to the IO complex. They have been labeled by the retrograde transport of HRP in cats. Such midbrain neurons were shown to be located in the ipsilateral mesencephalon, from the rostral pole of the red nucleus to the caudomedial border of the thalamus. Heavily labeled nuclear groups included the parvocellular red nucleus, the interstitial nucleus of Cajal, the nucleus of Darkschewitsch, and the caudomedial extremity of the subparafascicular nucleus (SaintCyr and Courville, 2004). A few cells were also labeled in the reticular formation lateral to the interstitial nucleus of Cajal, in the caudomedial parafascicular nucleus, in the nucleus of the fields of Forel, and in the central gray. Elsewhere in the midbrain, there were HRP-labeled cells in the deep layers (IV, VI) of the superior colliculus, predominantly on the side contralateral to the injection site, in the nucleus of the optic tract, and in the ipsilateral anterior and posterior pretectal nuclei. These latter groups were best labeled after caudally located injections into the olive. No labeled cells were found by Saint-Cyr and Courville (2004) in the basal ganglia, rostral raphe nuclei, and the nucleus of Edinger-Westphal, despite a previous report to the contrary.

6-4 Cells of Origin of Beaded Fiber Among beaded fibers so far identified, serotonergic fibers originate from the raphe nucleus, and noradrenergic fibers from the locus coeruleus. These aminergic fibers exert a unique facilitatory action on inhibitory neurons in the granular layer (Chapter 5, “Inhibitary Interneurons and Glial Cells in the Cerebellar Cortex”). Histaminergic fibers originate from a group of hypothalamic neurons. Histamine was shown to excite rat Purkinje cells via H2 receptors (Tian et al., 2000). Twenty-four different neuropeptides have been identified in the cerebellum (Ito, 2009), with many expressed in mossy fibers, climbing fibers, and several cellular structures. It is also known that angiotensin II and orexin are located in beaded fibers that originate in the hypothalamus. The stimulation of orexinergic neurons in the hypothalamus was shown to excite Purkinje cells in a small area (folium-p) of the flocculus. Hence, it was suggested that orexinergic fibers facilitate folium-p Purkinje cells together with brainstem structures involved in defense reactions (Nisimaru et al., 2010).

6 • Pre- and Post-Cerebellar Cortex Neurons

67

6-5 Cerebellar Nuclear Neurons Four types of neurons are packed into cerebellar nuclei. First are excitatory projection neurons, which release glutamate as the transmitter and project to the medulla and midbrain structures. These neurons have the largest somata and dendritic trees among cerebellar nuclear cells. Second, smaller GABAergic neurons project to the IO (De Zeeuw et al., 1989). In glutamate decarboxylase (GAD) 67-green fluorescent protein (GFP) knock-in mice, GABAergic neurons were identified as GAD-positive neurons. Compared with GAD-negative neurons, GAD-positive neurons generated broader action potentials, displayed stronger frequency accommodation, and did not reach high firing frequencies during depolarizing current injections. GAD-positive cells also displayed slower spontaneous firing rates but exhibited a longer-lasting rebound depolarization and associated spiking after a transient membrane hyperpolarization (Uusisaari et al., 2007). Third are glycinergic neurons recently found in the medial cerebellar (fastigial) nuclei of mice (Bagnall et al., 2009). Glycinergic fastigial neurons form projections to vestibular and reticular neurons in the ipsilateral brainstem, whereas their glutamatergic counterparts project contralaterally. These three types of projection neurons receive inhibitory inputs from Purkinje cells and excitatory inputs via collaterals of mossy fibers and climbing fibers. Fourth are the smallest cells, which extend their axons within the cerebellar nuclei (Chan-Palay, 1977). Both GABA and glycine transmitters were shown to co-localize in these cells (Chen and Hillman, 1993). Their input and output connections remain unknown, however. Among rat cerebellar nuclear cells, projection neurons (first type) show cyclic burst firing with underlying Na+ and Ca2+ plateau potentials, low-threshold Ca2+ spikes, and a slow Ca2+-dependent afterhyperpolarization. A small subset of cerebellar nuclear neurons lacks slow plateau potentials and low threshold spikes, and they have a lower rate of spontaneous discharge. These neurons were conjectured to be among the fourth type of cerebellar nuclear interneurons (Czubayko et al., 2001). Aizenman and Linden (1999) provided evidence that a characteristic feature of cerebellar nuclear neurons was their prominent rebound depolarization that occurred at the end of injected hyperpolarizing current pulses leading to a Na+ spike burst. This rebound depolarization depends on low-threshold T-type voltagegated Ca2+ channels and may involve also the activation of a hyperpolarization-activated cation current. Similar rebound was observed also in vestibular neurons receiving inhibition from flocculus Purkinje cells (Sekimjak et al., 2003). Such rebound has been interpreted as playing several functional roles, including timing, encoding information (pause coding; Steuber et al., 2007), and mediating synaptic

68

the cerebellum

plasticity (see Chapters 8 and 9). However, it has been claimed that such a rebound does not occur unless the test cell is slightly depolarized by microelectrode penetration (Alviña et al., 2008; Walter and Khodakhah, 2009). As such, the physiological meaning of this rebound is still a matter of discussion.

6-6 Vestibular Nuclear Neurons and Other Brainstem Neurons In the flocculus, Purkinje cells project directly to certain brainstem neurons rather than cerebellar nuclear cells. There are four studied cases. (1) Purkinje cells that project to the superior and medial vestibular nuclei (Fukuda et al., 1972; Kawaguchi, 1985; Sekimjak et al., 2003). Recall that these nuclei process the VOR. Such VOR neurons are either excitatory (glutamatergic) or inhibitory (glycinergic) in their action on oculomotor neurons (Chapter 10). (2) Purkinje cells in zone B of the vermis that inhibit the dorsal part of the lateral vestibular nucleus of Deiters (Ito et al., 1964), which in turn exerts excitatory actions on segmental motoneurons and interneurons (Wilson and Yoshida, 1969). (3) Purkinje cells located in the lateral edge of the nodulus that project directly to the medial portion of the ipsilateral parabrachial nucleus, which is located immediately rostral to the superior vestibular nucleus (Nisimaru et al., 1998). The parabrachial nucleus has been shown to mediate the vestibulo-sympathetic reflex and thereby maintain mean blood pressure constant during head movement. (4) Purkinje cells that have been shown to project directly to the lateral portion of the ipsilateral parabrachial nucleus (Nisimaru et al., 2010). This nucleus mediates the somatosympathetic reflex, which redistributes arterial blood flow in defense reactions and locomotion (Chapter 11).

6-7 Summary Morphological, immunocytochemical, and genetic analyses have advanced remarkably to reveal new properties of neurons in the cerebellum and its related structures. We now need more information about the activity of these neurons in behaving animals. It is hoped that this advance should be forthcoming in the near future through the development of new technologies for improved telemetry, optogenetics mapping, and molecular imaging.

7 Conjunctive Long-Term Depression (LTD)

7-1 Introduction Among electrical and chemical signaling capabilities of neurons, synaptic plasticity is a unique process that induces persistent changes in synaptic transmission efficacy in an activity-dependent manner. The best-known examples are long-term potentiation (LTP) and depression (LTD). A cascade of cellular/molecular events, which are initiated by an experience, generates these changes. They are considered to culminate in a durable form of altered synaptic transmission efficacy. “Conjunctive LTD,” which is induced postsynaptically by the combined activation of parallel fibers and a climbing fiber converging onto a Purkinje cell, is a unique type of synaptic plasticity in the cerebellum because of the involvement of climbing fibers that are unique to the cerebellum. The characteristic physiological properties of such conjunctive LTD and its underlying mechanisms are discussed in the following sections.

7-2 Properties of Conjunctive LTD Conjunctive LTD occurs in parallel fiber-Purkinje cell synapses when parallel fibers are activated simultaneously with a climbing fiber that converges onto the same Purkinje cell. The whole story of its discovery in the cerebellum was introduced in Chapter 3, “The Cerebellum as a Neuronal Machine.” In rat and mouse in vitro cerebellar slices, conjunctive stimulation at 1 Hz for 5 minutes (300 pulses) is optimal (Karachot et al., 1994) for reducing by ~30% the EPSPs (their rising slope) and EPSCs of Purkinje cells, as activated by parallel fiber input. A similar result is obtained when brief stimulation of parallel fibers by 8 pulses (100) precedes stimulation of a climbing fiber with a low frequency train of 3 pulses (20 Hz) (Schreurs et al., 1996). 69

70

the cerebellum

Conjunctive LTD involves postsynaptic glutamate receptors in parallel fiberPurkinje cell synapses. This finding is based on the following observations. When parallel fiber stimulation is replaced by the iontophoretic application of glutamate to directly activate the postsynaptic membrane of a Purkinje cell, the conjunction of this and climbing fiber stimulation depresses the glutamate sensitivity of the cell (Ito et al., 1982). In a cultured Purkinje cell devoid of dendritic spines, a combination of glutamate (or quisqualate) pulses and membrane depolarization, the latter causing Ca2+ entry, effectively depresses glutamate and AMPA sensitivity (Linden et al., 1991, 1995). In mice deficient in mGluR4, which is normally expressed in parallel fibers, parallel-fiber-induced paired-pulse facilitation and post-tetanic potentiation are impaired, but conjunctive LTD remains fully operative (Pekhletski et al., 1996). This showed that LTD does not require the mGluR4-regulated presynaptic mechanism to maintain synaptic efficacy during repetitive activation. Conjunctive LTD is a unique phenomenon at parallel fiber-Purkinje cell synapses: for example, it does not occur in synapses formed by ascending granule cell axons and Purkinje cell dendrites (Sims and Hartell, 2006). Furthermore, conjunctive LTD is input-specific. It occurs only in parallel fiber-Purkinje cell synapses involved in conjunctive stimulation (Ito et al., 1982; Ekerot and Kano, 1985). This input specificity is maintained in cultured Purkinje cells, in which LTD-like depression occurs only on the part of the dendrite exposed to quisqualate, but not in other unexposed parts (Linden, 1994). Nevertheless, in rat cerebellar slices, Wang et al. (2000b) estimated that conjunctive LTD spreads from a small number (~5-30) of stimulated parallel fiber synapses to ~20,000 unstimulated parallel fiber synapses. Thus, LTD must spread beyond conjunctively activated parallel fibers to about 600-fold (20,000/30) more parallel fiber synapses in the immediate neighborhood. The magnitude of this spread decreases to ~50% of the maximum 50 micrometers away from the conjunctively activated site. Such a spread, however, occurs only in neighboring synapses that are active at a very low rate of discharge (~0.2 Hz) and it has only a loose temporal relationship with climbing fiber activity (Reynolds and Hartell, 2000). Thus, the principle of input specificity for conjunctive LTD still holds in a broad sense. In attempts to analyze the mechanisms of conjunctive LTD, reduced forms of its manifestation are often examined. These have been induced by replacing parallel fiber- or climbing fiber stimulation with various pharmacological reagents. For example, iontophoresis of glutamate or quisqualate onto Purkinje cells, but not kainate and aspartate, has been used to replace parallel fiber stimulation (Ito et al., 1982; Kano and Kato, 1987). In another study, strong depolarizing pulses, which induced the entry of Ca2+ ions into Purkinje cells through voltage-gated channels, were used to replace climbing fiber stimulation (Crepel and Kruppa, 1988; Crepel

7 • Conjunctive Long-Term Depression (LTD)

71

and Jaillard, 1991). Other simple procedures for inducing LTD were to apply quisqualate (Yuzaki et al., 1994) that stimulates both AMPA receptors and metabotropic glutamate receptor type 1 (mGluR1) receptors, or to increase extracellular K+ concentration (Crepel et al., 1994). Intracellular photo-release of free Ca2+ ions from “caged” Ca2+ ions induced LTD effectively (Finch and Augustine, 1998). Also, Linden and his colleagues used to good effect glutamate/depolarization conjunction to induce LTD in cultured cells (Linden et al., 1991).

7-3 Major Signal Transduction Pathways Complex signal transduction underlying synaptic plasticity in cerebellar neurons has been analyzed extensively and in detail, particularly with respect to conjunctive LTD (Figure 19). Among the many molecules involved, some are indispensable for LTD induction, but others are required only under selected experimental conditions. According to the convention for LTP (Sanes and Lichtman 1999), such molecules will be defined as mediators and modulators, respectively (Ito, 2002). Ca2+ surge in Purkinje cell dendrites. The first step in the induction of conjunctive LTD can occur in two ways. One is by Ca2+ entry through the voltagegated Ca2+ channels, which are called “P/Q” channels. The P channel was discovered in cerebellar Purkinje cells and named P in keeping with the first letter in Purkinje (Llinás and Sugimori, 1980; Llinás et al., 1989). The Q channel is a phenotypic variant of a P channel that is generated by gene splicing (Bourinet et al., 1999). P/Q channels are distributed throughout Purkinje cell somata and their entire dendritic trees (Westenbroek et al., 1995). Ca2+ entry occurs during membrane depolarization caused by either climbing fiber responses or by parallel-fiberevoked EPSPs. The other way is for Ca2+ to be released from intracellular Ca2+ stored in the endoplasmic reticulum. This release is triggered by the following steps. Activation of mGluR1 activates, in turn, the G-protein (Gq/11) (Mailleux et al., 1992; Sternweis and Smrcka, 1993; Tanaka et al., 2000), which, again in turn, activates certain subtypes of phospholipase C (PLCβ3 or 4) (Roustan et al., 1995; Watanabe et al., 1998; Sugiyama et al., 1999). The latter then hydrolyzes the membrane’s phosphatidyl-inositol 4,5-diphosphate (PIP2) to produce two second-messengers; inositol-trisphosphate (IP3) and 1,2 diacylglycerol (DAG). Finally, IP3 acts on IP3 receptors located on the membrane of the endoplasmic reticulum and induces Ca2+ release. In addition, Ca2+ also affects the gating of the IP3 receptor channel as a co-agonist of the IP3 receptor, together with IP3 (Iino, 1990; Bezprozvanny et al., 1991; Finch et al., 1991). A kinetic model of Ca2+ dynamics within a Purkinje dendritic spine was formulated assuming that (1) IP3 was generated

72

the cerebellum

slowly via the metabotropic pathway of parallel fiber inputs; and (2) the climbing fiber-evoked Ca2+ influx triggered regenerative Ca2+-induced Ca2+ release from the internal stores. The delay in IP3 increase caused by the slow parallel fiber metabotropic pathway explained the optimal parallel fiber-climbing fiber stimulation interval (50–200 ms) for inducing LTD (Doi et al., 2005).

Figure 19 Signal transduction underlying LTD.

Chain of reactions induced by conjunctive stimulation of parallel fibers (PF) and climbing fibers (CF). I, II, and III signify the three phases of the signal transduction process. Additional abbreviations: AA, arachidonic acid; COX-2, cyclooxygenase type 2; cPLA 2α; cytosolic phospholipase A 2 alpha subtype; DAG, diacylglycerol; ERK, extracellular signal-regulated kinase (belong to MAP kinases); G11/Ge/Gi/Gq, subtypes of G proteins; IP 3, inositol 1,4,5-trisphosphate; IP 3R, IP 3 receptor; MEK, MAPK/ERK kinase; mGluR1, metabotropic glutamate receptor type 1; PGD 2/E 2, prostaglandin D 2 and E 2; PLC, phospholipase C (PLCβ3 or 4 subtypes); Raf, kinase related to MAP kinase; VSCC, voltage-sensitive Ca2+ channel.

Importantly, the coincident activation of parallel fibers and a climbing fiber converging onto the test Purkinje cell produces Ca2+ signals that are much larger than the linear sum of responses to either parallel fiber- or climbing fiber activation alone: that is, ~8 times in spines and ~4.5 times in dendritic shafts (Wang et al.,

7 • Conjunctive Long-Term Depression (LTD)

73

2000a). This supralinear summation of climbing-fiber-evoked Ca2+ entry and parallel-fiber-evoked release of Ca2+ leads to an abrupt surge in the Ca2+ concentration of Purkinje cell dendrites, as shown in Color Plate XII. Note in curves b–d in C that the Ca2+ indicator fluorescence sharply increases to a peak and then decreases to the baseline at each successive conjunctive pulse. The requirement of mGluR1-initiated signal transduction for the induction of conjunctive LTD has been shown by various means to interfere with signal transduction. These include (1) a specific antagonist of mGluR1, (RS)-α-methyl-4-carboxyphenylglycine (MCPG) (Hartell, 1994), (2) antibodies inactivating mGluR1 (Shigemoto et al., 1994), and (3) mGluR1-deficiency gene knockout (Conquet et al., 1994). In mice deficient in Gaq, conjunctive LTD was shown to be lacking (Miyata et al., 1998), albeit with the reservation that this mutant has climbing fibers that multiply innervate a single Purkinje cell (Offermanns et al., 1997). This aberrant innervation might impair the production of conjunctive LTD. On the other hand, the activation of mGluRs by an agonist (e.g., trans-ACPD or 1S, 3R-ACPD) has been shown to induce LTD when combined with depolarization-evoked Ca2+ spikes. This was shown to occur even in the presence of CNQX, an antagonist of AMPA receptors (Daniel et al., 1993; Hemart et al., 1995). In another study a PLC inhibitor greatly attenuated an increase in mGluR-agonist-induced intracellular Ca2+ concentration (Netzeband et al., 1997). The parallel-fiber-stimulation-evoked Ca2+ increase was also shown to be blocked in mice deficient in the PLCβ4 gene (Miyata et al., 2001). The importance of IP3-induced Ca2+ release for induction of LTD has been shown by the following observations. (1) Application of heparin, a nonspecific inhibitor of the IP3 receptor, blocked LTD induced by either glutamate/depolarization conjunction in cultured Purkinje cells (Kasono and Hirano, 1995) or parallel fiber/depolarization conjunction in Purkinje cells in a cerebellar slice (Khodakhah and Armstrong, 1997). (2) A specific antibody against the IP3 receptor blocked LTD induction, and mice with a disrupted IP3 receptor type 1 gene completely lacked LTD (Inoue et al., 1998). (3) Both LTD and IP3-mediated Ca2+ signaling in dendritic spines of Purkinje cells were absent in mice and rats with mutations in myosin-Va that prevented endoplasmic reticulum from entering the cell’s dendritic spines. This loss of LTD was rescued by photolysis of a caged Ca2+ compound (Miyata et al., 2000). (4) The photolytic release of caged IP3 in peripheral dendrites of a Purkinje cell produced a strong and persistent Ca2+ signal and evoked persistent LTD-like depression (Finch and Augustine, 1998). On the other hand, in cultured or freshly isolated Purkinje cells, IP3 was not required for inducing LTD by glutamate/depolarization conjunction. A potent and selective IP3 receptor channel blocker, xestospongin C, did not affect this LTD

74

the cerebellum

induction (Narashimhan et al., 1998). IP3 is therefore not required for LTD induction in cultured or freshly isolated Purkinje cells, and hence IP3 is a modulator, not mediator, for LTD induction. PKCα and lipid-signaling cascade. In Purkinje cell dendrites, a conjunction-evoked surge in Ca2+ ions should activate two Ca2+-sensitive enzymes: PKCα and cPLA2α. DAG generated by PLC from membrane PIP2, in parallel with the production of IP3 (see above), should also contribute to the activation of PKCα. The activated PKCα induces the phosphorylation of AMPA receptors in parallel fiber-Purkinje cell synapses. On the other hand, the activated cPLA2α initiates a positive feedback cycle consisting of (1) cPLA2α releasing arachidonic acid from membrane phospholipids, (2) arachidonic acid activating PKCα, (3) PKCα activating sequentially a series of kinases, Raf, MEK, and ERK1/2, and (4) these kinases then activating cPLA2α (Figure 19). This cyclic activation involving PKCα and cPLA2α has been postulated to strengthen the activation of PKCα (Kuroda et al., 2001; Tanaka and Augustine, 2008). However, it has been shown quite recently that arachidonic acid also acts on conjunctive LTD via a cascade involving cyclooxygenase-2 (COX-2) and prostaglandin D2 or E2 (PGD2/PGE2) (Le et al., 2010) (Figure 19). Because either a PKC inhibitor or a COX-2 inhibitor blocks LTD, it appears that PKCα and PGD2/E2 act synergistically on AMPA receptors to dissociate them from the cytoskeleton (see below). Events in AMPA receptors. AMPA receptors are the sole ionotropic glutamate receptors in parallel fiber-Purkinje cell synapses; there are no NMDA receptors except during an early postnatal period. In conjunctive LTD induction, AMPA receptors are phosphorylated at serine-880 of the GluR2 COOH terminus and become dissociated from the subsynaptic cytoskeleton (Matsuda et al., 2000; Xia et al., 2000). The dissociated AMPA receptors are eventually eliminated from dendritic spines by endocytosis (for review, see Carroll et al., 2001). Such shift of AMPA receptors from synaptic membrane to inside of dendritic membrane has been demonstrated by immunolabeling AMPA recptors with an antibody binding the extracellular domain of GluR2/3 (Color Plates XIII and XIV). The size of parallel fiber-Purkinje cell EPSCs reflects the number of AMPA receptor molecules present in the synaptic membrane (Linden, 2001). The number of AMPA receptors is maintained by the balance between endocytotic elimination and exocytotic insertion. To explain AMPA receptor behavior during conjunctive LTD, their stable and mobile pools in the synaptic membrane and their internal mobile pool are in dynamic equilibrium within each dendritic spine. It is assumed currently that during conjunction, a portion of the stable synaptic pool of AMPA receptors is

7 • Conjunctive Long-Term Depression (LTD)

75

shifted to the mobile synaptic pool and then moved to the internal mobile pool as schematically illustrated in (Figure 20).

Figure 20 AMPAR trafficking in parallel fiber-Purkinje cell synapses.

A schematic of the three-compartmental kinetic model of the constitutive trafficking of AMPAR receptors (see AMPAR at arrow), which consist of GluR2/3 subunits. In the postsynaptic membrane (half dark line), the stable form of AMPARs (on the right) changes to the mobile form (on the left) and vice versa. In the resting state, they are in equilibrium. The mobile forms of AMPARs in the synaptic membrane are also in equilibrium with their mobile forms in the internal pool. Symbols attached to AMPAR models represent GRIP (G), PICK1 (K), phosphor (P), and stargazin (S). (Courtesy of Kazuhiko Yamaguchi.)

7-4 Other Signal Transduction Pathways This section focuses on a number of other molecules related to LTD induction. It includes some molecules whose roles in LTD induction are now emerging but require more definitive study. Nitric oxide (NO). Parallel-fiber impulses activate NO synthase in parallel fibers, which releases NO that then diffuses to the postsynaptic side of parallel fiber-Purkinje cell synapses (Shibuki and Kimura, 1997). NO then activates a cascade that links guanylyl cyclase, cyclic GMP, and protein kinase G (PKG) and results in the phosphorylation of the G-substrate in Purkinje cells. G-substrate has been cloned, and its molecular structure has been determined (Endo et al., 1999; Hall et al., 1999). The phosphorylated G-substrate is a potent inhibitor of two protein phosphatases, PP1 and PP2A (Detre et al., 1984), and hence disinhibits kinases involved in conjunctive LTD induction (Endo et al., 2003).

76

the cerebellum

Two sets of observations in slice preparations indicate the involvement of NO in conjunctive LTD induction. First, this induction occurred following the bath application of an NO donor, sodium nitroprusside (Shibuki and Okada, 1991), or infusion of the NO donor, 4-ethyl-2-hydroxyamino-5-nitro-hexenamide (NOR3) into Purkinje cells (Daniel et al., 1993). LTD also occurred after the release of NO from its caged form within Purkinje cells in combination with depolarizationinduced Ca2+ entry (Lev-Ram et al., 1997a). Second, the induction of conjunctive LTD was readily blocked by bath application of NO synthase inhibitors (Crepel and Jaillard, 1990; Shibuki and Okada, 1991; Daniel et al., 1993). LTD induced by conjunctive stimulation of parallel fibers and membrane depolarization was also blocked by KT5823, a PKG inhibitor (Reynolds and Hartell, 2001). The targeted disruption of the NO synthase gene in mice has been shown to result in the near-complete loss of NO synthase activity in the cerebellum (Huang et al., 1993). As a result, conjunctive LTD did not occur in cerebellar slices obtained from these mice (Lev-Ram et al., 1997b). Disappointingly, however, the photolytic uncaging of NO and cyclic GMP inside Purkinje cells did not rescue LTD in these slices. A possible explanation for this failure is that a prolonged absence of NO synthase altered the signaling pathway downstream of cyclic GMP. Note that NO is required for LTD induction in cerebellar slices but not in cultured Purkinje cells. It has been shown that reagents that stimulate (sodium nitroprusside) or inhibit (hemoglobin, NG-nitro-l-arginine) NO signaling do not affect the reduced form of conjunctive LTD in cultured Purkinje cells (Linden and Connor, 1991). Also, LTD in cultured Purkinje cells derived from mice deficient in the neuronal isoform of NO synthase is indistinguishable from that in cultures from wild-type mice (Linden et al., 1995). This suggests that NO is not an indispensable mediator of conjunctive LTD, but is required to help mediators under certain conditions, such as in a slice preparation but not in a culture (Ito, 2002). Concerning the role of the NO-cyclic GMP-G-substrate cascade, a complication is that G-substrate-deficient mutant mice exhibit a unique age-dependent conjunctive LTD that is abolished at postnatal weeks 5 and 6. Before or after these weeks conjunctive LTD occurs normally (Endo et al., 2009). This finding indicates that the NOcyclic GMP-G-substrate cascade is also a modulator (not mediator) of conjunctive LTD because it is required for only a few postnatal weeks. δ2 receptors. The glutamate receptor δ2 (GluRδ2) is an orphan receptor predominantly expressed in Purkinje cells, in contrast to the δ1-subtype, which is found widely throughout the adult brain, but at low levels. The immunogoldlabeled δ1/2-subunits of glutamate receptors are present in parallel fiber-Purkinje cell synapses with a distribution pattern similar to AMPA receptors (Landsend

7 • Conjunctive Long-Term Depression (LTD)

77

et al., 1993; Petralia et al., 1998). In the dendritic spines of Purkinje cells, δ-receptors are anchored to actin filaments via spectrin, an actin-binding protein (Hirai and Matsuda, 1999; Hirai, 2000). However, when expressed alone or with other glutamate receptors, GluRδ2 does not form functional glutamate-gated ion channels, nor does it bind to glutamate analogs. Therefore, how GluRδ2 participates in cerebellar functions remains an open issue. Very recently, the ligand of δ2 receptors has been identified as cerebellin 1, a peptide secreted from granule cells. It binds directly to the N-terminal domain of GluRδ2. Exogenous application of cerebellin 1 to postsynaptic cells expressing GluRδ2 induces new synapses in vitro and in the adult cerebellum in vivo (Matsuda et al., 2010). Furthermore, another study revealed that GluRδ2 mediates cerebellar synapse formation by interacting with presynaptic neurexins through cerebellin 1 (Uemura et al., 2010). A role for δ2-receptors in LTD induction is suggested by three major findings. (1) Treatment of cultured Purkinje cells with an antisense oligonucleotide against the δ2-subunit mRNA blocked LTD induction without other noticeable effects on Purkinje cells (Hirano et al., 1994; Jeromin et al., 1996). (2) Purkinje cells in cerebellar slices (Kashiwabuchi et al., 1995) and tissue cultures (Hirano et al., 1995) derived from δ2-deficient gene-knockout mice did not exhibit LTD. (3) An antibody specific for the putative ligand-binding region of GluRδ2 induced AMPA receptor endocytosis, attenuated synaptic transmission, and abolished LTD (Hirai et al., 2003). Another open question is how the role of GluRδ2 in LTD is related to that in synapse formation and maintenance. Protein tyrosine kinases. The nonreceptor-coupled type of protein tyrosine kinase activity, a component of the intracellular signaling cascade in general, is prominent in the cerebellum. Purkinje cells express neuronal isoforms of c-src protein tyrosine kinase, pp60c-src (+) (Sugrue et al., 1990) and pp62c-yes, both being members of the src family (Zhao et al., 1991). Purkinje cells are also rich in tyrosine phosphatase, which antagonizes protein tyrosine kinases (Levy et al., 1993). Two protein tyrosine kinase inhibitors (lavendustin and herbimycin A) block conjunctive LTD (Boxall et al., 1996). Herbimycin A also prevents the depression of parallel fiber-EPSPs produced by intracellular infusion of a PKC activator, (-)-indolactan V. These findings indicate that protein tyrosine kinases, operating in association with PKC, are required for LTD induction. Protein tyrosine kinases may interact with PKC directly, but an indirect interaction via the G protein-PLC-DAG pathway is also possible because protein tyrosine kinase inhibitors block the Gq/11 proteincoupled receptor-mediated formation of IP3 (Umemura et al., 1999).

78

the cerebellum

Ca2+/calmodulin-dependent protein kinase (CaMKII). This serine/ threonine protein kinase is expressed at high concentrations and preferentially in neurons. One of the two isoforms, α, is predominantly expressed in the forebrain (a:b ratio, 3:1), whereas the other, β, is expressed in the cerebellum (α:β ratio, 1:4) (Miller and Kennedy, 1985). In the cerebellum, CaMKIIα is specifically concentrated in Purkinje cell somata and dendrites (Walaas et al., 1988). CaMKIIβ appears to function as an F-actin targeting molecule for localizing CaMKIIα/β hetero-oligomers to dendritic spines (Shen et al., 1998). Even though arachidonic acid and its metabolites are known to inhibit CaMKII (Piomelli et al., 1989), the involvement of CaMKII in conjunctive LTD has been unclear. A very recent report has suggested, however, that CaMKIIβ is important in controlling the direction of plasticity at parallel fiber-Purkinje cell synapses because a stimulus that induces synaptic depression in wild-type mice results in synaptic potentiation in CaMKIIβ knockout mice and vice versa. Possibly, this enzyme regulates calcium signals to produce synaptic depression or potentiation (van Woerden et al., 2009). Protein phosphatases. Two major types of serine/threonine-specific protein phosphatase are PP1 and PP2. PP1 specifically dephosphorylates the b-subunit of phosphorylase kinase and is inhibited by inhibitor-1 and inhibitor-2. In contrast, PP2 dephosphorylates the a-subunit of the phosphorylase kinase preferentially and is unaffected by the above-mentioned inhibitors. PP2 comprises three enzymes; PP2A, PP2B, and PP2C (Cohen, 1989). PP2A, like PP1, does not require cations for its activity and is sensitive to okadaic acid, whereas PP2B, also called calcineurin, is Ca2+/calmodulin-dependent and much less sensitive to okadaic acid. PP2C is Mg2+-dependent and insensitive to okadaic acid. PP1 consists of isoforms with different catalytic units. A strong immunoreactivity for a catalytic unit of PP1, PP1γ1, but not PP1δ or PP1α, has been observed in Purkinje cells, including their fine dendritic branches and spines (Hashikawa et al., 1995). Immunoreactivity for PP2A (α and/or β) was observed in Purkinje cell somata and thick dendrites, but not in the fine dendrites and spines. PP2Aα and PP2Aβ mRNAs are expressed in Purkinje cells (Abe et al., 1994). PP2B is present in the cerebellum (Sola et al., 1999). The phosphorylated G-substrate is a potent inhibitor of PP1 and PP2A (Endo et al., 2003). As would be expected, specific inhibition of postsynaptic PP2A by fostriecin or cytostatin induced a gradual and use-dependent decrease of synaptic current evoked by the stimulation of a single granule cell mimicking conjunctive LTD (Launey et al., 2004).

7 • Conjunctive Long-Term Depression (LTD)

79

c-Fos and jun-B. The induction of a reduced form of conjunctive LTD by the coadministration of AMPA and 8-Br-cyclic GMP to cerebellar slices is accompanied by an enhanced expression of c-Fos and Jun-B in Purkinje cells (Nakazawa et al., 1993). This suggests a role for active transcriptional complexes such as cFos/Jun-B in LTD induction. Application of the same compounds to the surface of the cerebellum in vivo, in conjunction with electrical stimulation of climbing fibers, induced the expression of Jun-B in Purkinje cells (Yamamori et al., 1995). Climbing fiber-parallel fiber conjunctive stimulation in the cerebellum in vivo also induced Jun-B expression, which was blocked by a nitric oxide synthase inhibitor (Yano et al., 1996). Jun-B/c-Fos forms an AP-1 complex, which acts as a transcriptional factor (Morgan and Curran, 1989). Possibly, this transcriptional factor plays a role in conjunctive LTD induction. Glial fibrillary acidic protein (GFAP). This intermediate filament protein is specifically expressed in astrocytes. In the cerebellum, it is expressed substantially in Bergmann glia, a subset of astrocytes. GFAP-deficient mice exhibit normal cerebellar architecture and synaptic transmissions, but not conjunctive LTD (Shibuki et al., 1996). What is missing in GFAP-deficient mice in terms of signal transduction for LTD induction is as yet unclear. However, because the membranes of Bergmann glia possess AMPA receptors and glutamate transporters, it may well be assumed that those of the Bergmann glia that respond to parallel fiber and/or climbing fiber activation secrete an as-yet-unknown diffusible factor, which may be required for LTD induction. Corticotropin-releasing factor (CRF). This neuropeptide containing 41 aminoacids is generally known as a stress hormone that plays a role in responses of the body to stress and is involved in affective disorders. CRF is contained in climbing fibers and cells of their origin in the inferior olive (IO), whereas Purkinje cells highly express CRF receptor type 1 (CRFR1) mRNA (Potter et al., 1994). Because specific CRFR1 antagonists, α-helical CRF-(9-41) and astressin, effectively blocked conjunctive LTD induction, CRF released from climbing fibers is critically involved in LTD-inducing mechanisms at parallel fiber synapses (Miyata et al., 1999). As would be expected, LTD was no longer observed when climbing fibers had been deprived by 3-aminopyridine intoxication, and it was restored by CRF replenishment. Because CRFR1 antagonists blocked LTD induced by conjunction of parallel fiber stimulation and depolarizing pulses without stimulating climbing fibers, it appears that CRF spontaneously released from climbing fibers, regardless of conjunctive stimulation, is sufficient for LTD induction. CRF release from climbing fibers may be a kind of stress response because errors represented by

80

the cerebellum

climbing fiber impulses imply discrepancy between intended and actual movements, and between demand for precision and discontent about inaccurate performance (Ito, 2009). However, what mechanisms organize these activities is unclear when climbing fibers and other CRF-containing neurons located in the amygdala and hypothalamus have no obvious neuronal connections. Insulin-like growth factor-1 (IGF-1). This basic peptide is involved in cell growth and differentiation and it operates at diverse synaptic sites. Purkinje cells exhibit IGF-1 immunoreactivity localized in the rough endoplasmic reticulum (Aguado et al., 1992). In the human cerebellum, IGF-1 immunoreactivity is pronounced throughout Purkinje cells and their extrusions, and it is also observed in the IO (Aguado et al., 1994). Electrical stimulation of the IO significantly increases the IGF-1 level in the cerebellar cortex. Whether IGF-1 is actually involved in LTD induction is uncertain, but because IGF-1 stimulates DAG production (Kojima et al., 1990), it is possible that DAG, in turn, stimulates PKC, which is required for LTD induction. IGF-1 may be involved in the maintenance of dendritic spine morphology because an IGF-1 antisense oligonucleotide injected into the IO was shown to induce a significant reduction in the size of the dendritic spines on Purkinje cells. Antisense oligonucleotide was also shown to evoke a significant and reversible decrease in IGF-1 level in the contralateral cerebellum (Nieto-Bona et al., 1997). Brain-derived neurotrophic factor (BDNF). Evidence is scarce regarding the involvement of this factor in conjunctive LTD induction. Nevertheless, when quisqualate application induces LTD, a significant increase in the level of BDNF’s mRNA expression occurs in cerebellar tissues, with a peak 4 hours after the application (Yuzaki et al., 1994). Even though the major source of this expression level increase is its granule cell fraction, the Purkinje cell fraction also contains BDNF’s mRNA. If BDNF is co-induced with LTD, it might play a role in the later phases of LTD.

7-5 Summary The basic properties and underlying signal transduction of conjunctive LTD have been analyzed in great detail. The complexity of this signal transduction should be a safeguard for its robust operation under variable conditions. The possibility that conjunctive LTD plays a crucial role in motor learning is examined in later chapters. The pharmacological and genetic tools required for analyzing such roles are now becoming available.

8 Multiplicity and Persistency of Synaptic Plasticity

8-1 Introduction In addition to conjunctive LTD in Purkinje cells, which was discussed in Chapter 7, “Conjunctive Long-Term Depression (LTD),” a rich variety of synaptic plasticity subtypes and underlying signal transductions has been revealed throughout cerebellar neuronal circuits. These subtypes should play their individual roles in cerebellar function. In this chapter, we address the above and ask critical questions about how conjunctive LTD persists and the extent to which it contributes in the formation of a memory trace.

8-2 Synaptic Plasticity in Purkinje Cells Homosynaptic LTD in parallel fiber-Purkinje cell synapses. LTD is induced in these synapses when a relatively large set of parallel fibers is stimulated repetitively (Hartell, 1996). Nevertheless, as long as parallel fibers are stimulated moderately, LTD occurs only after the conjunctive stimulation of parallel fibers and a climbing fiber. This ensures that conjunctive LTD is a prevailing physiological process in cerebellar networks and associated with a unique structure, that is, climbing fibers (Chapter 7). Homosynaptic LTD at parallel fiber-Purkinje cell synapses has been demonstrated to occur in the part of the fish cerebellum-like tissues that are devoid of climbing fibers (Bell et al., 1997) (see also Chapter 15, “Internal Models for Voluntary Motor Control,” Section 8). Long-term potentiation (LTP) in parallel fiber-Purkinje cell synapses. Presynaptic LTP occurs in parallel fiber-Purkinje cell synapses when parallel fibers are stimulated at 4–8 Hz, without being paired with climbing fiber activation. This 81

82

the cerebellum

form of LTP requires the activation of protein kinase A (PKA) by cyclic AMP but not NO (Salin et al., 1996; Lev-Ram et al., 2002). The knockout of the gene encoding the active zone protein, RIM1α, blocks presynaptic LTP. It is rescued, however, by the presynaptic expression of RIM1α. These findings suggest that the PKA-mediated phosphorylation of RIM1α at a single N-terminal site triggers presynaptic LTP (Lonart et al., 2003). A similar LTP occurs at parallel fiber-stellate cell synapses (Lachamp et al., 2009). Lev-Ram et al. (2002, 2003) demonstrated that LTP occurs postsynaptically in parallel fiber-Purkinje cell synapses after parallel fiber activation at 1 Hz unpaired with climbing fiber activation. Postsynaptic LTP does not require cyclic AMP or cyclic GMP acting on these synapses, but it does require NO activation, for example, when it is induced by bath application of an NO donor (NOR3) (Lev-Ram et al., 2002). Such NO-induced LTP requires soluble N-ethylmaleimide-sensitive fusion (NSF) protein attachment receptor (SNARE) proteins and GluR2-NSF interaction at the synapses (Kakegawa and Yuzaki, 2005). When postsynaptic LTD was saturated by repeated conjunction, a single bout of parallel fiber stimulation or application of NO reversed it and enabled conjunctive LTD to be re-induced (LevRam et al., 2003). Conversely, when postsynaptic LTP was saturated one bout of conjunction revitalized fresh postsynaptic LTP. Such interactions did not occur between presynaptic LTP and conjunctive LTD. The NO-evoked postsynaptic LTP was shown to be caused by excessive incorporation of AMPA receptors into the postsynaptic membrane of parallel fiber-Purkinje cell synapses (Yamaguchi 2009). Cannabinoid-receptor-mediated presynaptic LTD. The brain as a whole contains endogenous cannabinoids and their receptors. There is a particularly high density of these receptors in the cerebellum, largely in the molecular layer. As a result, cannabinoid intoxication leads to the degradation of motor coordination. Parallel fiber-evoked EPSCs in Purkinje cells are strongly inhibited by the bath application of a cannabinoid receptor agonist, and this inhibition is completely blocked by a cannabinoid CB1 receptor antagonist (Lévénès et al., 1998; Takahashi and Linden, 2000). The retrograde regulation of the endocannabinoid-mediated presynaptic LTD is mediated by postsynaptic mGluR1s (Maejima et al., 2001). Either postsynaptic Ca2+ concentration elevation or activation of Gq/11-coupled receptors induces the release of endocannabinoids (endogenous cannabinoids). When these two factors coincide, endocannabinoid release is markedly enhanced, this being attributed to the Ca2+ dependence of phospholipase Cβ (PLCβ) (Hashimotodani et al., 2007). A released endocannabinoid acts on parallel fiber cannabinoid receptors to inhibit Ca2+ influx into parallel fibers and thereby depresses the release of glutamate from parallel fibers (Brown et al., 2004).

8 • Multiplicity and Persistency of Synaptic Plasticity

83

Cannabinoid agonists impair LTD (Levenes et al., 1998) resulting, for example, in impaired eye-blink conditioning (Steinmetz and Freeman, 2010). In a recent study using gene-manipulated mice, the retrograde messenger was identified as 2arachidonoylglycerol (2-AG), which is produced by synthesizing the enzyme diacylglycerol lipase α (DGLα) (Tanimura et al., 2010). Homosynaptic LTD in climbing fiber-Purkinje cell synapses. Climbing fiber-Purkinje cell synapses undergo a modest degree of postsynaptic LTD after brief 5 Hz stimulation of climbing fibers (Hansel and Linden, 2000). LTD at both climbing fiber- and parallel fiber-Purkinje cell synapses requires an elevated postsynaptic Ca2+ concentration, and activation of mGluR1s and PKC. Rebound potentiation. A prolonged potentiation of basket/stellate cellmediated IPSPs in Purkinje cells is induced by coactivation of climbing fibers. This rebound potentiation is input-nonspecific (Kano et al., 1992). It is caused by a Ca2+dependent upregulation of postsynaptic GABA receptor function, and it involves the protein phosphatases, 1 and 2B, CamKII, PKA, DARPP-32 (a substrate of PKA and calcineurin, which inhibits protein phosphatase 1 when phosphorylated by PKA), and the GABAB receptor (Kawaguchi and Hirano, 2002). Rebound potentiation might cooperate with LTD in depressing the parallel fiber-mediated activation of Purkinje cells (Yamamoto et al., 2002). Synaptic plasticity induced by fear conditioning. The pairing of acoustic and nociceptive stimuli leads animals to express fear responses to otherwise neutral acoustic stimuli. In cerebellar slices obtained from fear-conditioned rats, both excitatory and inhibitory transmission to Purkinje cells were potentiated for up to 24 hours after conditioning (Scelfo et al., 2008). How these changes are induced is presently unknown. One possibility is that the neuromodulation is mediated by a hypothalamic neuropeptide (Chapter 4, “Input and Output Pathways in the Cerebellar Cortex”).

8-3 Synaptic Plasticity in Basket/Stellate Cells LTP and LTD in parallel fiber-basket/stellate cell synapses. Burst stimulation of a parallel fiber bundle unpaired with climbing fiber activity induces a long-lasting decrease in the size of the receptive fields of inhibitory interneurons. At the same time, it causes a large increase in the size of the receptive fields of Purkinje cells (Jörntell and Ekerot, 2002, 2003). In contrast, parallel fiber stimulation paired with climbing fiber activity induces a long-lasting increase in the size of receptive fields in interneurons, whereas it causes a long-lasting decrease in the size of the receptive fields of Purkinje cells. These findings suggest that parallel

84

the cerebellum

fiber synapses in inhibitory interneurons undergo LTP when parallel fiber synapses in Purkinje cells undergo LTD, and vice versa. Such a reciprocal pattern may have a synergistic effect, which augments the information storage capacity of cerebellar cortical networks. Alternatively, the parallel fiber-Purkinje cell synapses and parallel fiber-basket/stellate cell synapses may share different roles in the operation of cerebellar circuits. For example, a computer simulation study adopted such an assumption that synaptic plasticity first developed in parallel fiber-Purkinje cell synapses, and with additional training, this was transferred to parallel fiberbasket/stellate cell synapses for long-term memory storage (Kenyon, 1997).

8-4 Synaptic Plasticity in Other Cerebellar Cortical Synapses LTP in mossy fiber-granule cell synapses. High-frequency mossy fiber stimulation induces LTP in mossy fiber-granule cell synapses, providing postsynaptic NMDA receptors are activated. This LTP is accompanied by significant NO release in the granular layer, which is also dependent on NMDA receptor activation as well as NO synthase activation (Maffei et al., 2003). LTP is one of the processes that regulate the degree of divergence from a mossy fiber to granule cells (D’Angelo et al., 1999) in addition to the enhancement of intrinsic excitability (Armano et al., 2000) and Golgi cell inhibition (Chadderton et al., 2004). Long-term changes in Golgi cell responses. Effects of conjunctive stimulation of climbing fibers and peripheral afferents on Golgi cells have been examined (Xu and Edgley, 2008) because anatomical evidence has shown that Golgi cells are contacted by climbing fibers, and because climbing fiber signals may be transferred to Golgi cells by spillover of the transmitter or via synapses with NG2+ glial cells (Chapter 5, “Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex”). After the conjunction, Golgi cell responses to peripheral afferent stimulation were significantly reduced. The reductions developed progressively over 20 minutes of conjunctive stimulation and were persistent for up to 84 minutes. These results are in keeping with the idea that cerebellar circuits contribute to learning mechanisms.

8-5 Synaptic Plasticity in Cerebellar/Vestibular Nuclear Neurons LTD in inhibitory synapses. Inhibitory synapses of cerebellar nuclear neurons, presumably supplied by Purkinje cell axons, undergo postsynaptic LTD (Morishita and Sastry, 1996). This results from an input-nonspecific decrease in postsynaptic GABA sensitivity (not depending on the activation of GABAA receptors), which is

8 • Multiplicity and Persistency of Synaptic Plasticity

85

caused by an increase in intracellular Ca2+ concentration and activation of protein phosphatases. It may be worthy noting that this form of LTD explains the seemingly puzzling previous observation that destruction of the inferior olive depresses Purkinje cell inhibition on vestibular nuclear neurons remotely (Ito et al., 1979). Probably, in the absence of climbing fiber activity, Purkinje cells increased discharge (Colin et al., 1980) and this in turn caused LTD in the Purkinje cell inhibition of nuclear neurons. LTP/LTD in excitatory synapses. LTP has also been observed to occur in the interpositus and vestibular nuclear neurons of both acute and chronic rat in vivo preparations. This is seen when stimulating pulse trains are applied to the white matter at the entrance of inferior peduncle fibers to the cerebellum. The LTP lasted for at least 8 days in all but one of the preparations (Racine et al., 1986). However, a more recent study in rat in vitro preparations demonstrated that highfrequency burst stimulation of mossy fibers in the white matter, either alone or paired with postsynaptic depolarization, induced LTD of the mossy fiber-interpositus neuron synapse (Zhang and Linden, 2006). This form of LTD was blocked by infusion of a Ca2+ chelator through a micropipette to the interpositus neuron and required activation of mGluR1 and protein translation. In vestibular nuclei, a component (N2) of the field potentials evoked by electric stimulation of the vestibular nerve has been considered to represent the monosynaptic activation of vestibular nuclear neurons. After sustained optokinetic stimulation was applied, the N2 potential was enhanced, thereby suggesting the occurrence of LTP (Shutoh et al., 2006) (Chapter 10, “Ocular Reflexes”). Rebound depolarization. A robust rebound depolarization observed in cerebellar nuclear neurons following the offset of current-induced membrane hyperpolarization (Chapter 6, “Pre- and Post-Cerebellar Cortex Neurons”) has been related to an unusual form of synaptic plasticity. Pugh and Raman (2006) recorded EPSCs elicited in cerebellar nuclear neurons and showed that they were potentiated only when high-frequency synaptic activation was coupled with sufficient postsynaptic hyperpolarization evoked by Purkinje cells. This potentiation occurred as long as rebound current was in operation and even when postsynaptic spiking was prevented by a voltage-clamp. These data support Medina and Mauk’s (1999) postulate adopted in a large-scale computer simulation of cerebellar circuits. They proposed that inhibition from Purkinje cells controls potentiation of mossy fiber synapses in cerebellar nuclear neurons (see Chapter 9, “Network Models,” Section 7 for a possible learning rule in cerebellar nuclear neurons).

86

the cerebellum

8-6 Persistency of Synaptic Plasticity Activity-dependent synaptic plasticity is generally considered to be a transitory cellular process, which is eventually consolidated into a permanent memory trace, albeit by way of an as-yet-unknown mechanism. Information in at least initial memory is conventionally supposed to be stored as a patterned set of activitydependently modified synapses in a neuron or a neuronal network. On the basis of this Hebbian model of memory, efforts have been devoted to exploring the longlasting effects of activity-dependent synaptic plasticity. Several relevant findings are presented below. How long does synaptic plasticity persist? Using current technology, it is usually difficult to continuously follow a neuronal event, such as conjunctive LTD, for more than 2 hours. In Purkinje cells co-cultured with granule cells, miniature EPSCs (mEPSCs) can be intermittently observed by repenetration of the test cell. Murashima and Hirano (1999) found under this specific condition that a reduced form of conjunctive LTD lasted longer than 36 hours but shorter than 48 hours. Under the assumption that conjunctive LTD is required for adaptive behavior in reflexes, adaptation of optokinetic eye movement suggests that conjunctive LTD in the flocculus lasts one day, whereas LTP in vestibuloocular relay neurons lasts one week (Shutoh et al., 2006). A study of eye-blink conditioning suggests that the effect of conjunctive LTD lasts even up to one month (Chapter 10). Spine configuration. When one is studying the potential relation between synaptic plasticity and memory processes, a critical question is whether this plasticity involves an obligatory structural change in the tested synapse. Indeed, in dendrites of pyramidal neurons, LTP is associated with spine addition and enlargement, and LTD with spine loss and shrinkage (Lippman and Dunaevsky, 2005). In Purkinje cell dendrites, however, neither local synaptic induction nor global chemical induction of LTD changed spine number and size (Sdrulla and Linden, 2007). In contrast, a significant decrease in the number of excitatory synapses in the outer layer of the cerebellar cortex occurred by the third day of eyeblink conditioning in rabbits. This change was presumably brought about by the actions of conjunctive LTD (Connor et al., 2009). During long-term adaptation of the OKR, the number of parallel fiber-Purkinje cell synapses also decreased significantly in the mouse flocculus (by ~1/3 after 5 days). This certainly suggested that conjunctive LTD was followed by a loss of spine synapses (Nakadate et al., 2004). Silent synapses. A rather surprising recent finding is that stimulation of a single granule cell produced a detectable EPSC in only a small fraction of those Purkinje cells whose dendritic trees were traversed by the parallel fibers of the stimulated cell. Because ~54% of traversing parallel fibers form synaptic contacts

8 • Multiplicity and Persistency of Synaptic Plasticity

87

with one Purkinje cell, it was suggested that up to 80% of parallel fiber-Purkinje cell synapses are silent or have a very low transmission efficacy (Isope and Barbour, 2002). In another experiment on a sagittal slice of the rat cerebellum (Wang et al., 2000b), parallel fiber stimulation enhanced Ca2+ transients in a dendritic area 11 micrometers in radius (380 micrometers wide) containing 1,140–2,660 parallel fiber-Purkinje cell synapses (as determined from the anatomical finding that 1 um2 of a dendritic area contains 3–7 of these synapses). In contrast, parallel fiber stimulation evoked 60–370 picoampere EPSCs in a Purkinje cell. Because unitary parallel-fiber synaptic current is 12 picoamperes, 5–30 parallel fiberPurkinje cell synapses must have been activated. Hence, the electrical stimulation of parallel fibers in a small dendritic area of a Purkinje cell activates at the most 2.6% (30/1,140) of the parallel fiber-Purkinje cell synapses located within the test area. These findings are consistent with another observation obtained by an entirely different approach (Ekerot and Jörntell, 2001a,b). In the forelimb area of the C3 zone, each Purkinje cell responded via a mossy fiber-parallel fiber pathway to stimulation of only one of the 30–40 types of receptive fields that cover the forelimb skin in its entirety. The receptive fields of Purkinje cells vary even between neighboring Purkinje cells located along the same parallel fiber bundle. Hence, each Purkinje cell is activated via only ~2.5%–3% (1/40–1/30) of the total parallel fiber input received from forelimb skin. Thus, the three types of measurement described previously were consistent in their indication that each Purkinje cell computes on the basis of a small fraction of parallel fiber-Purkinje cell synapses. Of the 175,000 parallel fiber synapses each Purkinje cell receives (Chapter 4), 5,000 could thus be functional. Also surprising is the conspicuous effect on Purkinje cells of the repetitive high-frequency stimulation of parallel fibers (trains of 15 pulses @ 100 Hz were repeated at 1/3 s for 10 minutes) when unpaired with climbing fiber activation (Jörntell and Ekerot, 2003). A long-lasting, very large increase in the receptive field size of Purkinje cells then develops over 2–6 hours. In contrast, the stimulation of parallel fibers paired with climbing fiber activity reverses such changes in the size of the receptive field. The same repetitive stimulation of parallel fibers induces opposite effects on inhibitory interneurons. On the basis of these findings, Ekerot and Jörntell (2003) suggested that silent synapses are produced functionally by LTD and that learning converts silent synapses to active ones by the induction of LTP, the reverse occurring for LTD. The predominance of functionally silent parallel fiber synapses in Purkinje cells may imply that up to 97% of these synapses are long-term depressed during repeated learning trials. During this time “neurocomputing circuits” are emerging from the original randomly connected parallel fiber synapses.

88

the cerebellum

8-7 Protein Synthesis Whereas short-lasting memory is linked to functional changes in existing synapses, long-lasting memory is thought to be associated with a structural change such as the number of synaptic connections. The protein synthesis required for development and maintenance of synaptic structures should play a key role in the production and maintenance of a long-lasting memory (see Kandel, 2009). Proteins are synthesized not only in the somata but also locally in the dendrites of neurons in association with synaptogenesis and synaptic plasticity (Schuman et al., 2006). Accordingly, dendrites contain many different mRNAs, and the translational machinery includes ribosomes, polyribosomes, and elongation factor 2, as well as the endoplasmic reticulum and cisternae of the Golgi apparatus. In studies related to protein synthesis, translational inhibitors (anisomycin, puromycin, and cycloheximide) and transcriptional inhibitors (actinomycin D and 5,6-dichloro-β-D-ribofuranosyl-benzimidazole, DRB) are commonly used. The translational inhibitors depress conjunctive LTD in cultured Purkinje cells only in its late phase. This suggests a role for protein synthesis in sustaining conjunctive LTD, that is, providing proteins for structural changes that take place as activitydependent synaptic plasticity evolves (Linden, 1996). However, a puzzling observation was that the entire phase of conjunctive LTD in cerebellar slices was abolished by a 5-minute perfusion of translational inhibitors during conjunctive stimulation (Karachot et al., 2000). When a translational inhibitor was applied after a delay of 15 minutes or more, it no longer blocked conjunctive LTD. It appears that in slice conditions, a “quickly-turning-over” protein is required for the induction of LTD, probably in addition to that required for maintenance of LTD during its late phase. The composition of such a protein is not yet known, however. Cyclic AMP response element-binding protein (CREB). This nuclear protein regulates the transcription of genes with a CRE site in their promoter (Silva et al., 1998). When synaptic activation causes a high intracellular Ca2+ concentration, CREB is phosphorylated at the site of its transcriptional regulatory residue, serine 133, by involving CamKIV. When transfected with a dominant inhibitory form of CREB, which prevents DNA binding of endogenous CREB, or with dominant-negative constructs of CamKIV, Purkinje cells failed to develop the late phase of LTD, just as they did under the influences of transcriptional inhibitors (Ahn et al., 1999). CREB and CaMKIV are therefore suggested to contribute to the late phase of LTD. Genes encoding the a and b polypeptides of CaMKIV are highly expressed in cerebellar granule cells, but their expression in Purkinje cells is still unclear (Sakagami and Kondo, 1993). Nevertheless, CREB is a survival factor

8 • Multiplicity and Persistency of Synaptic Plasticity

89

for many neurons, and its deletion results in massive neuronal apoptosis. This might complicate the effects of CREB deletion on synaptic plasticity. Serum response factor (SRF) is another transcription factor that is regulated by synaptic activity, but its deletion in the adult brain does not cause neuronal death or gross malformation. Rather, such deletion produces a near-complete blockade of the induction of immediate early genes, including those that encode the synaptic protein, Arc. Smith-Hicks et al. (2010) secured the following findings. Deletion of the Arc gene blocked the late phase of LTD in cultured mouse cerebellar Purkinje cells, and the inhibition of SRF or its cofactor MAL displayed similar effects. Furthermore, when Arc-/- cells were transfected with a wild-type Arc, the late-phase LTD was rescued. However, mutation of one SRF-binding site in the Arc promoter, SRE 6.9, blocked this rescue. Co-transfection of wild-type Arc and SRF (engineered to bind mutated SRE 6.9) restored late-phase LTD in Arc-/-, SRE 6.9 mutant cells. Thus, SRF binding to SRE 6.9 in the Arc promoter is required for the late phase of cerebellar LTD.

8-8 Summary Various subtypes of synaptic plasticity found to this point must surely play their respective roles in the operational mechanisms of cerebellar neuronal circuits. The relationship between activity-dependent synaptic plasticity and memory traces remains unclear, and this issue stands out as a pressing future problem. Indeed, there seem to be endless possibilities for the discovery of novel components in cerebellar neuronal circuits!

9 Network Models

9-1 Introduction Having just considered the diverse cellular and molecular events occurring in the cerebellum, we can now examine how information is processed in its neuronal circuits. Since the monumental Marr and Albus contributions, various models have been proposed. In this chapter, we see how our understanding of such models has advanced recently.

9-2 Mossy Fiber-Granule Cell Relay Mossy fibers, the major inputs to the cerebellar cortex, make synaptic connections with the most numerous small neurons in this cortex, the granule cells. These connections follow a characteristic asymmetric pattern of divergence and convergence; one mossy fiber branches to contact with 400–600 or more granule cells, whereas each granule cell is contacted by 4–5 mossy fibers (cat, Palkovits et al., 1971; rat, Jakab and Hamori, 1988). Marr (1969) assumed that afferent information communicated by mossy fibers to the cerebellar cortex was returned to a combination of small subsets of active mossy fibers that he called “codons.” Each codon represented a feature of the input. The size of a codon that could activate a given granule cell varied from cell to cell, depending on its threshold, which could be regulated by Golgi cell inhibition or LTP (Chapters 5 and 8). A large-scale simulation of codon behavior was performed subsequently and it generally supported Marr’s model (Tyrrell and Wilshaw, 1992). Various ideas have been proposed for the functional significance of several mossy fibers converging onto single granule cells. Each granule cell may sample different types of mossy fiber input and associate them in patterns that are discriminated later by Purkinje cells. The small convergence number in the mossy fibergranule cell pathway may imply a “sparse coding” mechanism, which features 90

9 • Network Models

91

strong activation of a relatively small number of granule cells for each item of information. This would help transmit a complete contextual account of mossy fiber activity to a Purkinje cell and thereby minimize interference between the tasks being learned, a process that would augment information storage capacity (Brunel et al., 2004; Philipona and Coenen, 2004). On the other hand, recent physiological analysis of in vivo single granule cell discharge has not always supported the preceding models. Chadderton et al. (2004) reported that sensory stimulation produced bursts of mossy fiber EPSCs that summated to trigger bursts of granule cell spikes. This discharge pattern was evoked by only a few quantal EPSCs. However, spontaneous mossy fiber inputs did not trigger spikes unless Golgi cell inhibition of granule cells (Chapter 5, “Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex”) was reduced. These observations suggest that mossy fiber-granule cell synapses are characterized by a balance between the exquisite sensitivity of granule cells to mossy fiber input and a high signal-to-noise ratio. In the anesthetized mouse in vivo preparation, individual flocculus granule cell discharge was shown to represent faithfully the direction and velocity of whole-body motion via bidirectional modulation of EPSC frequency (Arenz et al., 2008). On the basis of differences in EPSC waveform being the signature of individual mossy fiber-granule cell synapses, Arenz et al. (2008) suggested that multiple vestibular, visual, and/or eye-movement-related signals converged onto single granule cells. They also emphasized that mossy fibergranule synapses did not simply relay incoming mossy-fiber spike rates but rather integrated vestibular information with other stimulus features. In a whisker-related cerebellar area, however, Rancz et al. (2007) found in rat in vivo preparations that sensory stimulation produced spike bursting at very high instantaneous firing frequencies (>700 Hz) in single mossy fiber terminals. This, in turn, drove rapid bursts of EPSC discharge in granule cells. Even a single mossy fiber firing at in vivo rates could produce the burst discharge of granule cells in in vitro preparations. These various observations suggest that a single presynaptic mossy fiber afferent transmits a sensory message to a granule cell in a “detonator” manner. Jörntell and Ekerot (2006) and Bengtsson and Jörntell (2009b) reported various features of mossy fiber-granule cell synapses in the C3 zone of decerebrate cats. In this zone, each granule cell receives three to four mossy fiber afferents arising from the same small cutaneous receptive field of a forelimb, with each afferent generating large EPSCs on stimulation. These authors suggested that in this way granule cells enhanced signal-to-noise ratio and, as such, they operated as a noisereducing device. Bengtsson and Jörntell (2009b) also demonstrated that the tested granule cells maintained the firing pattern of mossy fibers. In this “similar” coding,

92

the cerebellum

as they called it, granule cells processed mossy fiber signals, while not distorting the latter’s temporal pattern of activity. The preceding four cases show that the pattern of mossy fiber-to-granule cell convergence varies between different areas of the cerebellum, as if to suggest that each subserves a unique functional requirement.

9-3 Purkinje Cell Models Purkinje cells receive mossy fiber signals via granule cells and a unique set of parallel-fiber axons, and their axons serve as the final common output path of the cerebellar cortex (Chapter 4, “Input and Output Pathways in the Cerebellar Cortex”). Whereas each parallel fiber supplies excitatory synapses to ~300 Purkinje cells, each Purkinje cell receives excitatory synapses from ~150,000 parallel fibers. However, only 3%–15% of the latter synapses are functional, the remainder being silent. Parallel fiber-Purkinje cell synapses are characterized by bidirectional synaptic plasticity, involving conjunctive LTD and nonconjunctive LTP (Chapters 7 and 8), these presumably being the major mechanisms of information storage in Purkinje cells. In recent investigations of network mechanisms in Purkinje cells, advanced simulation technology is utilized to reproduce Purkinje cell behavior from neuronal circuit models. To calculate the information storage capacity of Purkinje cells, Brunel et al. (2004) used the classical perceptron model, a prototypical single-layer feedforward network with excitatory weights (Rosenblatt, 1962; Minsky and Papert, 1988) (Recall that the initial simple perceptron model was introduced in Chapters 1 and 3). Brunel et al. (2004) assumed that the operation of parallel fiber-Purkinje cell synapses was binary (zero or one action potential) despite the fact that parallelfiber-evoked EPSPs have a slow temporal profile, which is quite unlike binary pulses. They assumed that these EPSPs were trimmed by the IPSPs evoked in Purkinje cells via basket/stellate cells with a brief delay after the EPSPs (Chapter 5). They confirmed that these IPSPs were recruited concomitantly with EPSPs over a relatively wide range of stimulus intensities such that even the weakest of parallel fiber inputs could generate brief depolarizing pulses. Brunel et al. (2004) reasoned that because of the biphasic EPSP-IPSP responses to granule cell stimulation, the depolarizing pulses effectively summed in Purkinje cells only when they occurred coincidently. In this way, in the presence of basket/stellate cell inhibition, only coincident granule cell inputs could sum to excite effectively a Purkinje cell. They found that coincident inputs summed reasonably linearly, which matched the linearity approximation in the perceptron model.

9 • Network Models

93

Brunel et al. (2004) assigned to the perceptron model a task that consisted of learning the largest possible number of random input/output associations given a particular reliability level. They calculated the distribution of synaptic weights associated with the maximal storage capacity in their model and found that it contained >50% of the silent synapses. This fraction increased with storage reliability. Silent synapses (Chapter 8, “Multiplicity and Persistency of Synaptic Plasticity”) therefore appeared to be a necessary byproduct of optimizing learning and reliability. The distribution of synaptic weight theoretically derived by Brunel et al. (2004) resembled that obtained experimentally for granule cells-Purkinje cell synapses (Isope and Barbour, 2002). It was estimated that a Purkinje cell could learn up to 5 KB of information in the form of 40,000 input-output associations. To estimate the pattern recognition capacity of Purkinje cells, a multicompartmental model was developed by De Schutter and Bower (1994a,b). This model of a Purkinje cell received parallel fiber synapses on 150,000 dendritic spines (Harvey and Napper, 1988, 1991). All the spines were activated independently by a random sequence of parallel fiber inputs, firing at an average rate of 0.3 Hz. This background excitation was balanced by tonic background inhibition, and the model Purkinje cell fired simple spikes at an average frequency of 48 Hz, just as in real Purkinje cells. For a similar Purkinje cell model, Walter and Khodakhah (2006, 2009) added experimentally determined response variability and pattern size based on a linear algorithm. Each pattern was generated by randomly selecting 650 different inputs from the entire pool of 150,000 inputs. This model’s response was a linear function of the strength of its input, with the 650 inputs increasing the model’s firing rate by 200 spikes/s. Learning occurred in this Purkinje cell model by a process that mimicked LTD of the parallel fiber-Purkinje cell synapses. The result was a 50% decrease in the strength of all the inputs that comprised the learned pattern. The model was also used for evaluating the capacity of a Purkinje cell to participate in pattern recognition. This was accomplished by altering the number of patterns the model had to learn and quantifying its ability to distinguish between learned and novel patterns. To quantify the latter, the authors calculated the resulting signal-tonoise (s/n) ratio of the maximum firing rate of the model in response to learned versus novel patterns. From the results of this calculation, Walter and Khodakhah (2009) concluded that their linear algorithm provided efficient pattern recognition.

94

the cerebellum

To create yet another unique Purkinje cell model, Steuber et al. (2007) postulated that optimal pattern recognition capacity was obtained if Purkinje cells encoded information using pauses in their discharge rather than acceleration. Such a pause was observed in computer-simulated or experimentally recorded Purkinje cell discharges after applying a strong shock to granule cells. The applicability of this model to real cerebellar tissues under natural conditions has been questioned, however, because a relatively milder and temporally dispersed (i.e., more natural) granule cell activation failed to cause such a pause (Walter and Khodakhah, 2009). To compare with the pause coding, linear encoding of information would enable cerebellar nuclear neurons to use a simple averaging mechanism. In contrast to encoding patterns with pauses, Purkinje cells using a linear algorithm could recognize a large number of both synchronous and asynchronous input patterns in the presence or absence of inhibitory synaptic transmission. Walter and Khodakhah (2009) emphasized that under all conditions, the number of patterns recognized by Purkinje cells using a linear algorithm would be greater than that achieved by encoding information in pauses. Finally, another Purkinje cell model has been proposed for fish. In certain areas of the cerebellum-like structures of electric fish, climbing fibers are lacking, and parallel fiber stimulation alone induces homosynaptic LTD in Purkinje-like cells. When a fish moves voluntarily, it causes water perturbations that stimulate lateral line organs to generate sensory signals, which eventually cause the fish to move. Homosynaptic LTD has been shown to play a role in forming negative images of such sensory signals. Adding the negative images to actual sensory signals minimized the neural response to predictable sensory features (Bell et al., 1997).

9-4 The Granule Cell-Golgi Cell Loop Golgi cells have characteristically large divergence and convergence numbers. Each Golgi cell receives ~4,788 excitatory inputs to its dendrites in the molecular layer from parallel fibers and also ~228 mossy fiber terminals onto its descending dendrites (Pelionisz and Szentágothai, 1973). Each Golgi cell, in turn, extends a broadly branching axon to up to ~5,700 granule cells (cat; Palkovits et al., 1971). These mutual connections form a loop involving granule cells and Golgi cells. Marr (1969) emphasized the action of Golgi cells in regulating codon size. More recent researchers, however, have considered the contribution of Golgi cells to be more

9 • Network Models

95

important for the generation of signals with temporal profiles, which are essential for determining the timing of movements. Unlike earlier models based on conduction delays in parallel fibers, arrays of elements with different time constants, or populations of elements oscillating at different frequencies, recent models assume that temporal coding emerges from the dynamics of those cerebellar circuits that include Golgi cells. Fujita (1982a) proposed an adaptive filter model of the cerebellum. It incorporated a phase-converter that consisted of the mossy fiber-granule cell-Golgi cell circuit (Figure 16). This phase converter generated a set of multiphase versions of the mossy fiber input. This set was, in turn, conveyed by a set of parallel fibers and converged through modifiable synaptic connections to Purkinje cells, which provide the output signals of the cerebellar cortex. The learning principle used by these modifiable connections was that in each Purkinje cell the weight of the parallel fiber synapses was reduced when they were activated conjunctively with signals from climbing fibers. This learning principle insured that a Purkinje cell output converged to the “desired response” in order to minimize the mean square error of the performance. Since then, attempts have been made to substantiate the possible clock function of the granule cell-Golgi cell-granule cell pathway. In Buonomano and Mauk’s (1994) model for eye-blink conditioning (Chapter 11, “Somatic and Autonomic Reflexes”), the population vector of granule cell activity encodes both the particular mossy fiber input pattern and the time since its onset. For example, assume in this model that a particular periodic mossy fiber input pattern activates a subset of granule cells. This would then activate a subset of Golgi cells, which, in turn, would inhibit another, partially overlapping subset of granule cells. When this granule cell-Golgi cell-granule cell negative feedback loop retains realistic divergence/convergence ratios between granule cells and Golgi cells, it would create a dynamic, nonperiodic population vector of granule cell activity. Thus, the population vector of granule cell activity would encode not just the constellation of stimuli impinging on the organism, but also the time since the onset of the stimuli, and thereby serve as a clock. Yamazaki and Tanaka (2005) proposed another unique clock model of the granule cell-Golgi cell loop (Figure 21). It featured a set of random connections that generated ensemble patterns of the discharge of activated granule cells. As shown in Figure 22, this assembly of discharge patterns continued to change unless

96

the cerebellum

it was reset by a large mossy fiber input to the pathway. The time-dependent changes of the discharge pattern ensemble represented the passage of time as in a clock. When this Golgi cell clock was incorporated into a neuronal circuit model for the eye-blink reflex, it reproduced appropriately timed-conditioned responses of the reflex.

Figure 21 The granule cell-Golgi cell loop as an internal clock.

Schematic of the granular layer network wherein granule cells (small white circles) receive mossy fiber inputs and excite a set of Golgi cells (large gray circles), which are aligned in parallel with the granule cells’ axons; that is, the parallel fibers. Note how a single Golgi cell inhibits, in turn, a set of granule cells within its axonal arborization (dashed rectangle). If random connections between granule cells and Golgi cells occur, this recurrent inhibitory network then acts as an internal clock (an idea of Yamazaki and Tanaka, 2005). (Courtesy of Tadashi Yamazaki.)

9 • Network Models

97

Figure 22 Ensemble patterns of granule cell discharge change continuously.

Shown are some results from a simulation study using the circuit shown in Figure 21. The temporal profile of activity is presented for 100 granule cells lined up along the ordinate. Each cell underwent random transitions between periods of active firing (solid lines) and rest (breaks in lines) during 1,000 time steps after mossy fiber stimulation. Because the various granule cells had different temporal active-inactive transition patterns, the ensemble pattern of active cells changed over time, and the same pattern did not appear more than once throughout the simulation period. Hence, there is a one-to-one correspondence between the ensemble pattern of active granule cells and a given time step within the simulated period. (From Yamazaki and Tanaka, 2005.)

9-5 The Inferior Olive-Climbing Fiber System In the resting state, climbing fibers discharge irregularly at a low rate (~1 Hz). When the inferior olive (IO) receives a stimulus, the rate of climbing fiber discharges increases or decreases transiently. This can be detected in peristimulus histograms after many repeated bouts of stimulation. IO neurons seem to recode the high-frequency information carried by their synaptic inputs in the resting state into stochastic, low-rate discharge (Schweighofer et al., 2004; Kitazawa and Wolpert, 2005). This keeps the probability low of conjunctive LTD occurring in Purkinje cells, that is, so that no learning occurs. When climbing fibers discharge at a higher rate locked to the stimulus pattern, conjunctive LTD is induced in those of the parallel fiber synapses that are undergoing correlated activation. The mechanism that

98

the cerebellum

switches between the resting and learning states appears to be based on gap-junction-mediated electrical synapses between IO neurons (Chapter 6, “Pre- and PostCerebellar Cortex Neurons”). Activity in these electrical synapses determines whether climbing fiber activity is rhythmic, random, or synchronous (Kitazawa and Wolpert, 2005). An accessory circuit, composed of GABA-containing neurons located in the cerebellar nuclei, also regulates the activity of IO neurons. These GABA-containing neurons may modulate electrical synapses in in vivo neurons by the release of GABA, which has been shown to increase nonjunctional membrane conductance by shunting between the neurons. A computer simulation study showed that this inhibition of climbing fibers by cerebellar output combines with LTD/LTP to selfregulate spontaneous climbing fiber activity to an equilibrium level at which LTP and LTD are in balance and the expected net change in parallel fiber-Purkinje cell synaptic weights is at zero. This may be a unique network mechanism for regulating the modes of activation of in vivo neurons, and thereby regulate learning in cerebellar circuits (Kenyon et al., 1998a,b). Another accessory circuit is the excitatory projection to the IO from certain midbrain neurons located from the rostral pole of the red nucleus to the caudomedial border of the thalamus. These neurons are located primarily in the parvocellular red nucleus, the nucleus of Cajal, the nucleus of Darkschewitsch, and the caudomedial extremity of the subparafascicular nucleus (Saint-Cyr and Courville, 2004). This accessory circuit may supply positive feedback to in vivo IO neurons, thereby antagonizing inhibition mediated by cerebellar nuclei. These inhibitory and excitatory inputs to IO neurons (Figure 23) seem to modify the in vivo network that generates climbing fiber signals and hence regulates LTD-based learning. Heslow and Ivarsson (1996) demonstrated that cerebellar output during performance of a conditioned response inhibited the in vivo IO. Thus, studies of eye-blink conditioning have provided an opportunity to define the roles of the nucleo-olivary inhibitory projection, as explained in Chapter 11.

9 • Network Models

99

Figure 23 A microcircuit in the inferior olive.

Three IO neurons and their gap junction couplings are illustrated. “Half” junctions extending toward the outer edges indicate that the gap junction couplings include additional IO neurons that are not depicted in the drawing. Also illustrated are an excitatory and an inhibitory afferent. Some of the latter terminate precisely on the sites of gap junctions. Symbols: + in a circle, excitatory synapse; - in a circle, inhibitory synapse. (From Bengtsson and Jörntell, 2009a.)

9-6 Multilayered Network Models Albus (1971) developed a simple perceptron model of the cerebellum based on the principle of a pattern-classification device (Figure 6). Later, he elaborated the model mathematically and proposed a cerebellar model articulation controller (CMAC) for application to engineering problems (Albus, 1975). He suggested that the mossy fiber to granule cell to Golgi cell input network performed an expansion recoding that enhanced the cerebellum’s pattern-discrimination capacity and the learning speed of Purkinje cells. Parallel fiber synapses of the dendritic spines of Purkinje cells, basket cells, and stellate cells were all postulated to have their own variable responses to climbing fiber activity. This variability was assumed to be the mechanism of pattern storage. Albus (1971) speculated further that in order for the learning process to be stable, pattern storage had to be accomplished principally by weakening synaptic weights rather than by strengthening them. This viewpoint was supported later by the finding of conjunctive LTD in the cerebellum.

100

the cerebellum

In more general terms, a Purkinje cell in an adaptive filter model acquires a filtering function on the basis of multiple pairs of input signals (via parallel fibers) and corresponding desired output signals (via a climbing fiber). Adaptive filter models are advantageous in being able to accommodate temporally varying signals. This is not possible when using a simple perceptron model. Whereas a simple perceptron model is suitable for discrimination of spatial patterns of mossy fiber inputs, an adaptive filter model is better for recognition of the temporal patterns of these inputs. Indeed, an adaptive filter model has been shown to be effective for simulating adaptation in the VOR (Fujita, 1982b; Dean et al., 2010). As emphasized by Irry and Keele (1989), the cerebellum has a unique capability of timing to contribute to skills in general. This capability could arise from the adaptive filter network of the cerebellum. A liquid state machine model has been proposed for describing computations in biological networks of neurons, particularly for adaptive computational systems (Maass et al., 2002). It provides a method for employing randomly connected circuits in meaningful computations. It also provides a theoretical context for heterogeneous, rather than stereotypical, local gates or processors to increase the computational power of a circuit. It is also a method for multiplexing different computations (on a common input) within the same circuit. Yamazaki and Tanaka (2007) applied such a model to the cerebellum. In Figure 24, external contextual information is fed to a network called a “reservoir,” in which a number of neuronal units are randomly connected to constitute a random recurrent network. The number of reservoir neurons is assumed to be much larger than that of afferent inputs so that the spatiotemporal activity pattern of reservoir neurons encodes afferent input signals in much higher dimensions. (“Dimension” refers to the number of parameters used to define the state of the network.) Then, the signals mapped in the reservoir and external error signals are fed to neuronal units called “readouts.” These readout neurons extract time-varying information using the error and produce the final output. It designates the functional role of the granular layer as a reservoir and the Purkinje cells as the generator’s readout neurons. It bears emphasis that a liquid state machine possesses a more powerful information processing capability than does a perceptron.

9 • Network Models

101

Figure 24 The structure of a liquid state machine model of the cerebellum.

This model shows external contextual information being fed to the recurrent network shown in Figure 21 and called here a “reservoir.” It features a number of neurons that are connected randomly. In such a reservoir, the activity pattern of afferent input signals is represented by the population activity of reservoir neurons that evolves spatiotemporally. Because the number of reservoir neurons is assumed to be much greater than that of afferent inputs, the reservoir expands the dimension of input signals into a much higher one by means of the population activity of reservoir neurons. Another group of neurons called “readouts” receives the signals mapped in the reservoir (just as do Purkinje cells, and they also receive external error signals from teacher neurons in the IO. The readout circuit undergoes supervised learning by way of these external error signals and projects the desired signals (specified by the error signals) as the final output of the cerebellar cortex in response to the external contextual information. (Model of Yamazaki and Tanaka, 2007; courtesy of Tadashi Yamazaki.)

9-7 Nuclear Circuits Evidence presented in Chapters 8, 10, and 11 indicates that in addition to the cerebellar cortex, vestibular and cerebellar nuclei also provide memory sites. Why is such a dual memory system required? One prevailing view is that synaptic plasticity in the cerebellar cortex serves for quick and short-lasting adaptation that is needed from time to time, whereas synaptic plasticity in the nuclei sustains the memory required for slow and long-lasting adaptation. Another view is that relatively simple nuclear circuits are capable of retaining the memory of single parameters such as a gain increase or decrease but not the memory required for precise timing. The latter is presumably stored in more sophisticated cortical circuits. A role of nuclear memory in “savings” (the ability to relearn faster than the first time) has been demonstrated in modeling studies (Medina et al., 2001; Masuda and Amari, 2005).

102

the cerebellum

To model the cortical and nuclear dual memory mechanism in the cerebellum, Yamazaki and Nagao (2008) proposed a hybrid model that used a liquid state machine for the cerebellar cortex and a simple perceptron for vestibular and cerebellar nuclei (Figure 25). The authors assumed that teacher signals for the simple perceptron were derived from Purkinje cell output as in the Medina and Mauk’s (1999) model, but with different cellular processes. The hybrid model enabled the computer simulation of short- and long-term gain adaptation of the OKR.

Figure 25 Hypothetical hybrid organization of a cerebellar circuit.

This model involves a combination of the cerebellar cortex as a liquid state machine (Figure 24) and the precerebellar-cerebellar nuclei as a simple perceptron (Figure 5). Mossy fiber-cerebellar nuclear cell synapses are considered to undergo plastic changes using teacher signals fed by Purkinje cells to the cerebellar nuclei. (Courtesy of Tadashi Yamazaki.)

9-8 Microcomplexes Oscarsson (1979) delineated a narrow longitudinal zone as a structural-functional unit of the cerebellar cortex and called it a “microzone.” Today, we consider the microzone concept to also pertain to other small groups of neurons, for example, cerebellar and vestibular nuclear neurons as recipients of Purkinje cell inhibition and IO neurons as the source of climbing fibers (Figure 26). A small group of neurons of the parvocellular red nucleus that project to the neocerebellum via the IO may also be attached to a microzone (see also Color Plate V). These structures conjointly constitute what can be termed a “cortico-nuclear microcomplex” and shortened to microcomplex (Ito, 1984). The presence of such a microcomplex has been demonstrated in various regions of the cerebellum. In the intermediate part of the

9 • Network Models

103

cerebellar hemisphere, a microcomplex has been visualized by a combination of zebrin immunohistochemical and tracer techniques (Pijpers et al., 2005). Injections of small amounts of biotinylated dextran amine into either the medial or dorsal accessory olives label strips of climbing fibers and also patches of climbing fiber collaterals in the interpositus nucleus. On the other hand, gold-lectin injection into selected parts of the interpositus nuclei retrogradely labels both Purkinje cells and in vivo neurons. On the basis of the cellular data reviewed in Chapters 4–6, microcomplexes seem to function as follows. They transform mossy fiber inputs to nuclear neuron outputs and are equipped with an in vivo input for instruction on error learning via climbing fibers. They also have neuromodulatory inputs for switching the behavioral expression of their neuronal circuits. Assuming that a microzone occupies an area of ~0.2 × 50 millimeters squared (Andersson and Oscarsson, 1978), the human cerebellum, which is 50,000 millimeters squared in area, is composed of ~5,000 microcomplexes. Each of these complexes is presumed to be incorporated into a functional system composed of the spinal cord, brainstem, and cerebral cortex (Ito, 1984).

Figure 26

A cerebellar microcomplex functioning as an embedded computing system.

A, This schematic details the major components of a cerebellar microcomplex. From left to right, these components and their role include the following: Input, via precerebellar neurons; Mode setting, beaded fibers release a monoamine or a peptide as a neuromodulator (e.g., serotonin, orexin) to modify the mode of operation of the circuits of the microcomplex; MF, mossy fibers serve as a major input to the cerebellar cortical circuit; LTD, long-term depression serves as a memory process; CF, climbing fibers convey an error signal; IO, the inferior olive serves as a teacher; Error signal, informs about errors in the performance of the entire system, including the microcomplex; Output, signals from nuclear neurons that are sent from the cerebellum to postcerebellar neurons; VN/CN, vestibular and cerebellar nuclear neurons.

104

the cerebellum

Mapping of the flocculus effects in response to various electrical microstimulations revealed microzones associated with each component of the VOR (horizontal, vertical, and rotatory) (Nagao et al., 1985), an eye blink and neck muscle contraction (Nagao et al., 1984), and a rise in mean blood pressure (Nisimaru et al., 2010) (Figure 29). The microcomplexes involved in the VOR have been analyzed with various techniques, as will be reviewed in Chapter 10, “Ocular Reflexes”; in particular that in the horizontal VOR has been labeled systematically by retrograde transneuronal tracing with the rabies virus from the medial rectus muscle in guinea pigs (Graf et al., 2002). In another study, the microcomplex structure in the C3 zone of the anterior lobe of the cerebellum has been analyzed intensively in connection with the withdrawal reflex (Ekerot et al., 1995; Garwicz et al., 1998; Jörntell and Ekerot, 2003). The C3 zone, which receives mossy fiber and climbing fiber inputs from receptive fields in the forelimb skin, contains 30–40 longitudinal microzones lying side by side, each 50–150 micrometers wide. Purkinje cells and inhibitory interneurons in each microzone receive climbing fiber inputs from the same cutaneous receptive field. Climbing fibers in adjacent microzones are activated from adjacent skin areas, forming a detailed somatotopic map of the ipsilateral forelimb’s skin, particularly its distal parts. Adjacent microzones innervate, in turn, adjacent cell groups in the anterior interpositus nucleus. Through further projections to the red nucleus, these microzones control movement components that have specific relationships with the location of climbing fiber receptive fields (see Chapter 11).

9-9 Summary Computational modeling has been fruitful in reproducing theoretical principles conceived for complex neuronal networks. Bottom-up, experiment-based realistic simulation has also been of value. Unique models of the cerebellum have thus been developed. Given the complex diverse features of the cerebellum, however, further advanced models are clearly needed. There is also a possibility that hardware models of the cerebellum can be developed using silicon analogues of neurons (Rachmuth and Poon, 2008).

10 Ocular Reflexes

10-1 Introduction An eyeball is often likened to a single-joint limb, with three degrees of freedom for movement (horizontal, vertical, rotatory). For securing stable accurate vision, however, eyeballs are equipped with four major reflexes. These are evoked by vestibular and visual afferent signals, and they compensate for movements of the head and its environment. In this chapter we review current knowledge about these ocular reflexes and their cerebellar control. The four reflexes provide relatively simple control system models of cerebellar adaptation and compensation, and they have helped considerably in the design of robust experimental paradigms for testing several aspects of the function of the cerebellum.

10-2 The Vestibuloocular Reflex (VOR) The VOR is mediated by a tri-neuronal arc consisting of vestibular afferents, VOR relay neurons, and motoneurons supplying extraocular muscles. To control eye movement and eye position in three-dimensional space, the input VOR afferents are those that originate in the three semicircular canals (horizontal, anterior, posterior) and the two otolith organs (utricle, saccule) in each labyrinth (Uchino et al., 1996; Isu et al., 2000). The output of VOR motoneurons acts on six extraocular muscles (medial/lateral rectus, superior/inferior rectus, superior/inferior oblique) for each eye (Figure 27). Under experimental conditions, yaw, pitch, and roll of the head induce, respectively, the horizontal, vertical, and rotatory components of the VOR. Utriculo- and sacculo-ocular reflexes are evoked by linear motion and static tilt of the head. Under freely moving conditions, the VOR is mediated by the concerted operation of the parallel pathways that link the ten sensors in two labyrinths to the twelve muscles in the two eyes (Ezure and Graf, 1984b). 105

106

the cerebellum

Figure 27 Canal specific pathways of vestibuloocular reflexes.

The A–F drawings show VOR pathways converging onto six different extraocular muscles including A, Sr (superior rectus); B, Ir (inferior rectus); C, So (superior oblique); D, Io (inferior oblique); E, Mr (medial rectus); and F, Lr (lateral rectus). Other abbreviations: Ac, anterior canal; E 1-6/I 1-6, secondary vestibular neurons (E, excitatory; I, inhibitory); FL, flocculus; Hc, horizontal canal; MV, medial vestibular nucleus; Pc, posterior canal; SV, superior vestibular nucleus. (From Ito et al., 1977.)

Among the component pathways of the VOR, those functioning in the horizontal VOR have been investigated the most extensively. Hence, unless otherwise indicated, the VOR discussed in this chapter is for its horizontal component. The horizontal semicircular canals are stimulated by ipsilateral head rotations, and they send their neural signals via the primary vestibular nerve fibers to VOR relay cells located in the vestibular nuclei. These cells send, in turn, either excitatory or inhibitory signals to the lateral/medial rectus motoneurons located in the abducens/oculomotor nuclei. The excitatory transmitter for the primary vestibular nerve is glutamate (Raymond et al., 1984), and those for the secondary VOR relay neurons are glutamate and aspartate (Kevetter and Hoffman, 1991). The inhibitory transmitter for the secondary VOR relay neurons is glycine (Spencer et al., 1989). When the head is rotated to one side, contraction of the ipsilateral medial rectus and contralateral lateral rectus muscles and relaxation of the ipsilateral lateral rectus and contralateral medial rectus muscles cause both eyes to rotate in the direction opposite to that of head rotation.

10 • Ocular Reflexes

107

Velocity storage. Head rotation in darkness at a constant velocity induces a nystagmus that consists of alternating slow phases representing the VOR and quick phases that reset eye position. In this situation, impulses evoked in primary vestibular afferents attenuate rapidly within 5 seconds. In parallel, however, the slow-phase velocity of nystagmus decays slowly with a time constant of ~20 seconds. To explain this discrepancy, the VOR arc has been assumed to involve a velocity storage mechanism. It might store the initial head velocity as transduced by the semicircular canals and maintain it despite decay in the firing rate of the canal afferents (Raphan et al., 1979). Because electrical stimulation of the nodulus and uvula induces a rapid decrease in horizontal slow-phase velocity, these cerebellar regions might control velocity storage (Solomon and Cohen, 1994) in contrast to the control of VOR gain by the flocculus. Neural integrator. In vestibular nuclei and motoneurons for extraocular muscles, the head velocity signals generated by the labyrinth are converted to eyevelocity and eye-position signals. Hence, the VOR arc must be equipped with a neural integrator (Cannon and Robinson, 1985), which, for the horizontal VOR, may involve the nucleus prepositus hypoglossi and/or commissural inhibitory projections between bilateral vestibular nuclei (Arnold and Robinson, 1997). The integrator for the vertical VOR is provided by the interstitial nucleus of Cajal in the midbrain (Fukushima et al., 1992). Velocity storage and neural integrator action may represent functions of the same neuronal circuit, but this has yet to be clarified.

10-3 VOR Adaptation and the Flocculus Figure 28 illustrates the wiring diagram of the VOR pathway and its connection with the flocculus, fitted to a framework of an adaptive control system. Vestibular nuclear neurons relaying the VOR arc placed as the controller, whereas motoneurons, extraocular muscles, and eyeballs are taken together as the controlled object. We found early on that Purkinje cells in the flocculus directly inhibited relay neurons of the VOR (Fukuda et al., 1972; Kawaguchi, 1985). This led to the proposal of the flocculus hypothesis—that is, that this structure controls adaptively the VOR (Ito, 1982). One aspect of this hypothesis was that Purkinje cell inhibition was exerted on one of the two VOR arcs converging onto each extraocular muscle (Figure 27; Ito et al., 1977). These Purkinje cells were shown to form VOR-specific microzones in the flocculus, as mapped in Figure 29. Microzones have now been demonstrated in various species (rabbits, Nagao et al., 1984; Van der Steen et al., 1994; De Zeeuw et al., 2004; cats, Sato et al., 1983, Sato and Kawasaki, 1984;

108

the cerebellum

guinea pigs, Graf et al., 2002; rats, Billig and Balaban, 2005; mice, Schonewille et al., 2006). Retrograde transneuronal tracing using the rabies virus from the left medial rectus muscle demonstrated vividly the H-zone in the flocculus (Graf et al., 2002).

Figure 28 Wiring diagram for cerebellar control of the horizontal vestibular ocular reflex.

This drawing shows the tri-neuronal arc of the horizontal VOR (broken line rectangle cs) and the flocculus superimposed upon it. The broken line rectangle mc encloses eVN and iVN (excitatory and inhibitory vestibular nuclear neurons) as a controller and the flocculus which provides its adaptive mechanism). The box on the right encloses the controlled object component. Other abbreviations: AOS, accessory optic system; CF, climbing fibers; IO, inferior olive; LR, lateral rectus muscle; MF, mossy fibers; Mr, medial rectus muscle; PC, Purkinje cells; VA, vestibular afferents. Symbols: filled circles, excitatory connections; square, inhibitory connections. Note that the broken line rectangle contains the microcomplex, which acts as an adaptive controller.

The marked adaptation in horizontal VOR gain was demonstrated when the vestibular-visual relationship was changed using Dove prism goggles or magnifying lenses, or by exposing an animal to an in-phase/out-of-phase combination of horizontal turntable movements and screen rotation (Chapter 3, “The Cerebellum as a Neuronal Machine”). Adaptation has also been demonstrated to occur in the vertical VOR when an animal was fixed to a chair with the animal’s right or left side down and rotated horizontally in the center of a horizontally rotating screen (squirrel monkeys; Hirata and Highstein, 2001). In another study, cross-axis adaptation occurred in the cat’s horizontal VOR when horizontal turntable rotations (yaws) were combined with vertical optokinetic motions. This resulted in the VOR (measured in the dark) acquiring a vertical component. Also, the sensitivity of vestibular

10 • Ocular Reflexes

109

nuclear neurons increased significantly to yaw in the dark, but not to vertical plane rotations (Quinn et al., 1996). Similarly, cross-axis adaptation in the monkey was shown by having the animal track a vertically moving target spot synchronized with horizontal whole body rotation. Before training, the horizontal trapezoidal rotation resulted in a collinear VOR with a mean latency of 15 ms. After training, the collinear VOR remained unchanged but an orthogonal, cross-axis VOR developed. It had a mean latency of 42 milliseconds with a gain (eye velocity/chair velocity) of 0.2, which decayed with a mean time constant of 80 milliseconds (Sato et al., 1999).

Figure 29 Microzones mapped in the fllocculus.

Lateral surface of a rabbit flocculus on the left side. Stimulation through a glass pipette electrode evoked site-dependent effects from the shaded or dotted areas: horizontal (H), vertical (V), or rotatory (R) eye movement, eye blink (B), contraction of neck muscles (N), or a transient increase in the mean blood pressure (S). Based on Nagao et al. (1984, 1985) and Nisimaru et al. (2010).

Pharmacological or genetic methods have been used to block conjunctive LTD in order to test its potential role in horizontal VOR adaptation. For example, such adaptation was abolished in rabbits and monkeys when their flocculus was superfused with hemoglobin, which blocks conjunctive LTD induction by absorbing NO (Nagao and Ito, 1991). VOR adaptation was also blocked in mutant mice null-deficient in Purkinje-cell-specific PKG-1 (Feil et al., 2003) and in transgenic mice overexpressing in Purkinje cells the pseudo-substrate protein kinase C inhibitor (De Zeeuw et al., 1998). Conjunctive LTD was absent in cerebellar slices derived from these two types of mice. The signaling effects of flocculus Purkinje cells, which underlie VOR adaptation, have been analyzed as described in the following text.

110

the cerebellum

10-4 Neuronal circuit for VOR adaptation Vestibular mossy fiber input. The flocculus has long been known to be that part of the vestibulocerebellum that receives the projection of primary vestibular afferents in the form of mossy fibers (Brodal, 1972). This notion, however, was reexamined subsequently in various species. In rabbits, the presence of primary vestibular afferents to the flocculus was negated (Gerrits et al., 1989). Later, however, it was shown to engage 2% of the primary vestibular afferents, whereas 64%–89% of these afferents were shown to project to the nodulus and uvula (Barmack et al., 1993). In cats, the primary vestibular projection to the flocculus was described as being substantial but less pronounced than that to the nodulus and uvula (Korte and Mugnaini, 1979). In monkeys, the same projection to the flocculus was measured to be 6% (Nagao et al., 1997a), and in rats the value was 10% (Osanai et al., 2000). This relative paucity of primary vestibular afferents to the flocculus, however, could be supplemented by secondary projections from vestibular nuclei (Barmack, 2003). Head velocity-sensitive neurons have indeed been found in vestibular nuclei (Scudder and Fuchs, 1992). Some mossy fiber terminals in the monkey flocculus and ventral paraflocculus were shown to respond primarily to vestibular stimuli (Lisberger and Fuchs, 1978; Noda and Warabi, 1987; Markert et al., 1988). These observations support the idea that primary and/or secondary mossy fiber inputs convey vestibular signals to the flocculus. Climbing fiber input. A prominent pathway arises from the retina and reaches the flocculus in the form of climbing fibers (Maekawa and Simpson, 1973). This pathway is mediated by the nucleus of the optic tract and the accessory optic system (Simpson, 1984). The functional role of this pathway has been assumed to convey retinal slips as sensory errors to the flocculus. However, this idea was confounded by the finding that flocculus Purkinje cells often modulate climbing fiber discharges during a VOR in the dark (Ghelarducci et al., 1975; Belton et al., 2002; Simpson et al., 2002). This implies the involvement of vestibular/oculomotor components in the signals conveyed by the climbing fiber pathway. Normally, experiments on the VOR are complicated by the difficulty in isolating motor signals from sensory signals because the latter cause movements that generate motor signals. In recent studies, efforts have been made to break the normally tight relationship between instantaneous retinal slip and eye movement. The results of such experiments will be discussed in Chapter 12, “Adaptive Control System Models,” in relation to the general question of whether climbing fibers represent sensory or motor signals. Here, one may ask about the biological role of the climbing fiber discharges during a VOR in the dark. A relevant observation was that if rotation in darkness occurred immediately followed learning, the gain of the VOR reverted toward its

10 • Ocular Reflexes

111

prelearning value, thereby indicating that expression of the memory was disrupted. If, after gain-down learning, the cat spent another 60 minutes stationary without form vision (the cat was stationary either in the light with the lenses covered by filter paper or in complete darkness), subsequent disruption did not occur. This suggested that the memory had been consolidated (Titley et al., 2007). Eye-movement-related signals. In addition to the previously mentioned simple-spike discharges in response to vestibular mossy fiber inputs, Purkinje cells in the flocculus also exhibit simple-spike discharges representing either eye acceleration, eye velocity, or eye position (Mizukoshi et al., 2000; Omata et al., 2000; Hirata and Highstein, 2001; Blazquez et al., 2003). The possibility has been entertained that such eye-movement signals are in response to proprioceptive feedback from extraocular muscles. This was supported only in part because the blockade of muscle afferents by local anesthesia was shown to reduce simple-spike discharges related to eye velocity by only 31% (rabbits: Miyashita, 1984). This finding suggested that the remaining two-thirds of such discharge must have an intrinsic origin. This issue is still open, with the following three possibilities on the forefront of consideration. First, the source of such an intrinsic mechanism might be eye-velocity- and eye-position-sensitive neurons located in vestibular nuclei (Scudder and Fuchs, 1992). These neurons might convert head velocity signals to eye-velocity signals. Additionally, eye position signals might be generated from eye-velocity signals by means of a neural integrator, as defined above. It was also reported that some mossy fibers in the flocculus/ventral paraflocculus respond to eye movements induced by a saccade, head rotation, or smooth tracking of a moving target (Lisberger and Fuchs, 1978; Noda and Warabi, 1987). Hence, one hypothesis is that VOR relay neurons convert head velocity signals to eye-velocity signals and send them to the flocculus/ventral-paraflocculus via recurrent axon collaterals (Miles and Lisberger, 1981). The second possibility is that the simple spikes of Purkinje cells might express eye-movement-related signals received via mossy fibers from sources other than VOR relay neurons. For example, midline paramedian tract neurons were reported to send vertical eye-velocity signals to the flocculus via mossy fibers (Nakamagoe et al., 2000). It remains unknown, however, if other paramedian tract neurons mediate horizontal eye-velocity signals. The third possibility is that eye-movement-related signals are generated in the flocculus (Ito, 1998). Indeed, simple-spike activity in the flocculus changed rapidly preceding the eye-velocity change that occurred in response to a change in the direction of the visual surroundings. These rapid responses should represent

112

the cerebellum

sensory signals from which the eye-movement-related simple-spike activity was generated (cat; Mizukoshi et al., 2000). Such a conversion might be secured by error learning as driven by the climbing fibers’ retinal error signals. The latter contain eye-movement-related components because retinal errors represent a discrepancy of movements between the eyes and their visual surroundings relative to the head. This possibility is supported by observations about the OFR, as described later. In brief, simple-spike discharges in ventral paraflocculus Purkinje cells during the OFR are modulated to become a mirror image of the temporal profile of climbing fiber signals. This suggests that during the OFR, climbing fiber signals refine simple-spike activity, possibly by inducing conjunctive LTD (Shidara et al., 1993). Memory sites. Studies of the VOR have emphasized an important problem concerning its adaptation mechanisms. Several groups have considered whether the memory of VOR adaptation is stored in the cerebellar cortex, vestibular nuclei, or both (Miles and Lisberger, 1981; Lisberger and Sejnowski, 1993; Paster et al. 1994; Raymond et al., 1996; Ito, 1998; Broussard and Kassardjian, 2004). In evaluating this issue, it is well to remember that two phases of VOR adaptation, fast and slow, have been distinguished. In studies on goldfish (McElligott et al., 1998) and monkeys (Nagao and Kitazawa, 2003), the short-term (fast) VOR adaptation, which developed during about one hour of continuous rotation, was abolished by inducing localized anesthesia of the flocculus. In contrast, slow VOR adaptation was established by repeated trials of VOR adaptation for several days (van Alphen and De Zeeuw, 2002). The slow VOR adaption persisted after the flocculus was injected with either an AMPA/kainate receptor blocker, CNQX (cats; Kassardjian et al., 2005) or a local anesthetic (monkeys; Anzai et al., 2010). Studies on the OKR have also provided important results (Shutoh et al., 2006; see later). The implication of these results will be discussed later as common problems for the VOR, OKR, and OFR (Chapter 12, Sections 3-4).

10-5 Vestibular Compensation It has long been known that a unilateral labyrinthectomy causes the development of pronounced nystagmus (Vidal et al., 1998). After the lesion, the static deficits generally disappear in a few days, whereas restoration of the dynamic, vestibularrelated synergies is much slower and only partial. When the flocculus was lesioned unilaterally 40–70 days earlier, recovery from the effects of the contralateral labyrinthectomy was severely delayed. When such lesions were made after recovery for 16–60 months, they produced only a transient asymmetry of the vestibuloocular responses. These observations suggest that the flocculus is required for initiating

10 • Ocular Reflexes

113

(but not maintaining) the compensatory process following peripheral lesions of the vestibular system (cat; Courjon et al., 1982). The behavioral recovery from unilateral labyrinthectomy was accompanied by an asymmetric expression of isoforms of PKC in flocculus Purkinje cells. There was also a regionally selective increase in the number of PKC-immunoreactive Purkinje cells contralateral to the lesion (rat; Goto et al., 1997). In another study, the compensation was retarded after application of PKC inhibitors through the cerebral ventricle (rat; Balaban et al., 1999). It thus appears that conjunctive LTD, which requires PKC, has a role in vestibular compensation. Certain changes related to vestibular compensation have been found to occur also in the brainstem. Unilateral labyrinthectomy caused an increase in the number of GABAergic neurons in vestibular nuclei (cat; Tighilet and Lacour, 2001), an increase in mRNA coding BDNF in medial vestibular nucleus (mouse; Li et al., 2001), and an increase in glycinergic quantal current amplitudes and frequency of glycinergic quantal events in medial vestibular nuclear neurons (mouse; Lim et al., 2009). A neuronal network simulation also suggested a change in commissural inhibitory connections after this lesion (Graham and Dutia, 2001). Together with cerebellar adaptation by the flocculus, considerable reorganization of the vestibular neuronal network appears to underlie vestibular compensation.

10-6 Optokinetic Eye-Movement Response (OKR) The OKR moves an eye to follow a relatively slowly moving visual environment. When the subject moves in the light, this reflex and the VOR are evoked simultaneously, and they act synergistically. It has been shown that the VOR is especially effective for higher frequencies of a changing visual environment, whereas the OKR is more effective during lower frequencies. For any given frequency, the two systems in combination produce a rather constant gain (~0.8) within the frequency range tested (0.05–2 Hz) (rabbit; Collewijn and Grootendorst, 1979). The OKR shares a large part of the neuronal circuit for the VOR. Neurons responding to horizontal optokinetic stimuli have been located in the nucleus reticularis tegmenti pontis (NRTP) (rabbit; Kano et al., 1991). In another species, however, these neurons responded to both horizontal optokinetic and vestibular stimulation. During optokinetic stimulation, the response of these neurons was either unidirectional (51%) or bidirectional (49%). All such neurons were shown to exhibit a response during sinusoidal head rotation. Phase and gain analyses suggested that neurons in the (NRTP) convey a head velocity signal (rat; Taillanter and Lannou, 1988). Yet to be investigated is whether this nucleus contains a group of neurons specifically devoted to OKR adaptation.

114

the cerebellum

The OKR may appear to be a feedback control mechanism, but in view of the relatively long loop time (~50 ms) consumed for visual information processing, feedback control may be efficient only at a relatively low velocity of optokinetic stimulation. Furthermore, while the OKR has a relatively high open loop gain at a relatively low velocity of the optokinetic stimulus, the open loop gain was shown to decrease steeply at a relatively high velocity (Collewijn, 1969). With such a retinal delay and a low open loop gain, the OKR seems unable to perform efficient feedback control by itself at high stimulus velocities. The flocculus appears to intervene in the OKR to cope with these difficulties. The OKR has indeed been shown to exhibit an adaptive gain increase toward unity during prolonged sinusoidal rotation of the visual environment around a stationary subject (rabbit and mouse; Nagao, 1988, 1989; Katoh et al., 1998) (Figure 30). In another study, OKR adaptation was blocked by the local injection of a NO synthase inhibitor that blocked conjunctive LTD (Chapter 7, “Conjunctive Long-Term Depression (LTD)”). This adaptation was absent in mutant mice that lacked conjunctive LTD because of their lack of the neuronal isoform of NO synthase (Katoh et al., 2000). Also, it was shown that intravenous injection of cyclooxygenase-2 inhibitor (nimesulide), which blocked conjunctive LTD in cerebellar slices, effectively suppressed OKR adaptation (mouse; Le et al., 2010). It is also worth noting that LTD induction is facilitated in Delphilin-deficient mutant mice, in which the gain-increase adaptation of the OKR is also enhanced (Takeuchi et al., 2008). This may suggest that LTD induction at parallel fiber-Purkinje cell synapses is a crucial rate-limiting step in OKR adaptation. The OKR provides a convenient means of examining cerebellar adaptation because it is so easy to measure (Figure 30). Its short-term (fast) adaptation was shown to develop during a 1-hour screen rotation that diminished throughout the subsequent 24 hours. In contrast, long-term (slow) adaptation of the OKR was established by repeated bouts of screen rotation throughout seven days, and it persisted for a week even after the flocculus was injected with locally acting lidocaine. There are two possibilities for the relationship between fast and slow adaptation of the OKR. They may occur independently of each other, or consecutively (i.e., one after the other). It has recently been found that protein synthesis inhibitors injected locally into the flocculus had little effect on fast adaptation, but they blocked slow adaptation (Okamoto et al., 2011). This finding suggests that the memory of fast OKR adaptation formed in the flocculus may subsequently induce the memory of slow adaptation in vestibular nuclear neurons. Slow OKR adaptation has been shown to be underlain by LTP in vestibular nerve-VOR relay neuron synapses (mouse; Shutoh et al., 2006). A puzzling electron microscope study showed that fast OKR adaptation was accompanied by a significant decrease in the

10 • Ocular Reflexes

115

AMPA receptor density of parallel fiber-Purkinje cell synapses in the flocculus, whereas slow OKR adaptation was accompanied by a decrease in the number of parallel fiber-Purkinje cell synapses in the flocculus (mouse; Nakadate et al., 2004). These observations suggested that the flocculus maintained a trace of conjunctive LTD in the form of a loss of synapses, yet there was no sign of this trace in OKR gain. Clearly, more information is needed to advance understanding of the neuronal mechanisms that underlie fast and slow OKR adaptation.

Figure 30 Adaptation of the optokinetic eye-movement response (OKR).

In an awake mouse, eye movements were recorded via a half mirror using an infrared television camera for real-time recording. A left side, side view of the recording arrangement. IR, infra-red. A right side, top-down view of the animal’s relation to an oscillating screen. “Screen” indicates movement of the screen. B, OKR gain measured before () and after ( ) daily training for one hour, repeated for five consecutive days. These mice were kept in the dark during trials. Average values for 12 mice are plotted. Recovery thereafter is followed for two weeks in 4 mice ( ). The OKR gain increase during one hour of screen oscillation implies the occurrence of short-term adaptation (see downward arrows), whereas the increase over five days (thick arrow) indicates long-term OKR adaptation. C shows records of the eye movements of one mouse during screen oscillation, at the first, third, fourth, and sixth day of training. Data for ten or more cycles are averaged. (Courtesy of Soichi Nagao.)

•

•

116

the cerebellum

10-7 The Ocular Following Response (OFR) The OFR is another eye-movement reflex elicited by brief, unexpected movements of a visual scene (monkey; Kawano and Miles, 1986; Miles and Kawano, 1986). It may involve but is distinct from the OKR. This was revealed in a study wherein bilateral removal of the cerebral occipital lobes depressed the OFR to a residual response that resembled the OKR (monkey; Zee et al., 1987). OFRs involve Purkinje cells in the ventral paraflocculus (Shidara et al., 1993; Gomi et al., 1998). Unlike the OKR, however, the OFR has a visual perception mechanism located in the cerebral association cortex. Cells in this area were shown to discharge during brief, sudden movements of a large-field visual stimulus that elicited an OFR. These cells were located mainly in the medial superior temporal area (MST) and also partly in the middle temporal area (MT). The response properties of MST neurons during ocular following were similar to those of dorsolateral pontine nucleus neurons. This suggested that MST neurons relayed visual information to the ventral paraflocculus via dorsolateral pontine nucleus neurons (Kawano et al., 1994). The OFR is an example of a control system in which an elaborate cerebral cortical mechanism is incorporated as an accessory into the primary controller, which is located in the brainstem. At first glance the OFR operates by feedback, but in reality it is a feedforward mechanism. This is known because of its long loop time (50 ms; i.e., long for visual information processing across the retina and MT/MST areas) in response to 100 milliseconds ramp changes in the visual scene. One relevant study in monkeys, which was designed to induce OFR adaptation, involved application of doubleramp sequences: the first ramp to initiate the OFR and the second to apply visual errors that evoked adaptive mechanisms in the OFR. Both ramps lasted 150 milliseconds and were delivered at different speeds and directions (speed steps and direction steps, respectively). Repeated exposure to increasing versus decreasing speed steps increased and decreased, respectively, the OFR (Miles and Kawano, 1986). Another study analyzed simple spike discharges of Purkinje cells using the multiple linear regression technique for system identification. It was found that dorsolateral pontine nucleus neurons, the origin of mossy fiber inputs to the ventral paraflocculus, encoded some selective aspects of visual stimuli (Takemura et al., 2001). It was found also that simple spike activity in ventral paraflocculus Purkinje cells encoded eye movements in terms of their acceleration, velocity, and eye position, thereby representing inverse dynamics of the actual eyeball movements (Shidara et al., 1993). The meaning of these findings will be considered in Chapter 12.

10 • Ocular Reflexes

117

10-8 Integrated Control of Ocular Reflexes by the Cerebellum Whereas the VOR in rabbits is adaptively controlled by the flocculus, three specific types of ocular reflexes (VOR, OKR, and OFR) in monkeys are under the influence of a cerebellar area composed of the flocculus and ventral paraflocculus. How these reflexes are differentially represented in this area and integrated to perform purposeful eye movements is still a matter of conjecture. One difficulty is that the flocculus and ventral paraflocculus exhibit a particularly large species variation in folial morphology (Voogd, 2004). A practical solution for arguments of folial structures is to ignore them and lump these areas as one floccular complex, as adopted in physiological mapping in monkeys (Lisberger, 2009). For a radical solution, nevertheless, precise mapping is required for topographical demarcation of the cerebellar cortex in association with precise knowledge of the relevant neuronal circuits and functions. For example, the area defined as the proper flocculus receives major mossy fiber inputs from (1) the primary vestibular nerve, (2) vestibular nuclei, (3) the NRTD, and (4) the central part of the mesencephalic reticular formation. In contrast, the area defined as the ventral paraflocculus receives major mossy fiber inputs from pontine nuclei and NRTP (Gerrits and Voogd, 1989; Glickstein et al., 1994; Nagao et al., 1997a). It is apparent that the former area mediates the VOR/OKR, and the latter the OFR. On the other hand, Purkinje cells in the flocculus project to the medial and ventrolateral parts of the medial vestibular nucleus, superior vestibular nucleus, and the y group, which contain VOR relay neurons. Purkinje cells in the ventral paraflocculus also project to the medial and ventral parts of the medial vestibular nucleus, superior vestibular nucleus, and y group (Balaban et al., 1981; Nagao et al., 1997b). There is also a projection to the caudoventral part of the posterior interpositus and dentate nuclei. Hence, it is possible that the OFR is also mediated by VOR relay neurons. These findings are the basis for the assumption that the VOR, OKR, and OFR are integrated into a multi-input system sharing the controller and the controlled object (Figure 9A).

10-9 Saccadic Eye Movement A saccade is a quick, simultaneous movement of both eyes in the same direction to catch a visual target by small foveal areas of the retinae for high-acuity vision. It is part of the overall orienting response to the sudden onset of a novel or behaviorally interesting stimulus, which orients the eyes, external ears, head, and/or body toward the source of the stimulus. Saccades are the fastest movements produced by the human body. (The peak angular speed of the monkey eye during a saccade

118

the cerebellum

can reach 1,000 degrees/second). Saccades to an unexpected stimulus normally take ~200 milliseconds to initiate, and they last for ~20–200 milliseconds. Under certain laboratory conditions, the latency of saccade production can be cut nearly in half, that is, to ~100 milliseconds (for express saccades). The brainstem neuronal circuit for the saccade generator has been dissected in detail (see Moschovakis, 1996; Scudder et al., 2002; Ramat et al., 2006; Williams and Hilmas, 2010). In brief, omnipause neurons in the superior colliculus are activated by a visual stimulus and, in turn, activate excitatory burst neurons located mainly in the caudal pontine reticular formation and inhibitory burst neurons in the medullary reticular formation (Strassman et al., 1986a,b). For horizontal saccades, excitatory burst neurons activate ipsilateral abducens motoneurons (supplying the ipsilateral lateral rectus muscle) and also contralateral oculomotor neurons (supplying the contralateral medial rectus muscle), the latter effect being mediated by intranuclear neurons in the abducens nucleus. Inhibitory burst neurons inhibit contralateral abducens motoneurons and, at the same time, ipsilateral oculomotor neurons via intranuclear neurons. An excitatory/inhibitory neuron pair thus provides reciprocal innervation to appropriate extraocular muscles to make a saccade in one horizontal direction. Some of these neurons are included in a model of cerebellar control of the saccade generator (see Figure 40). A neural mechanism in the superior colliculus directs a saccade to a target in the visual field. Electrical microstimulation in a deep layer of the superior colliculus evokes a saccade, whose amplitude and direction are functions of the site of stimulation and independent of its intensity and frequency. As the stimulating site in the superior colliculus is moved caudally and medially, respectively, the amplitude of the saccade increases and the direction of the saccade shifts from downward to upward (for a two-dimensional motor map of saccades, see Robinson, 1972; Sparks and Nelson, 1987). The timing of the onset and cessation of the discharge of excitatory and inhibitory burst neurons varies according to the site of activation in the superior colliculus. The amplitude and direction of the saccade so induced is appropriate for a foveal catching of the target in the visual field. A group of neurons called long-lead burst neurons has been found to be concentrated in the rostral pons. These neurons are characterized by low-level, unstructured activity preceding (by ~30 milliseconds) the intense peri-saccadic burst. Long-lead burst neurons appear to play multiple roles in saccades; for example, some of them may mediate a trigger to suppress omnipause neurons’ discharge (see previous description) during small saccades (Kaneko, 2006). The involvement of the posterior vermis (lobules VI and VII) in saccade control has been determined by the observations that (1) local electrical stimulation of

10 • Ocular Reflexes

119

this area induced saccadic eye movements, and (2) this effect was lost when Purkinje cells in the same area were destroyed by a local kainate injection (Noda and Fujikado, 1987). Purkinje cells of the posterior vermis project their axons to the posterior part of the fastigial nucleus, which, in turn, projects to the previously described saccade-generator circuit in the lower brainstem. Saccadic eye movements are so rapid that their precision control cannot be based on visual feedback. Rather, cerebellar mechanisms are required for this purpose. For example, saccades were shown to become grossly inaccurate (hypometria) three days after ablation of the monkey cerebellum. Three or more months later, the average saccade amplitude was recovered near completely, but it remained more variable than before the ablation (Barash et al., 1999). When the GABA agonist, muscimol, was injected unilaterally or bilaterally into the caudal fastigial nucleus, the saccades became hyper- or hypometric and their trajectories and end points became more variable (Robinson et al., 1993). Posterior vermis lobules VI/VII and the caudal fastigial nucleus are therefore presumed to act on the saccade generator in the lower brainstem to make saccades more consistent and accurate. The adaptive control function of the cerebellum for saccades has been tested by a paradigm called “saccadic adaptation.” In control trials, a first jump of a visual target by 15° caused an accurate saccade. In adaptation trials, while a similarly induced saccadic movement was under way, the target was displaced from 15° to 20°, the resultant eye position ending 5° short of the target. After 1,000 consecutive adaptation trials, eye position overshot the original 15° position of the target and ended closer to the 20° position. Lesioning of the posterior vermis abolished permanently this saccadic adaptation (McLaughlin, 1967; Barash et al., 1999). Close connections of the cerebellum to the saccade generator have also been demonstrated by showing that a significant proportion of mossy fibers carries a signal very similar to excitatory burst neuron activity in the saccade generator (Ohtsuka and Noda, 1992). Moreover, HRP injected into the posterior vermis labeled, among many other brainstem structures, the paramedian pontine reticular formation (Yamada and Noda, 1987), which contains excitatory and inhibitory burst neurons (see previous description). The population response of a large group of Purkinje cells in the posterior vermis was shown to provide a precise temporal signature of saccade onset and offset. Modeling revealed that changing the relative contributions of individual Purkinje cells (i.e., by having them discharge at different times throughout the saccade) translated directly into changes in the amplitude of the saccade (Thier et al., 2000; Kojima et al., 2010). The IO plays a role in conveying an error signal in the current hypothesis of cerebellar learning (Chapters 3 and 9). The pathway that appears to convey such

120

the cerebellum

errors in saccadic adaptation was found to be located in the monkey midbrain tegmentum. Its weak electrical stimulation delivered ~200 milliseconds after a saccade in one horizontal direction produced progressively more marked changes in saccade gain, this being similar to changes induced by adaptation to real visual errors (Kojima et al., 2007). Even though the anatomy of this pathway has not yet been identified, a very likely candidate is the collicular output fibers that project to subnucleus b of the medial accessory nucleus of the inferior olive (MAO). This structure has been shown to project, in turn, to the posterior vermis (Yamada and Noda, 1987). A computer simulation using these data reproduced the adaptation of saccadic gain to repeated presentations of dual-step visual targets (see Chapter 12).

10-10 Summary Neuronal circuits for the four types of ocular reflex (VOR, OKR, OFR, and saccades) have been dissected in terms of both their structure and function. In addition, the roles and mechanisms of the cerebellum in their elaboration have been studied in detail. The findings to this point provide a firm basis for clarifying several adaptive control mechanisms of the cerebellum in Chapter 12.

11 Somatic and Autonomic Reflexes

11-1 Introduction In this chapter we will consider several types of somatic and autonomic reflexes whose operation is improved when modulated by the cerebellum. The prototypical adaptive cerebellar control of these reflexes is similar to that for the ocular reflexes analyzed in Chapter 10, “Ocular Reflexes.” In particular, eye-blink conditioning, together with VOR/OKR adaptation, provide excellent examples of cerebellar function.

11-2 The Stretch Reflex and Posture Under Selected Conditions A stretch reflex causes the contraction of a muscle in response to its stretch. Interestingly, its mediators are quite complex. First, three types of motoneurons are involved. α-motoneurons innervate extrafusal muscle fibers, whereas γ-motoneurons innervate intrafusal muscle fibers. In addition, β-motoneurons innervate both types of fibers. Second, three types of muscle afferents come into play (Ia, Ib, and spindle II afferents). Ia and Ib afferents are the fastest conducting myelinated axons, and they are subdivided by a combination of their physiological responsiveness to brief muscle stretch and conduction velocity. Spindle II afferents are the muscle spindle component of the group II afferent population (Matthews, 2010). Third, spinal interneurons come into play during long-latency components of the stretch reflex. Group Ia afferents arise from annulospiral endings in the middle area of muscle spindles. They make direct excitatory connections with the α-motoneurons of that muscle and its synergists (they also have disynaptic inhibitory connections with antagonist α-motoneurons that are discussed later and several other connections not covered in this chapter). Hence, the impulse signals of stretchactivated Ia afferents from a muscle may, under some conditions, induce 121

122

the cerebellum

monosynaptic excitation of α-motoneurons that, in turn, cause contraction of that muscle; that is, the short-latency Ia (also called “tendon tap”) stretch reflex, as it may be called, can provide a negative feedback control of muscle length (length feedback; see Nichols and Ross, 2009) (Figure 31A). In several vertebrates, including the human, this reflex can be demonstrated (again under certain conditions) by a brief tendon tap (stretch) or in the form of an H reflex, which results from applying an electric shock to largely the group Ia axons of a peripheral muscle nerve (Pierrot-Deseilligny and Burke, 2005).

Figure 31 Hypotheses on key neural control mechanisms for muscle activation.

Sketches show four hypotheses of particular relevance to this chapter. (A) Length follow-up servo. Commands first activate γ-motoneurons, followed by reflex activation of α-motoneurons. (B) Coactivation of both α- and γ-motoneurons for length servoassisted motion. (C) Negative force feedback from tendon organs and negative length feedback from muscle spindles cooperating to stabilize muscle stiffness. INN, inhibitory interneuron. (D) Fusimotor set of static (γ s) and dynamic (γ d) motoneurons. For further information, see the text. (Based on ideas of Loeb,1984; Nichols and Ross, 2009; Prochazka and Ellaway, 2011.)

Group Ib afferents, on the other hand, arise from Golgi tendon organs that sense the force of muscle contraction and, within the spinal cord, their impulse signals may, under some conditions, inhibit α-motoneurons via inhibitory interneurons. Hence, this type of Ib stretch reflex (with short and long latency components) may provide a negative feedback control (force feedback; see Nichols and Ross, 2009), which can cooperate with the Ia stretch reflex to regulate muscle stiffness (Figure 31C). However, whereas this control occurs during “resting” states in spinal and decerebrate cats, and several other species, the reflex effects of Ib afferents switch to excitatory during terrestrial locomotion (Pearson and Collins, 1993).

11 • Somatic and Autonomic Reflexes

123

This results in group Ib afferents providing positive force feedback, which strengthens the stance phase of the step (for further details, see Prochazka et al., 1997; Hultborn, 2001; Nichols and Ross, 2009). Based on these observations on Ia and Ib afferents, four hypotheses have been proposed to explain different, largely spinal neural control mechanisms for muscle contraction. Figure 31A shows the classical length follow-up servo, in which γ-motoneurons play the role of initiating muscle contraction; a change in their firing rates will cause a similar reflex-induced change in the activation level of extrafusal muscle fibers (Merton, 1953). The second hypothesis shown in Figure 31B is length-servo-assisted motion, in which α- and γ-motoneurons are coactivated by the same command signals (Granit, 1975). This process prevents the length-servo from resisting muscle contraction. The third hypothesis shown in Figure 31C involves negative force feedback from tendon organs and negative length feedback from muscle spindles cooperating to stabilize the stiffness of the muscle (Houk, 1979). As mentioned previously, this Ib effect changes to positive feedback during locomotion, thus assisting in the stance phase of the step. The fourth hypothesis of “fusimotor set” shown in Figure 31D is based on the finding that γ-motoneurons have two subtypes; static and dynamic. Activity of dynamic γ-motoneurons is enhanced in certain occasions when spindle endings may fire at very high rates during large fluctuations in muscle length (Prochazka et al., 1985). Fusimotor set may also be adjusted to optimize spindle sensitivity according to anticipated variations in kinematics for particular movements (Loeb et al., 1990). Each of these four hypotheses accounts for only a highly constrained class of muscle functions, and a general theory is still awaited (see Loeb, 1984; Prochazka and Ellaway, 2011). Spindle group II afferents arise from an area off the middle of muscle spindles. In the spinal cord, they make mono-, di-, and trisynaptic connections with αmotoneurons. In the spinal cord, they may excite or inhibit flexor muscles and do the reverse to extensor muscles, depending on the nature and context of the movement. For example, it has been shown in a study on human treadmill locomotion that when the soleus muscle was stretched by applying small amplitude dorsiflexion perturbations shortly after heel contact, short and medium latency responses were observed. The short latency response was velocity sensitive, whereas the medium latency one was not. The medium latency component was more sensitive than the short latency one to nerve cooling, but more resistant to ischemia. Two hours after the ingestion of tizanidine, an α2-adrenergic receptor agonist known to selectively depress transmission in the group II muscle afferent pathway, the

124

the cerebellum

medium latency reflex was strongly depressed, whereas the short latency component was unchanged. Based on these observations, it was suggested that the medium latency component of the stretch reflex represented responses of spindle II afferents induced by an unexpected perturbation (Grey et al., 2001). A general theory on spindle afferent II function is also yet to emerge. Myotatic reflexes. Each joint is attached by a group of agonist muscles on one side and another group of antagonist muscles on the other (recall Figure 3A). As stated previously, the short-latency part of the stretch reflex in a muscle evokes group-Ia afferent signals, which excite α-motoneurons of that muscle and its agonists. At the same time, the group-Ia afferent signals excite Ia inhibitory neurons mediating “Ia reciprocal inhibition” of α-motoneurons supplying antagonist muscles. In this way, the short-latency stretch reflex in a muscle is sometimes accompanied by the contraction of agonist muscles and the relaxation of antagonist muscles. Each joint is equipped with a pair of such stretch reflex-Ia inhibition combinations, this being called a myotatic reflex (Figure 32). In each joint of a limb, myotatic reflexes determine under some conditions the position of the joint by balancing two muscle groups—that is, extensor muscles acting to extend the limb and flexor muscles acting to flex the limb. Feldman’s equilibrium point hypothesis (1986) proposed that the position of each joint is set by the equilibrium of the spring-like tension-length relationship between two antagonistic muscles crossing the joint. It may further be assumed that, involving both the length- and force-feedback pathways from muscle spindles and Golgi tendon organs, respectively, the force-length relationship for each muscle will be determined by control signals from the CNS. The set of the equilibrium points so determined for the agonist-antagonist pairs of muscles crossing a joint represents the position of the joint, and changes in patterns of these equilibrium points in time would represent trajectories leading a movement (for a recent evaluation of Feldman’s hypothesis, see Nichols and Ross, 2009). Descending control. The functional pattern of excitatory inputs from several descending pathways (see following description) and sensory input received by motoneurons that supply one set of muscles appears to be replicated onto the Ia inhibitory interneurons projecting to the motoneurons that supply the muscles that are antagonist to the homonymous set. This similarity in the convergence onto “corresponding” motoneurons and Ia inhibitory interneurons led Lundberg and his colleagues to hypothesize that both motoneurons and Ia inhibitory interneurons are activated in parallel during voluntary movements to secure a coordinated contraction of agonists and relaxation of antagonists (Hongo et al., 1969; Lundberg, 1970). Myotatic reflexes and spinal pattern generators are modulated by several descending pathways including the cortico-, reticulo-, rubro- (weak in humans), and

11 • Somatic and Autonomic Reflexes

125

Figure 32 Wiring diagram for cerebellar control of the short-latency (tendon-tap) myotatic reflex.

Some CNS connections are shown for a pair of agonist (extensor) and antagonist (flexor) muscles when the agonist is exhibiting a tendon-tap reflex (correction) brought on by a perturbation. The right box shows that muscle spindles (sp) in the agonist activate their own motoneurons (m1 and m2) via Ia afferents (IaAF) and inhibit their antagonist motoneurons (m3) via Ia inhibitory interneurons (IaIN). The circuitry also includes Renshaw cells (Rc), which exert recurrent inhibition in the m3 motoneurons. This segmental circuit is controlled by Deiters neurons via the lateral vestibular nucleus (LVN) and the lateral vestibulospinal tract (LVST). The LVN is modulated by the B zone of the cerebellum, which receives proprioceptive and exteroceptive information via the dorsal and ventral spinocerebellar tracts (SCT). The B-zone and LVN forn a microcomplex, which serves as an adaptive controller for stretch reflexes. Proprioceptive information from muscles and tendons as well as cutaneous signals evoked by a perturbation of posture are transferred to the cerebellum via the spinoolivary tract (SOT). Additional abbreviations: CF, climbing fibers; MF, mossy fibers; PC, Purkinje cells.

vestibulo-spinal descending tracts. Among these, descending signals of the lateral vestibulospinal tract facilitate extensor motoneurons supplying hindlimb, forelimb, and neck motoneurons (Figure 32). (Note, however, that the monosynaptic linkage between vestibulospinal fibers and motoneurons is mainly with respect to neck motoneurons; Wilson and Yoshida, 1969.) Signals of the lateral vestibulospinal fibers also excite the Ia inhibitory interneurons responsible for Ia reciprocal

126

the cerebellum

inhibition from extensors to flexors (Grillner et al., 1971). In addition to these connections, various effects in terms of excitation/inhibition and latency were induced in spinal motoneurons by stimulating reticulospinal tracts (Peterson et al., 1979). Stimulation of the rubrospinal tract also induced excitation of flexor and inhibition of extensor motoneurons, but a mixture of EPSPs and IPSPs was found in many motoneurons (Hongo et al., 1969). These effects were mediated by segmental interneurons. It has also been shown that the lateral vestibulospinal tract and part of the reticulospinal tract project to contralateral motor nuclei via common commissural inhibitory interneurons (Krutki et al., 2003). It has recently been reported that locally injected orexin A excites lateral vestibular nuclear neurons and improves vestibular-related motor behavior in rats, including vestibular-mediated posture, motor balance, and negative geotaxis. To estimate the negative geotaxis, rats were placed on a 40° incline with their head pointing down the slope, and the time spent for a turn of 180° upward against the direction of the gravitational field was measured. The motor effects of orexin A were prominent when an animal faced a major motor challenge such as when rats traversed an inclined balance beam or stayed on an accelerating rota-rod, as compared to during rest and the execution of everyday movements (Zhang et al., 2011). Note also that orexins are also involved in defense reactions (Section 6). Cerebellar function. The possibility that stretch reflexes operate under the influence of the cerebellum via spinal descending tracts has been examined by recording a short-latency stretch reflex in a muscle stimulated by sinusoidal stretch at 0.1–10 Hz. The stretch reflex so evoked displayed a phase advance relative to the stretch stimuli. This advance was reduced by acute and chronic cerebellectomy (decerebrate cat; Higgins, 1987). Apparently, this effect reflects the contribution of the cerebellum to improve dynamic characteristics of the short-latency stretch reflex. In the dorsal area of the lateral vestibular nucleus of Deiters, neurons of origin of lateral vestibulospinal tract fibers receive Purkinje cell inhibition directly from the B-zone of the cerebellum (Figure 32). These dorsal Deiters neurons also receive excitatory synapses from spinal ascending tracts (Akaike et al., 1973). This shows that Deiters neurons mediate a supraspinal loop superposed on a segmental stretch reflex, and thereby mediate an adaptive action from the B-zone as illustrated in Figure 32. Stretch reflexes may also be under the influence of the cerebellar A-zone, which, via the fastigial nucleus, regulates adaptively pontine reticulospinal tract axons (Ito et al., 1970). In humans, ankle stretch evokes a reflex contraction of calf (soleus and gastrocnemius) muscles. In view of its long latency (120 ms), this response must be mediated by a supraspinal pathway, which is called the “functional (or long-latency) stretch reflex.” This reflex has task and context dependencies. For example, Nashner

11 • Somatic and Autonomic Reflexes

127

(1976) showed that when a human standing upright on a platform was subjected to a sudden backward movement of the platform, the functional stretch reflex was evoked in calf muscles, which helped reduce postural sway. On the other hand, when the platform was suddenly inclined forward and upward, the reflex was depressed because it would have enhanced a backward postural sway. Following an unexpected change in the usefulness of stretch reflexes, certain subjects progressively altered reflex gain during the succeeding three to five trials, always in the direction that optimized the reflex response. The cerebellum appeared to be involved in assessing the stabilizing effect of the functional stretch reflex because clinically diagnosed cerebellar deficits diminished the extent of reflex adaptation (Nashner, 1976). The functional stretch reflex might be produced in a neuronal circuit involving the B-zone, but this issue is still open. A subtle mechanism, presumably involving the cerebellum, has been suggested to operate in humans standing freely while maintaining a stable, quiet posture. When small, unobtrusive mechanical perturbations, mimicking those occurring naturally, were applied to the foot using a piezoelectric translator, the posture remained stable but the perturbations produced no stretch reflex response in calf muscles. Also, intrinsic (non-neural) ankle stiffness of the foot, attributable to the Achilles’ tendon and several aponeuroses, was shown to be insufficient to stabilize the body. Hence, the stabilization must have been maintained by additional neural modulation of ankle torque (Loram and Lakie, 2002). This modulation may have been provided by predictive control using a cerebellar internal model (Loram et al., 2005).

11-3 Nociceptive Withdrawal Reflex Noxious stimuli applied to the skin of a cat forelimb causes flexion of that forelimb. These stimuli evoke mossy fiber and climbing fiber signals, which reach the C1 and C3 zones of lobule V. Purkinje cells in these areas project to the anterior interpositus nucleus, which, in turn, projects to the magnocellular red nucleus and primary motor cortex. Finally, descending signals along the rubrospinal and corticospinal tracts evoke withdrawal of the forelimb (Figure 33 illustrates only the red nucleus pathway). The lobule V/C3 zone contains 30–40 longitudinal microzones lying side by side, each 50–150 micrometers wide. Purkinje cells in each microzone receive simple-spike inputs from a small receptive field on the skin and adjust the withdrawal movements to be appropriate for avoiding the stimuli to this part of the skin. If not appropriate, the nociceptive stimuli may not be avoided and may stimulate climbing fiber receptive fields. Responses of Purkinje cells to such inappropriate stimuli could then be suppressed by the induction of conjunctive LTD. Climbing fibers in adjacent microzones are activated from adjacent skin areas, forming a detailed somatotopic map of the ipsilateral forelimb’s skin, particularly

128

the cerebellum

Figure 33 Neural wiring diagram for the limb withdrawal reflex.

The cerebellum provides adaptive control to optimize limb withdrawal in response to nociceptive stimulation of the skin and/or other leg parts. In this case, cerebellar nuclear neurons are in the position of the controller, while the magnocellular portion of red nucleus (RNm) is nested in the controlled object. In spite of this complexity, the general framework of the adaptive cerebellar control applied to the VOR (Figure 28) and stretch reflex (Figure 32) is applicable to the limb’s withdrawal. Abbreviations: a, CF receptive field in the forearm; AIP, anterior interpositus nucleus; b, forelimb flexor muscle; c, MF receptive field; C 1, C 3, names of the longitudinal cerebellar zones involved in withdrawal reflexes; CF, climbing fiber; FM, flexor motoneuron; MF, mossy fiber; PC, Purkinje cell; RST, rubrospinal tract. RST is interrupted to indicate that its connection to FM is mediated by segmental interneurons. (Based on Apps and Garwicz, 2005.)

its distal parts. Adjacent microzones innervate, in turn, adjacent cell groups in the anterior interpositus nucleus. Through further projections to the red nucleus, these microzones control movement components that have specific relationships with the location of climbing fiber receptive fields (Garwicz et al., 1998; Ekerot and Jörntell, 2003) (Chapter 9, “Network Models”).

11-4 Locomotion In the classic half center model of a CPG, as proposed by Anders Lundberg and his colleagues (Jankowska et al., 1967a, b; Stuart and Hultborn, 2008), it was assumed

11 • Somatic and Autonomic Reflexes

129

Figure 34 The organization of reflex pathways from spindle Ia afferents in extensor muscles to motoneurons supplying the same muscles during walking in the cat.

In this schematic, note the large and medium-sized hollow grey circles, and the small hollow circles that represent excitatory neurons and excitatory synapses, respectively. Likewise, the middle-sized filled circles and the small black rectangles represent inhibitory neurons and inhibitory synapses, respectively. Pathways (1) and (2) are the well-known monosynaptic excitatory and disynaptic inhibitory pathways from group 1a and 1b afferents, respectively, that were first revealed in barbiturate anesthetized in vivo cat preparations. Excitatory pathways (3) and (4) are open during locomotion and probably most other natural movements. Transmission in the disynaptic excitatory pathway (3) occurs during extension where it presumably functions, for example, to reinforce ongoing extensor activity during the stance phase of the step. One function of pathway (4) is to regulate the duration of the stance phase of the step by exciting the extensor half center of the locomotor rhythm generator. Activity in group 1b afferents during the stance phase of the step prevents the onset of the swing (flexion) phase of the step until the extensor muscles are unloaded. The shaded interneurons in pathways (3) and (4) have not yet been identified, and the connections indicated by the dashed lines are the most parsimonious for describing the functional effect of extensor group I afferents. The rhythm generator is assumed to include mutually inhibiting extensor and flexor half centers. (Based on Pearson, 1995; Hultborn, 2001; Stuart and Hultborn, 2008; Nichols and Ross, 2009.)

130

the cerebellum

that each limb is controlled by a separate half center and that each half center contains two groups of excitatory interneurons that directly project to flexor and extensor motoneurons, respectively. Because of the mutual inhibitory interconnections between the two groups, only one group can be active at a time (Figure 34). The CPG circuit for lamprey locomotion has recently been analyzed in depth and reconstructed by simulation (Grillner et al., 1991, 2007; Grillner and Jessel, 2009). A substantial but less complete model has also been achieved for cat locomotion (McCrea and Ryback, 2008). The involvement of the cerebellum in locomotion had been suggested by classic lesion studies in quadrupeds (Chapter 2, “Traditional Views of the Cerebellum”), which impaired coordination of the four limbs and other parts of the body as well. The modern approach to studying the mechanisms of locomotion started when steady quadrupedal locomotion was successfully evoked in high decerebrate cats on a treadmill belt during repetitive electrical stimulation of a specific site within the midbrain, the mesencephalic locomotor center (Shik et al., 1966). Stimulation of a restricted region along the midline cerebellar white matter in a decerebrate cat was later shown to evoke a generalized augmentation of postural muscle tone on a stationary treadmill belt and locomotion on a moving one. This effect appeared to be mediated by crossed fastigioreticular fibers that excited presumably the same reticulospinal pathway activated by the mesencephalic locomotor center (Mori et al., 1999). During mesencephalic-locomotor-center-evoked locomotion in the high decerebrate cat, there is phasic activity in rubro- (Orlovsky, 1972b), reticulo- (Perreault et al., 1993), and vestibulospinal (Orlovsky, 1972a) pathways. This rhythmic activity is dependent on an intact cerebellum and appears to be mediated by spinocerebellar pathways conveying information on the activity of the locomotor-generating spinal networks (Arshavsky et al., 1983, Arshavsky et al., 1986). Similar results on movement-related phasic activity in rubro- and vestibulospinal neurons have also been described for fictive scratching (Arshavsky et al., 1978a,b, 1988). Spike discharges have also been found to occur in DSCT and VSCT neurons in synchrony with motor rhythms (cat; Arshavsky et al., 1972a,b) (Chapter 6, “Preand Post-Cerebellar Cortex Neurons”). Magnocellular red nucleus neurons, the origin of the rubrospinal tract, also exhibited spike discharges related to each cycle of locomotion. The average firing frequency was minimal at the transition between the rhythmical alternating discharge of extensor and flexor efferents, and it increased progressively to reach a maximum in the second phase of the flexor burst (Arshavsky et al., 1988). Apparently, the rubrospinal tract mediates the function of the C1/C3 zone via the anterior interpositus nucleus when it adaptively controls the swing phase of limbs during locomotion. Lateral vestibulospinal tract neurons were

11 • Somatic and Autonomic Reflexes

131

also found to discharge during controlled locomotion. Most of the neurons projecting to the cervical segments were shown to have two frequency modulation peaks in their discharge: one in the late swing or early stance phase of a step by the ipsilateral forelimb and the other in the late stance or early swing phase (Udo et al., 1982). In decerebrate cats, a stretch reflex of the soleus muscle and its H-reflex were reported to be modulated during quadrupedal treadmill locomotion, reaching its peak at or before the peak in soleus EMG activity associated with the stance phase of the step (Akazawa et al., 1982). Marked modulation of the H reflex in the soleus muscle has also been demonstrated in humans during walking and running (Edamura et al., 1991). Evidence indicated that the locomotion-related modulation represented postsynaptic changes in α-motoneuron activity or in presynaptic inputs to these motoneurons, but not to changes in muscle spindle discharge (Akazawa et al., 1982). The modulation of stretch reflexes observed during controlled locomotion might be induced, at least partly, via the lateral vestibulospinal tract whose activity is, itself, known to be modulated by locomotion (Udo et al., 1982). In B-zone Purkinje cells, Udo et al. (1981) showed that climbing fiber responses occurred during controlled locomotion. During stable locomotion at a belt velocity of 36 centimeters per second, climbing fiber responses were modulated slightly: a small increment in their discharge rates occurred during the swing phase of the ipsilateral forelimb (Yanagihara and Udo, 1994). In an experiment on decerebrate ferrets, the perturbation of locomotion by a bar extended into the trajectory of the right forelimb in a specific phase of the step cycle also caused climbing fiber discharges (Lou and Bloedel, 1992). In mice, the extent to which their locomotion on a normal treadmill kept pace with increasing belt speed was taken as a measure of adaptation. This distinguished control mice from mutants deficient in mGluR1, which is required for conjunctive LTD induction (Ichise et al., 2000). In another experiment, a cat walking on a horizontally placed ladder was perturbed when one of the stepped-on rungs dropped 2 centimeters. Such drops usually resulted in some Purkinje cells generating climbing fiber responses at a short latency. Importantly, these responses were well time-locked to the onset of the unexpected rung drops but not to their cessation, which they often preceded (Andersson and Armstrong, 1987). A unique method of testing for cerebellar-evoked adaptability during locomotion was introduced by the use of split-belt treadmill locomotion in which the speed of each belt (i.e., for the right- versus left-side legs) could be controlled independently. While decerebrate cats were performing controlled locomotion on such a device, both fore- and hindlimbs on one side were suddenly subjected to a higher belt velocity. The cat gradually adapted using an unusual (atypical) pattern of limb coordination to retain stable locomotion. An inhibitor of NO synthase or a scavenger of NO applied to the cerebellar locomotion area abolished this adaptation. Because NO is required for the

132

the cerebellum

induction of conjunctive LTD, this finding supported the hypothesis that learning in locomotion was based on the induction of conjunctive LTD (Yanagihara and Kondo, 1996). The adaptive behavior described here has been well reproduced by simulation using an autonomous distributed control model of locomotion (Ito et al., 1999). In human patients with cerebellar lesions, two types of gait abnormality were categorized: one featured by impaired leg placement and the other by deficiencies in dynamic balance control. The former abnormality included inadequate adaptation to added weight loads. The cerebellar lesions in patients with the former deficiencies usually involved the interpositus and adjacent dentate nuclei. In contrast, patients with the latter abnormality usually had lesions in the fastigial nuclei (and to a lesser degree the interpositus nuclei) (Ilg et al., 2008). Split-belt treadmill locomotion has also been applied to humans to study their interlimb coordination (Diezt et al., 1994). In Morton and Bastian’s paradigm, belt speed was switched in sequential steps from both-slow (0.5 meters/second), to both-fast (1 meter/second), to both-slow, to the split-belt condition (one fast, the other slow), and finally to both-slow. Both normal control subjects and cerebellar patients were able to quickly adjust their stride length (distance traveled by one of the ankle markers from the time of initial contact to the time of lift-off of one limb) and percentage length of the stance phase (stride length in percentage of the total time from initial contact to the next initial contact). In contrast, the step length (difference between the distances of the leading and trailing limb ankle markers at the time of initial contact on the leading limb) and the double limb support time (time from initial contact on one limb to lift-off of the other limb, expressed as a percentage of the total stride time of the lift-off limb) changed slowly in the control subjects, and these adaptations to the split-belt condition could not be achieved in the cerebellar patients. Hence, it appeared that the former two parameters were determined by reactive feedback-driven adaptation and the latter two parameters by predictive feedforward adaptation in the cerebellum (Morton and Bastian, 2006).

11-5 Eye-Blink Conditioning The classical conditioning of the eye-blink reflex, particularly “delay conditioning” in which conditioned and unconditioned stimuli are contiguous, has been introduced as a simple and robust model of cerebellar learning (Lincoln et al., 1982; Yeo et al., 1985a,b,c). Conditioned eye-blink responses develop during repeated trials. Once established, they remain (consolidation), unless they are extinguished by unpaired conditioned and unconditioned stimuli. The conditioned reflex pathway (Figure 35) is mediated by a group of neurons in the anterior interpositus nucleus. These neurons in the rabbit receive conditioned

11 • Somatic and Autonomic Reflexes

133

stimuli such as a tone pip and, in turn, send signals via the red nucleus to trigeminal premotor and motor neurons. These latter cells induce eyelid closure, eyeball retraction, and a resulting passive extension of the nictitating membrane. Anterior interpositus neurons are normally inhibited by Purkinje cells in lobule HVI, which are activated by mossy fiber-mediated tone-pip signals. When, however, unconditioned stimuli, such as an air puff, were applied to the cornea to evoke climbing fiber signals, conjunctive LTD released the anterior interpositus neurons from Purkinje cell inhibition. This was shown to occur at the precise moment that the air puff was applied (Jirenhed et al., 2007; see also McCormick and Thompson, 1984; Berthier and Moore, 1986). Thus, an eye blink was evoked in a timely manner to protect the cornea from the air puff. Later studies revealed that impaired eye-blink conditioning occurred in mutant mice that lacked conjunctive LTD (Shibuki et al., 1996; Kishimoto et al., 2001, 2002; Miyata et al., 2001).

Figure 35 Wiring diagram for eye-blink conditioning.

Tone pip-evoked auditory signals as the conditioned stimulus (CS) enters the cerebellum via projections from the pontine nuclei (PN). Air puff as the unconditioned stimulus (US) enters the cerebellum via projections from the inferior olive (IO). The CS and US pathways converge onto the nucleus interpositus (AIP). The postsynaptic conditioned response CR pathway projects from the AIP to the magnocellular red nucleus (RNm) and ultimately connects to neurons (TPM) that cause eye blinking. TPM includes trigeminal premotor neurons, motoneurons for the eyelid muscles and, in rabbit, the nictitating membrane. Additional abbreviations: CS, broken line square enclosing control system structure; MC, that for microcomplex; NV(sp), spinal trigeminal nucleus; VCN, ventral cochlear nucleus. Symbols: broken line, polysynaptic connections. Other conventions are similar to Figure 33. (Based on Thompson, 1988.)

“Blocking” studies of eye-blink conditioning provide the opportunity to define the roles of the nucleo-olivary inhibitory projection (N-O in Color Plate V).

134

the cerebellum

Recordings of cerebellar neuronal activity have shown that the inputs of the IO to the cerebellum become suppressed as learning occurs. This corresponds to the behavioral phenomenon of “blocking.” It occurs when a conditioned stimulus (CSA) is first paired extensively with an unconditioned stimulus (US), and then a second conditioned stimulus (CSB) is combined with CSA and the same US. Virtually no conditioning then occurs for the CSB. This blocking effect suggests that if an US is already fully predicted by one stimulus, and if the addition of a new stimulus provides no new information about the US, then the US will not activate or support the learning process responsible for establishing a new CS-US association. Because disrupting the inhibition of the IO by local infusion of a GABA antagonist, picrotoxin, prevented blocking in rabbits, the nucleo-olivary negative feedback process could be the neural mechanism mediating blocking (Kim et al., 1998). The disruption of nucleo-olivary inhibition also prevented the extinction of eye-blink conditioning, whereas that of excitatory inputs induced extinction (Medina et al., 2002). These findings are interpreted as indicating that the nucleo-olivary inhibition is required for setting the climbing fiber discharge rate at a level appropriate for driving extinction. However, caution is needed in interpreting these observations because the manipulation of olivary neuronal activity influences the background activity of Purkinje cells. This might affect learning and its behavioral manifestation (Bengtsson et al., 2004). The effects of the removal of the cerebellar cortex and the local application of pharmacological agents on eye-blink conditioning have been controversial in studies carried out to determine whether memory is stored in the cerebellar cortex or alternatively in the anterior interpositus nucleus (for review, see Christian and Thompson, 2003). The importance of the cerebellar cortex was emphasized because the acquisition training for eye-blink conditioning was blocked when the anterior lobe of the cerebellar cortex was lesioned (Garcia et al., 1999) or when the lobule HVI was inactivated by localized infusion with an AMPA/kainate receptor blocker, CNQX (Attwell et al., 2001). A specific contribution of the cerebellar cortex to the timing of eye-blink conditioning was suggested because a reversible blockade of cerebellar cortical outputs via the infusion of picrotoxin into the interpositus nucleus disrupted the timing to accurately coincide conditioned responses with the unconditional stimulus. In the same animals, conditioned responses were abolished by the infusion of a GABA agonist, muscimol (Garcia and Mauk, 1998; Bao et al., 2002). In mice lacking conjunctive LTD owing to the Purkinje cell-specific inactivation of PKC, conditioned responses substantially remained, but accurate timing was lost (Koekkoek et al., 2003). The abnormally timed, short-latency conditioned responses remaining after the blockade of Purkinje cell outputs could be generated by an enhanced excitation

11 • Somatic and Autonomic Reflexes

135

of anterior interpositus neurons. This postulate was adopted in simulating eye-blink conditioning (Medina et al., 2000). Four lines of experimental evidence supported this idea. (1) Anterior interpositus nuclear neurons displayed an associative form of learning plasticity (Ohyama et al., 2006). (2) Synapse formation occurred in the interpositus nucleus in association with eye-blink conditioning (Kleim et al., 2002). (3) The injection of a protein synthesis inhibitor, anisomycin, in the bilateral cerebellar hemispheres reversibly blocked the acquisition, even though it did not block the performance of a conditioned response (Bracha et al., 1998). (4) Eye-blink-conditioned animals exhibited a new category of cell-divisioncycle-2 (cdc2)-related kinases in interpositus neuron mRNAs (Gomi et al., 1999). A more recent study showed that consolidation was prevented by the application of the GABA agonist, muscimol, shortly after training had inactivated the cerebellar cortex. Post-training inactivation of the interpositus nucleus, in contrast, allowed conditioning to develop normally (Attwell et al., 2002). These observations indicate that a cortical circuit is essential for memory consolidation. A complication arises, however, from the possibility that memory consolidation is cortical in the cerebellum but sensitive to the increased excitability of nuclear neurons following the loss of cortical inhibition. When cerebellar lobule HVI and the anterior interpositus nucleus of rabbits were inactivated simultaneously during the post-training period, consolidation was still impaired. This suggests that the disinhibitory effects of cortical inactivation on the interpositus nucleus are irrelevant for consolidation (Kellett et al., 2010). A crucial role of conjunctive LTD in eye-blink conditioning has been suggested in two ways. Activation of PKC, which is required for conjunctive LTD induction (Chapter 7, “Conjunctive Long-Term Depression (LTD)”), occurred within the molecular layer of lobule HVI in rabbits given this conditioning (Freeman et al., 1998). On the other hand, mice lacking conjunctive LTD owing to the Purkinje cell-specific inactivation of PKC retained their conditioned responses, but its accurate timing was lost (Koekkoek et al., 2003). Note, however, that the relationship of eye-blink conditioning to conjunctive LTD is not so simple because a mutant mouse lacking the fragile X mental retardation type 1 protein (FMR1) exhibited impaired eye-blink conditioning in its timing when they exhibited unusually enhanced LTD (Koekkoek et al., 2005). When the HVI areas of the rabbit cerebellum were sliced 24 hours after the rabbit had been trained for eye-blink conditioning, sequentially applied parallel fiber and climbing fiber stimuli failed to induce conjunctive LTD, whereas it occurred in cerebellar slices dissected from control rabbits (Schreurs et al., 1997). This suggests that conjunctive LTD underlying eyeblink conditioning continues for at least 24 hours and occludes the elicitation of another conjunctive LTD. An increased excitability of Purkinje cell dendrites in

136

the cerebellum

the rabbit lobule HVI was detected and sustained for one month following the initial acquisition of eye-blink conditioning (Schreurs et al., 1998). This learningrelated excitability increase is presumably caused by changes in K+ current, possibly mediating an IA-like current, but its relationship to conjunctive LTD is unclear. Finally, a significant decrease in the number of excitatory synapses in the molecular layer of the cerebellar cortex has been reported in eye-blink-conditioned rabbits (Connor et al., 2009). Whereas evidence continues to accumulate for cerebellar plasticity playing a key role in eye-blink conditioning, a recent study by Nakanishi’s group suggested a more complex possibility. This study used reversible neurotransmission blocking (RNB) mice (Wada et al., 2007). In animals treated this way, only granule cells expressed tetanus toxin in a doxycycline (DOX)-dependent manner. Simple spikes of Purkinje cells elicited by granule cell inputs were turned off and on by DOX treatment and its withdrawal, respectively. It was found that blockade of granule cell inputs to Purkinje cells abolished eye-blink conditioned responses in a DOXdependent manner. Thus, a cortical circuit via Purkinje cells appeared to be essential for expressing the conditioned responses. Peculiarly, however, from the beginning of the reconditioning process, when granule cell inputs recovered by removal of DOX, seemingly normal conditioned responses were immediately produced in the reconditioned mice. Bilateral destruction of the interpositus nuclei prior to eye-blink conditioning abolished the learning. These results can be explained on the basis of a bilateral membrane-potential dependent synaptic plasticity at mossy fiber-nuclear neuron synapses (depressed by depolarization and potentiated by hyperpolarization), as recently reported by McElvain et al. (2010). It might happen that mossy fiber synapses in the interpositus nucleus are depressed by the depolarization caused by DOX turning off of Purkinje cell signals. The effect might then be reversed to potentiation quickly when membrane hyperpolarization is caused by reactivated Purkinje cells. The potentiated conditioned responses might look like learned responses.

11-6 Sympathetic Reflexes Several areas of the cerebellum are now known to affect cardiovascular functions (Nisimaru, 2004), with two of them studied in detail. One is the lateral edge of lobules X (nodulus) and IX (uvula), which is linked to the vestibular nucleus and a medial part of the parabrachial nucleus. It functions in the vestibulosympathetic reflex. The other is folium-p in the flocculus (Figure 28), which connects with the lateral part of the parabrachial nucleus and functions in connection with the somatosympathetic reflex.

11 • Somatic and Autonomic Reflexes

137

The vestibulosympathetic reflex acts to maintain blood pressure stability during head and body movement. For example, head tilting with the nose upward in awake rabbits causes a slight decrease in mean arterial blood pressure, which recovers to the control level within 3–5 seconds. The lateral nodulus-uvula is considered to adaptively improve the dynamics of the vestibulosympathetic reflex. Indeed, after bilateral lesioning of the lateral nodulus-uvula, recovery is retarded in the just-cited head-tilt-evoked blood pressure changes (Nisimaru et al., 1998). The most lateral zone of the nodulus and uvula receives mossy fiber afferents from the ipsilateral vestibular nuclei and climbing fiber afferents from the contralateral vagal and aortic nerves via the medial accessory olive (MAO). This zone, in turn, sends Purkinje cell axons to the parabrachial nucleus and vestibular nuclei. These connections subserve adaptive control of the vestibulosympathetic reflex. The somatosympathetic reflex has a very different role. During movement, it helps redistribute arterial blood flow among working and non-working muscles and viscera. For example, during defense reactions of a rabbit to electric foot shocks, the rate of arterial blood flow to active muscles increases, whereas it decreases to inactive muscles and viscera. The entire neuronal circuit for this behavior has recently been analyzed in detail (Nisimaru et al., 2010). The reflex arises from the activity of high-threshold afferents supplying muscles, joints, and skin. This activity is conveyed to the parabrachial nucleus, the coordinating center of the reflex, whose output, in turn, activates the descending sympathetic system. Therefore, the folium-p appears to be embedded to the somatosympathetic reflex as its adaptive mechanism. Interestingly, orexin-containing beaded fibers arise from the hypothalamus and pass to the flocculus, most densely in the folium-p. Here, orexins appear to act as neuromodulator in the fight-or-flight situations of defense reactions invoving the somatosympathetic reflex and folium-p (Nisimaru et al., 2006, 2010). The involvement of the cerebellum in respiration has long been discussed. In a recent study on anesthetized rats, stimulation of the fastigial nucleus, but not the interpositus and lateral nuclei, significantly affected respiration, primarily by increasing its frequency. Lesions of cerebellar nuclei by microinjection of kainic acid did not significantly alter eupneic breathing, but fastigial nuclear lesions attenuated the respiratory responses to hypercapnia and sodium cyanide. This implies that fastigial nuclear neurons uniquely modulate respiration independent of cardiovascular effects and facilitate respiratory responses mediated by activation of CO2 and O2 receptors. Hence, it has been suggested that a microcomplex in the A-zone-fastigial system functions to make the respiratory chemoreflexes adaptive (Xu and Frazier, 2000).

138

the cerebellum

An ongoing study has revealed prominent respiratory rhythmic activity emanating from the cerebellar cortex in newborn rats, using block preparations of the cerebellum-pons-medulla-spinal cord. These blocks were excised from rats at postnatal week 1 and maintained in a perfusion chamber. Optical imaging with voltagesensitive dyes and whole cell patch clamp recording from Purkinje cells showed that a respiratory rhythm was present in many parts of the cerebellum, especially in the lateral part of paraflocculus and lateral end of the vermis. Similar inspiratory and postinspiratory activities were seen in the dorsal part of the IO and the parabrachial nucleus (Arata et al., 2009). The functional meaning of these respiratory activities, which do not occur in the adult, is unclear. It is nonetheless an interesting observation suggesting developmental changes deserving of full study.

11-7 Summary After reading Chapters 10 and 11, one might think that the cerebellum originally evolved to facilitate reflexes becoming readily adaptable to ever-changing environments. Indeed, studies of reflexes do provide simple, robust experimental paradigms for evaluating the mechanisms and roles of cerebellar circuits. These paradigms will continue to be useful, particularly when testing cerebellar functions in behaving animals and patients with cerebellar lesions.

12 Adaptive Control System Models

12-1 Introduction After reviewing the involvement of the cerebellum in various reflexes in Chapters 10 and 11, we can now address the question of how the intricate neuronal circuit integrated in the form of a microcomplex is incorporated into reflex control systems for a subtle adaptive mechanism. Here, we consider major adaptive control system models proposed to this point, and the experimental evidence on which these models are based.

12-2 Adaptive Control of Ocular Reflexes In various ocular reflexes, neuronal circuit components corresponding to the controller, controlled object, and adaptive mechanism have been identified on the basis of anatomical structures, lesion effects, and activities of Purkinje cells and other involved neurons. These components are illustrated in Figures 36 to 38. As reviewed in Chapter 10, “Ocular Reflexes,” visual feedback plays no role in the VOR. It may operate in the OKR, OFR, and saccades, but only quite inefficiently because of its relatively long delay (50–100 milliseconds) (Smith et al., 1969) (see Chapter 10). Hence, to maintain an optimal gain over a given range of vestibular, visual, and oculomotor conditions, these ocular reflexes need the intervention of a microcomplex as an adaptive controller. In both the VOR and OKR, the same group of vestibular neurons acts as the common controller (Figures 36 and 38), which is the target of flocculus Purkinje cells. The controller for the OFR has not been well explored, but it should consist of neurons targeted by ventral paraflocculus Purkinje cells (Figure 38). Because

139

140

the cerebellum

these Purkinje cells project to vestibular nuclei in a pattern similar to that of flocculus Purkinje cells (Balaban et al., 1981), it is conceivable that a group of vestibular nuclear neurons serve as the controller of the OFR (Chapter 10). This group of vestibular neurons could be identical or, even if not, at least akin to VOR/OKR relay neurons. The combination of the VOR, OKR, and OFR provides a good example of a multi-input control system, that is, sharing the controller and the controlled object (Figure 9A). Note that even though the OFR involves the parietal association cortex (MT/MST area), it is an elaborate visual perception mechanism and by no means designed for voluntary motor control.

Figure 36 Control system scheme for adaptation of the vestibuloocular reflex.

The block diagram for adaptive VOR control is based on the neuronal wiring diagram in Figure 28. The shade on the left is for the sensory system (Ss) processing information for the controller. The shade in the middle encloses the microcomplex (Ac), which uses VN and the flocculus to act as an adaptive controller. The shade on the right is for the controlled object (Co). The dotted arrow represents climbing fibers. Abbreviations: a, sensory mossy fiber pathways; AOS, accessory optic system; b, recurrent mossy fiber pathways (shown by broken lines because of the ambiguity of their presence); IO, inferior olive; OM, oculomotor neurons; VN, VOR relay neurons in the vestibular nuclei; Ve, eye velocity; Vh, head velocity. (Based on Ito, 1972, 1974, 1998, 2006.)

12 • Adaptive Control System Models

141

Figure 37 Control system scheme for adaptation of the optokinetic reflex.

The OKR shares the controller, controlled object, and flocculus with the VOR but responds to visual signals. This is an example of a multi-input reflex system, as explained in Figure 9A. Abbreviations: NRTP, nucleus reticularis tegmenti pontis; Vs, screen velocity. For other abbreviations and symbols, see Figure 36. (Based on Ito, 2006.)

Figure 38 Control system scheme for the ocular following response.

This block diagram shows that adaptive control of the OFR involves the cerebral cortex (VC/MT/MST) and the ventral paraflocculus (VPFL). Other abbreviations: DLPN, dorsolateral pontine nucleus; LGN, lateral geniculate nucleus; MST, medial superior temporal area; MT, medial temporal area; VC, visual cortex. For further abbreviations, see Figure 36. The possible connection from DLPN to VN is indicated by a broken line because it has not been confirmed. (Based on Kawato, 1999; Ito, 2006.)

142

the cerebellum

12-3 Two Models of the Flocculus-VOR System: Forward Versus Recurrent Pathways In the flocculus hypothesis first proposed in the early 1970s, we assumed that the primary drive to flocculus Purkinje cells came from sensory signals originating in the vestibular organ (Ito, 1972, 1982). In this hypothesis, the flocculus is a modifiable side path to the major VOR pathway, as schematically shown in Figure 39A. Miles and Lisberger (1981) challenged this hypothesis, as schematically shown in Figure 39B, and the debate that followed centered on two issues (Ito, 1993a; Lisberger and Sejnowski, 1992, 1993). The first concerned the source and functional role of eye velocity signals in flocculus Purkinje cells. The second issue (discussed in the next section) concerned the memory sites for VOR adaptation. The first issue required asking whether eye velocity signals were mediated by recurrent collaterals of VOR relay neurons in a form of positive feedback (Miles and Lisberger, 1981; Lisberger and Sejnowski, 1992, 1993), or whether vestibular mossy fiber inputs to the flocculus served primarily as a forward side path to the major VOR pathway (Ito, 1972, 1982, 1993a). Meanwhile, observations about the ventral paraflocculus revealed that the major simple-spike signals driving Purkinje cells were of sensory (visual) origin (Takemura et al., 2001) and that this structure provided a parallel pathway to the forward vestibular pathway proposed in the flocculus hypothesis (Tabata et al., 2002) (see Figure 39C). Hence, there is a dichotomy about the possible role of the flocculus. It may provide either a recurrent positive feedback mechanism (Figure 39B) or be a mechanism for converting forward (side path/parallel-pathway) signals to Purkinje cell simple-spike discharges (Figure 39A, C) and thereby represent an inverse model of the eyeballs (see Section 12-5). Three to four decades have passed since then. We still have no experimental basis, however, for the model-oriented idea that the flocculus receives mossy fiber recurrent collaterals of VOR relay neurons that provide positive feedback of eye velocity signals. Recurrent collaterals are common in cerebellar nuclear neurons, but this cannot be generalized to vestibular nuclei because Deiters neurons in the lateral vestibular nucleus seem to lack recurrent collaterals that project to the cerebellar cortex. In early studies, Deiters neurons were never seen to respond antidromically to stimuli from the cerebellar cortex (Ito and Yoshida 1966; Ito et al., 1969). This was in contrast to the antidromic responses of cerebellar nuclear neurons to stimuli from their overlying cortex (Ito et al., 1964). A side issue is that recurrent collateral input from VOR relay neurons is often assumed to be in accordance with von Holst’s (1954) classic concept of “efference copy.” It proposed the existence of an internal centripetally projecting copy of the motor command to

12 • Adaptive Control System Models

143

cancel the sensory effect of the induced movement (re-afference) (for review, see Bridgeman, 1995; Houk et al., 1996). However, no true efference copy appears to accompany the VOR because we see the world moving during the slow phases of vestibular nystagmus. If an efference copy is operative, the world should continue to appear stable during such nystagmus (Bridgeman, 1995). This emphasizes the need for caution when suggesting that efference copy is one of the mechanisms of the VOR.

Figure 39 Circuit models of adaptive control of ocular reflexes.

(A) Modifiable side path model for the VOR, assuming a forward path. Based on the ideas of Ito (1972, 1974, 1982). (B) Recurrent positive feedback model for the VOR proposed by Miles and Lisberger (1981) and Lisberger and Sejnowski (1992). (C) Parallel pathway model for the OFR developed by Tabata et al. (2002). Major pathways in each model are shown by thick lines. Abbreviations: CF, climbing fiber; FL, flocculus; T and F, units inserted to the circuit; MST, medial superior temporal area; OM, motoneurons for extraocular muscles; VA, vestibular afferent system; Ve, eye velocity; Vh, head velocity; Vs, screen velocity; VN, vestibular nuclear neuron; VPFL, ventral paraflocculus. Symbols: arrows indicate fixed synaptic actions or point to output responses; closed circle head, modifiable synapse; square head, inhibitory action of Purkinje cells; broken arrows, climbing fiber pathway; -, inhibition.

144

the cerebellum

12-4 Two Memory Sites for VOR/OKR Adaptation Miles and Lisberger (1981) suggested that a memory site is located in vestibular nuclei (Figure 39B), as recently confirmed, in addition to parallel fiber-Purkinje cell synapses in the flocculus. A calculation based on their model with two memory sites, however, indicated that parallel fiber-Purkinje cell synapses should undergo an adaptive change in a direction opposite to that expected from the flocculus hypothesis. This led to doubts about the functional role of the flocculus in VOR adaptation. To address this issue, Highstein’s group performed a system identification analysis of Purkinje cells in the monkey flocculus. Their initial measurements, using acute adaptation to a visual-vestibular mismatch, detected a change in Purkinje cell behavior that contradicted the flocculus hypothesis (Hirata and Highstein, 2001). However, a later measurement, using chronic adaptation by the wearing of minifying or magnifying lenses, revealed that after their adaptation, Purkinje cells changed their sensitivity to eye position, eye velocity, and head velocity. These combined changes at the Purkinje cell level contributed to a net modulation that was appropriate for supporting the learned VOR gains (Blazquez et al., 2003). These analyses, however, were based on a theoretical circuit that assumed the presence of an efference copy pathway. This efference pathway is still a modelistic postulate, as emphasized above. As a result, another open question concerns what occurs if eye velocity signals are derived otherwise: for example, from (1) midline paramedian tract neurons, (2) the conversion of the retinal error signals of climbing fibers, or (3) the input of supplemental signals from other relevant source(s) (Chapter 10).

12-5 Inverse Model of the Eyeballs A remarkable finding in the monkey ventral paraflocculus is that during the OFR, simple-spike discharges of Purkinje cells represent an inverse model of an eyeball (Shidara et al., 1993; Yamamoto et al., 1997; Gomi et al., 1998). This model is expressed by a linear summation of eye acceleration, eye velocity, and eye position. The idea and the finding are considered below in relation to the internal model control of voluntary movements (Chapter 15, “Internal Models for Voluntary Motor Control”). Here, it is considered from the perspective of the feedforward control of reflexes. In an adaptive feedforward control such as the OFR (see earlier), the role of the cerebellum is to tune the overall gain of the control system (equal to -1) at which value the imposed visual target movement is compensated completely by the generated eye movement. This condition is attained when the

12 • Adaptive Control System Models

145

input/output relationship of the adaptive controller (f) is equated to the output/input relationship of the controlled object (G): that is, f= -1/G; see Figure 7C). In the OFR, the ventral paraflocculus should tune f to be equal to 1/G, the reciprocal of the eyeball’s dynamics (G). A problem about the preceding argument. The correctness of f = 1/G can be determined if one records from the final output neurons of the adaptive controller, which should be the controller neurons of the OFR. However, in the aforementioned Shidara et al.’s (1993) study, the recordings were made in ventral paraflocculus Purkinje cells, and not from the controller neurons, which are located most likely in the vestibular nuclei (Chapter 10). Generally speaking, controller neurons are driven by excitatory signals from mossy fibers and inhibitory signals from Purkinje cells, and hence the activity of Purkinje cells should deviate from that of the real controller neurons. Nevertheless, in the Shidara et al. (1993) study, pontocerebellar mossy fibers mediating the OFR did not seem to provide any potent excitatory input to vestibular nuclear neurons (see Figure 38). Hence, such deviation might have been minimal. It is therefore conceivable that the spike discharges recorded from Purkinje cells by Shidara et al. (1993) represented an approximate inverse model of the eyeballs. Similar measurements have also been performed in flocculus Purkinje cells during the vertical OKR (Mizukoshi et al., 2000) and horizontal VOR and OKR (Omata et al., 2000). The results generally conformed to the inverse model representation of the eyeballs. However, in view of the prominent mossy fiber inputs to the controller neurons of the VOR/OKR (see Figures 36 and 37), it is difficult to apply the previous approximation based on the OFR. Recently, recording was made in the Y group of vestibular nuclear neurons (Y neuron) during squirrel monkey’s chronic VOR adaptation (Blazquez et al. 2006). In this experiment, Y neurons did not change eye-velocity sensitivity after training, whereas head-velocity sensitivity increased significantly after low-gain but not high-gain training. In contrast, Purkinje cell eye velocity sensitivity changed significantly after low-gain training, and head-velocity sensitivity changed significantly after high-gain training, Apparently, there is no simple reciprocity between Y neurons and Purkinje cells even though they are connected by inhibition. Even though the results still do not allow a simplistic interpretation, it will be valuable to continue these efforts to examine the hypothesis that a microcomplex represents an inverse model of the eyeballs.

146

the cerebellum

12-6 The Saccade Control System When compared with the previously considered ocular reflexes, the control system structure for saccadic eye movement suggested by experimental data (Chapter 10, Section 8) is somewhat more complicated. However, if we take the fastigial nucleus as a controller, we need to include the excitatory and inhibitory burst neurons together with the oculomotor neurons, extraocular muscles, and eyeballs as the controlled object (Figure 40). This nesting is analogous to the previously presented models for limb withdrawal (Figure 33) and eye-blink conditioning (Figure 41). The fastigial neurons form a microcomplex with A-zone Purkinje cells, which would function as an adaptive controller, that is similar to the models shown in Figures 28 and 33. The superior colliculus may serve for visual information processing just as the MT/MST area functions in OFR control (Figure 38).

Figure 40 Control system scheme for a horizontal saccade.

This block diagram is drawn based on the data described in section 10-8 for saccadic eye movements. Abbreviations: SC, superior colliculus; Fas, fastigial oculomotor region. SG, saccade generator. Note that the illustrated microcomplex (Ac) is of the Ntype characterized by the involvement of cerebellar nuclear neurons in the controller and also by the involvement of three units (SG, OM, and Eye) in the controlled object (Co). These features contrast to those in the V-type, which involves vestibular nuclear neurons as the controller and two units in the controlled object (compare with Figures 36–38).

To model the saccadic system, it is assumed that errors detected via IO climbing fiber signals interact with simple spikes reflecting saccade-generating commands and thereby induce conjunctive LTD (Schweighofer et al., 1996a,b). The problem of a so-called temporal credit assignment occurs because climbing fiber signals representing consequence errors can reach Purkinje cells only after the arrival of parallel fiber activity related to the saccade-generating command. This problem can be avoided by assuming that parallel fiber activity results in long-lasting chemical signal transduction effects, in particular, the generation of IP3

12 • Adaptive Control System Models

147

(Chapter 7, “Conjunctive Long-Term Depression (LTD)”). In a recent model of saccadic adaptation, Fujita (2005) assumed that a consequence error detected at the end of the first saccade was converted to a motor command in order to initiate a corrective saccade and that the corrective command triggered climbing fiber activity. In spite of the fact that such climbing fiber activity arises after the first saccade, it may still induce conjunctive LTD in the parallel fiber-Purkinje cell synapses involved in the first saccade. The reason is that parallel fiber-evoked signal transduction in Purkinje cells lasts long enough to interact with the climbing fiber signals that arrive somewhat later. The conjunctive LTD so induced modifies the first saccade adaptively to “catch up with” the second saccade. When a large, realistic cerebellar neuronal network was incorporated into a saccade generator model, the complex spatiotemporal behavior of the neuronal subpopulations implicated in adaptive saccadic control was modeled in a manner that was consistent with the experimental data (Schweighofer et al., 1996b).

12-7 Adaptive Control of Somatic Reflexes Similar control system models can be applied to somatic reflexes. Deiters neurons, via the lateral vestibulospinal tract, function as the controller of the controlled object provided by the segmental circuit for extensor muscles (Figure 32). In this system, the peripheral proprioceptive and cutaneous sensory signals are sent to Deiters neurons via spinocerebellar tracts. Deiters neurons, in turn, send descending signals to excite extensor motoneurons and inhibit their inhibitory interneurons. As a result, a limb can remain extended during a standing posture or during the supporting (stance) phase of stepping. Attached to this long loop reflex, the Bzone receives mossy fiber afferents from the spinocerebellar tracts and, in turn, projects Purkinje cell axons directly to Deiters neurons. Climbing fiber responses occur in B-zone Purkinje cells when an error is sensed (Chapter 10). For the withdrawal reflex of a forelimb, the anterior interpositus nucleus acts as a controller, whereas the magnocellular red nucleus neurons, forelimb flexor motoneurons, and associated segmental neurons are nested as a controlled object (Figure 33). A microcomplex involving the C1/C3-zone and the anterior interpositus nucleus constitutes an adaptive controller. The unconditioned eye-blink reflex is evoked by an air puff, which drives trigeminal premotor neurons and motoneurons and induces eyelid closure. The magnocellular red nucleus forms a long loop pathway superposed on the eye-blink reflex pathway. Based on the neuronal circuit so far dissected (Figure 35), the block diagram in Figure 41 is formulated. There, the anterior interpositus nucleus,

148

the cerebellum

together with the HVI area of the C1- and C3-zones, form an adaptive controller of the eye-blink reflex, whereas the magnocellular red nucleus is nested to the controlled object together with trigeminal premotor neurons and motoneurons. Normally, tone pip stimuli do not drive the eye-blink reflex, but after classic eye-blink conditioning, they evoke an eye blink in a timely and predictive manner in order to protect the cornea from the air puff. This adaptive mechanism is now explained by a combination of conjunctive LTD that is induced by tone-pip-induced simple spikes, and air-puff-induced climbing fiber spikes in Purkinje cells. Another form of learning is also involved. It occurs in those synapses supplied by the tone-pipemediating mossy fibers to the anterior interpositus nuclear neurons (see Chapter 9, Section 7 and Chapter 11, Section 5).

Figure 41 Control system scheme for eye-blink conditioning.

This block diagram is based on the wiring diagram illustrated in Figure 35. The cerebellum provides an adaptive controller for the controlled object, which in this case is the eye-blinking system including the magnocellular red nucleus (RNm). Whereas tone stimuli provide mossy fiber inputs, air puff stimuli provide climbing fiber inputs. Abbreviations: Ac, adaptive controller; AIP, anterior interpositus nucleus; AUD, auditory system; C1/C3, zones of lobules VI-VII responsible for eye-blink conditioning; Co, controlled object; IO, inferior olive; N-O, a nucleo-olivary inhibitory projection, which is presumed to be responsible for the “blocking” phenomenon (see text); TPM, trigeminal premotor neuron and motoneuron; VS, visual system. (Based on Ito, 2006.)

12-8 Two Prototypes of Adaptive Control of Reflexes When the wiring and block diagrams for various reflexes are compared, two prototypes of microcomplexes stand out. For the VOR, OKR, OFR (Figures 28, 36–38) and myotatic reflex (Figure 32), the controller is served by certain vestibular

12 • Adaptive Control System Models

149

nuclear neurons, and hence the microcomplex has been called “V-type.” In contrast, for the limb withdrawal reflex (Figure 33), saccadic eye movement (Figure 40), and eye-blink conditioning (Figure 41), the controller is served by cerebellar nuclear neurons (fastigial or anterior interpositus), and the microcomplex involved has been called “C-type” (Ito, 1984). The V-type prevails in the flocculonodular lobe and B-zone (Figures 28, 32), whereas the C-type is found in the A- and C1/C3zones (Figures 33, 40, 41). These two types of microcomplexes may function similarly to each other as an adaptive controller, but they act on the controlled objects of different structures; V-types act on one-step-nested controlled objects, whereas C-types act on two-step-nested controlled objects (see Figures 11 and 12). We may consider that the C-type is more advanced than the V-type in an evolutionary sense in view of the higher integration level of the C-types’ controlled objects.

12-9 Summary The cerebellum can now be viewed as an assembly of numerous microcomplexes, many of which are associated with reflexes to afford them a substantial degree of adaptability. This feature somewhat resembles the recently developed “embedded computing system,” in which a tiny computer is lavishly used in each instrument, such as a portable phone, in contrast to a large computer commonly shared by many users. Much is yet to be learned about the specific neuronal processes that operate in each microcomplex.

13 Voluntary Motor Control

13-1 Introduction We now proceed to the examination of neuronal circuits for another category of movements: the voluntary control of arms, hands, and fingers, as commanded by the primary motor cortex. Importantly, this control involves the cerebrocerebellar communication loop that links the primary motor cortex and the intermediate part of the cerebellar hemisphere. This system enables humans to acquire skills for multijointed limb movements with many degrees of freedoms (27 for an arm with a hand and five fingers). Various test paradigms have been designed to study this form of cerebellar control. They include noting movement deficits caused by lesions of the cerebellum or TMS disruption of cerebellar function, and studying the potential functional meaning of simple- and complex-spike discharges in Purkinje cells during the elaboration of skillful voluntary movements. In most cases, control mechanisms have been studied in both monkeys and humans.

13-2 Load Compensation and Reaction Time Task Load compensation task. In Gilbert and Thach’s (1977) test, a monkey was trained to hold a lever stationary in the face of flexion and extension wrist forces developed by a torque motor. A change of load at some unpredictable time induced complex spikes in Purkinje cells located in the intermediate cortex of lobules III to V. These spikes occurred just after the load switch (at 50–150 milliseconds), and they apparently represented errors caused by the sudden change. Simple-spike discharge in these Purkinje cells changed inversely to their complex-spike discharge, and it remained decreased after the latter discharge had returned to normal. These observations provided early experimental evidence supporting Albus’ (1971) theoretical hypothesis (Chapters 3, 9). 150

13 • Voluntary Motor Control

151

Reaction time. “Simple reaction time” is the time taken from the presentation of a sensory stimulus (typically light or sound) to a subsequent behavioral response (typically pressing a button). It is considered to be an index of the speed of information processing in the relevant sensory-motor pathways. This time in college-age individuals is ~160 milliseconds for an auditory stimulus and ~190 milliseconds for a visual stimulus (Brebner and Welford, 1980). In a “serial reaction time” task, now widely used to measure the acquired degree of skill, a visual cue can appear at any one of four positions (1–4) arranged horizontally on a computer screen. When a cue appears as the go signal, the testee taps the appropriate button on a response pad, thereby ending the trial. The trials are repeated in 5 blocks of 100 trials/block. In blocks 1 and 5, the go signals are presented in random order. In blocks 2 through 4, the go signals represent a sequence of ten cues whose order (e.g., 2-3-1-4-3-2-4-1-3-4-2-1) is repeated ten times in each block of trials. The duration of each trial, defined by the participant’s response time, is the primary task measure, and its gradual shortening during the task is considered to be an index of learning (see, e.g., Gómez-Beldarrain et al., 1998; Robertson, 2007). In a group of normal subjects performing a serial reaction time task, repeated TMS (rTMS) over the lateral cerebellum significantly impaired learning. This interference with right cerebellar hemisphere activity induced a significant decrease in learning regardless of the hand used to perform the task, whereas left cerebellar hemisphere activity seemed to be more linked to learning using the ipsilateral hand (Torriero et al., 2004). In contrast, patients with single, unilateral vascular lesions in the territory of the posterior-inferior or superior cerebellar artery were examined as compared to age- and sex-matched control subjects in a onehanded version of a serial reaction time task. Learning was impaired for the hand ipsilateral to the lesion, but it was normal for the contralateral hand (Gómez-Beldarrain et al., 1998). The lateral cerebellar hemisphere thus appears to be involved in learning processes for a serial reaction time task for only the ipsilateral hand. However, another study on patients with unilateral focal cerebellar lesions reported that these patients were defective in learning the task with both hands (Molinari et al., 1997). Clearly, the interesting problem of functional lateralization in the cerebellar hemispheres is still an open issue.

13-3 Multijoint Arm Movements for Reaching The well-known finger-to-nose test of cerebellar dysmetria was reproduced when a monkey wearing wedge prisms was trained to try to reach a target by hand. Adaptation in the reaching depended critically on visual information about errors

152

the cerebellum

consequent to the movement at its end point. The effect of this visual information on consequence errors was largest at 0–10 milliseconds, and it decayed significantly in 50–100 milliseconds (Kitazawa and Yin, 2002). Thus, this paradigm represents typical consequence error-based learning. In another study by Kitazawa and his colleagues, a monkey was trained to perform short-lasting reaching (~200 milliseconds in duration) to touch a visual target, which appeared at random locations on a screen. The monkey saw its hand and fingers and the target before and after completion of the movement, but the movement itself was performed without visual feedback. In this case, Purkinje cells in cerebellar lobules HIV–HVI exhibited multiply timed climbing fiber responses at three different stages of the movement (first, second, and third responses). The third response occurred at the end point of the movement, apparently representing visually perceived deviations between the target and the reaching finger’s end position. However, the first and second Purkinje cell responses appeared too early to be interpreted similarly (Kitazawa et al., 1998). They apparently arose within the neuronal circuit controlling the hand-reaching movement (see Chapter 15, “Internal Models for Voluntary Motor Control,” for further details). Reaching experiments have also been performed in cats to test the unique function of a set of propriospinal interneurons that have been studied extensively by Anders Lundberg and his colleagues (Chapter 1, “Neuronal Circuitry: The Key to Unlocking the Brain”). These interneurons, which are located in the C3–C4 spinal segments, receive major excitatory inputs from the corticospinal tract and also from the rubro-, reticulo-, and tectospinal tracts. The cells mediate their converged input, and that from other sources, to forelimb motoneurons (Illert et al., 1977, 1978). Importantly, these C3–C4 propriospinal neurons have ascending collaterals to the lateral reticular nucleus (Chapter 6, “Pre- and Post-Cerebellar Cortex Neurons”), a major mossy fiber source for the cerebellum (Illert and Lundberg, 1978). The neurons thus appear to control forelimb movements while receiving adaptive modulation from the cerebellum. To test this hypothesis, cats were trained to make a fast, visually guided reaching movement toward a small tube containing a morsel of food, while the trajectory and velocity of the wrist movement were being monitored. A set of two or three such tubes were positioned in front of the animal at shoulder level, and the correct tube for retrieving the food was indicated by light-emitting diodes. The light was randomly switched off in one tube after a variable delay and switched on in one of the other tubes. The cat initiated a fast correction of the movement trajectory when switching from the left to the right tube. The switching latencies observed were as short as 40–60 milliseconds (Pettersson et al., 1997). After delays for retinal processing and electromechanical

13 • Voluntary Motor Control

153

delays in the muscles were subtracted, it seemed likely that within a time span of less than 20 milliseconds, information was processed to estimate the amount of correction needed to reach the new target position and give new command signals via subcortical systems to the propriospinal neurons (Alstermark et al., 2007). How the cerebellar circuit computes such new signals is still an open question (see Chapter 18, “Concluding Thoughts,” Section 3). In another study, Purkinje cell activity was observed in cats trained to use the left paw to track a visual target moving from the left to the right in front of the animal. For successful reaching, the cat had to predict where the target would be when it received the go signal (brightening of a light-emitting diode) for initiation of the tracking movement. Some (23/50) of the Purkinje cells sampled in the lateral cerebellar D1-zone of crus I displayed an increase or a decrease in simple-spike discharge rate in response to the onset of target movement. Other of the cells (14/33 tested) responded to the go signal with modulation, the great majority (12/14) with an increase in discharge rate. The results were interpreted to suggest that these Purkinje cells were predicting the initial movement of the target (Miles et al., 2006). TMS has been applied to disrupt briefly ongoing neuronal processes in the cerebellum. In the Miall et al. (2007) study, participants viewed a virtual image of a static target in three-dimensional (3D) space ahead of them and started each trial by lifting their right index finger from a start key and moving steadily toward their right. Liquid crystal device (LCD) goggles blocked the view of their hand and of the target as soon at the start key was released. An auditory go cue, 500–1500 milliseconds after trial onset, instructed them to make a rapid upwardand leftward-pointing movement to the virtual target (Color Plate XV). Their index finger had typically moved laterally 10–40 centimeters from its original position when the go cue was delivered. Final positional errors on control trials were small and averaged 4.2 centimeters across all conditions. Thus, participants were normally able to compensate for their initial lateral arm motion and reach the target despite the lack of visual feedback. For a random 50% of trials in each block, three TMS pulses were delivered at 50, 100, and 150 milliseconds within their reaction time after the auditory go cue. Errors in setting the initial direction and final finger position of this target-reaching movement were significantly larger during cerebellar stimulation than during control movements. From this measurement, it was calculated that the reaching movement was planned and initiated, as based on hand position, which was estimated to be 138 milliseconds before the hand actually started to move toward the target (Miall et al., 2007). Throwing a ball. Human subjects’ ability to accurately throw a ball of clay at a visual target is improved by daily training. It has been shown that when wearing

154

the cerebellum

wedge-prism spectacles, control subjects initially threw in the direction of the prism-bent gaze, but after repeated throws they became adapted and were able to hit the target. Immediately after removal of the prisms, the adapted throw persisted, this being the so-called negative aftereffect. Repeated throws were required to adapt back to the control situation. Patients with lesions in the cerebellum or related structures showed impaired or absent prism adaptation (Martin et al., 1996a). Furthermore, training control subjects with the right hand did not generalize to left hand performance, and overhand training seldom generalized to underhand throwing. When two subjects threw with the same hand (right) and the same type of throw (overhand) alternately, with and without prisms, over a period of six weeks, they gradually learned to hit the target on the first throw, with and without prisms. The two gaze-throw calibrations (prism and no-prism) were retained for >27 months (Martin et al., 1996b).

13-4 Hand Grip The skill in using fingers (digital dexterity) varies over seven ranks along the evolution of mammals to primates (examples in brackets): (1) Fused retrained fingers [horse]; (2) Nonconvergent digits [cat]; (3) Nonprehensile digits [rat]; (4) Nonopposable thumb [marmoset]; (5) Pseudo-opposable thumb [bushbaby]; (6) Power grip [chimp]; (7) Precision grip [man] (Napier and Napier, 1967). Attempts have been made to correlate this rank of digital dexterity to the size of the corticospinal tract, but in vain (Heffner and Masterson, 1975). Digital dexterity might well be related to the development of the structures involved in the cerebrocerebellar communication loop. The following studies have provided knowledge about neuronal mechanisms for the highest ranks (6 and 7) of digital dexterity. In grasping, lifting, and replacing an object by hand, humans exhibit subtle grip force control with sufficient yet minimal force, thus minimizing the risk of crushing or inadvertently dropping the object. When human subjects lift small objects using a precision grip between the tips of the fingers and thumb, the ratio between the grip force and the load force (i.e., the vertical lifting force) takes into account friction between the object and the skin. Johansson and Westling (1987) provided direct evidence that signals in tactile afferent units are utilized for this purpose. Tactile afferents were readily excited by small but distinct slips between the object and the skin, which were sensed as vibrations of the object. Such adaptation in grip force versus load force coordination should involve the cerebellum, and indeed, cerebellar lesions in stroke patients exhibited slowed grip force development and impaired grip force versus lift force coordination (Müller and

13 • Voluntary Motor Control

155

Dichgans, 2002). In monkeys trained to perform a grasp-lift-hold task, inactivation by muscimol injection into the anterior interpositus nucleus region produced pronounced dynamic tremor and dysmetric movements of the ipsilateral arm when the animal performed unrestrained reaching and grasping movements. In contrast, no immediate deficits were observed during a 15–20 minute period after muscimol injection in the dentate nucleus, albeit some effects were observed after several hours (Monzée et al., 2004). Thus, grasp-lift-hold tasks appear to be controlled primarily by the anterior interpositus nucleus, and not the dentate nucleus. To examine neural events underlying the preceding type of task, researchers trained monkeys to use a precision grip in order to lift and hold for 2.5 seconds an instrumented object at a fixed height. In some blocks of 20–30 trials, a downward force-pulse perturbation simulating object slip was applied to the object after 1.5 seconds of stationary holding. The animals were required to resist the perturbation to obtain a fruit juice reward. The perturbations produced two types of responses. (1) They elicited invariably a reflex-like, time-locked increase in grip force at latencies between 50 and 100 milliseconds. These persisted as long as the perturbation was applied. (2) When the trials were repeated, a preparatory increase in grip force emerged prior to the onset of the perturbation, and it was extinguished slowly after the perturbation was removed (Monzée and Smith, 2004). This preparatory response may represent a learned component of the grasping movement that develops during repeated trials. During repeated trials of a grasping-lifting-holding task, single cells were recorded in and around the interpositus and dentate nuclei and the overlying cerebellar cortex (Espinoza and Smith, 1990; Dugas and Smith, 1992; Monzée and Smith, 2004). Among those nuclear cells whose activity was related to grasping and lifting, the vast majority responded to a downward force-pulse perturbation simulating object slip (see preceding description). The majority of these cells displayed both reflex-like and anticipatory responses to the perturbation. They were confined to the dorsal anterior interpositus nucleus, that is, adjacent to, but not within, the dentate nucleus (Monzée and Smith, 2004). The object-slip perturbation also increased powerfully simple-spike discharges in Purkinje cells recorded in the anterior paravermal and lateral cerebellar cortex. The strong responses of these Purkinje cells suggest that the cerebellum participates in corrective responses. Repetitive perturbations elicited preparatory short-latency responses from many of these Purkinje cells (Dugas and Smith, 1992), which might represent preparatory motor responses in cerebellar circuits. In an fMRI study, control human subjects performing a precision grip with the right arm and hand displayed activity only in the right, anterior, and superior

156

the cerebellum

cerebellum and/or biventer of the left cerebellum (Kawato et al., 2003). These cerebellar regions have been implicated in forward models of arm movement (Chapter 15). In patients with cerebellar lesions, however, task performance was impaired as revealed by a disruption of temporal coordination between proximal (lifting) muscles and (gripping) fingers. This result suggested that the cerebellum helps coordinate the timing of multijoint movement sequences. Damage to the dentate nucleus or, in particular, its afferent input, leads to an increase in grip force (Fellows et al., 2001). Modeling has been carried out on the control process in humans undertaking the preceding types of task, which included variable initial grip apertures and perturbations and variations in object size, location, and orientation. This modeling incorporated slip sensations as error signals in the cerebellum to adapt phasic motor commands to tonic force generators associated with output synergies controlling grip aperture (Ulloa et al., 2003, 2010; Ulloa and Bullock, 2003). A computational model has also been constructed that involves the cerebellum in learning anticipatory grip force control by referring to grip force error estimation and sensory input on deformation of the finger tips (De Gruijl et al., 2009).

13-5 Operant Conditioning In operant (instrumental) conditioning, humans and animals learn to behave in a specific manner to obtain rewards or avoid punishments. The major mechanism for this process is generally considered to reside in the cerebral cortex. The cerebellum, however, may also be involved, as revealed by an experiment in which a monkey was successfully trained, using operant conditioning, to repeatedly lift a lever in response to visual stimuli. At an early stage in the learning process, when the monkey still lifted the lever randomly, short-latency electrical responses to a light stimulus appeared bilaterally in certain areas of the prefrontal, premotor, and prestriate cortices. These responses became gradually larger as the monkey’s training progressed. When the monkey started to respond to the stimulus with the appropriate movement, premovement potentials appeared in the forelimb motor cortex, and the size of responses in the premotor cortex increased. As the movement became faster and more skillful, late premovement potentials emerged and became even more marked, larger, and steeper in the forelimb motor cortex contralateral to the moving hand. A cerebellar hemispherectomy contralateral to the motor cortex eliminated these potentials. This suggested the participation of the neocerebellum at an advanced stage of the learning process (Sasaki et al., 1981; Sasaki and Gemba, 1982).

13 • Voluntary Motor Control

157

According to a standard definition of “skill” as being an adaptive behavior acquired through practice (Chen et al., 2005), operantly conditioned changes in the short-latency (tendon tap) spinal stretch reflex and H-reflex constitute simple motor skills. They can be manipulated to increase or decrease the size of response to an operant conditioning task, as demonstrated in monkeys and rats (Chen and Wolpaw, 2005). The two reflexes changed when the testee was exposed to an operant conditioning protocol that gradually decreased (down-conditioning) or increased (up-conditioning). For example, when the H-reflex of the soleus muscle was elicited in freely moving rats by stimulating the posterior tibial nerve, downconditioning was induced by giving a food reward 200 milliseconds after the Hreflex amplitude became smaller than a criterion value (Wolpaw and Chen. 2006). Concerning neuronal mechanisms for this form of adaptation, it has been reported recently that down-conditioning of the H reflex is accompanied by an increase in the number of involved GABAergic interneurons and their synaptic terminals in the spinal cord (Wang et al., 2006, 2009). This may have caused the down-conditioning, but on the other hand, ablation of the interpositus and lateral cerebellar nuclei in down-conditioned rats caused an immediate increase and a delayed increase with an overshoot in the H-reflex. This effect was observed for as long as 150 days. Therefore, the cerebellum appears also to play an essential role in the maintenance of down-conditioning (Wolpaw and Chen, 2006). Because a lesion experiment indicated that down-conditioning requires the corticospinal tract and does not require other major descending pathways (Chen and Wolpaw, 2002), the cerebellum may use this pathway in the down-conditioning process.

13-6 Source(s) of Central Instruction Signals Voluntary control systems differ from reflex control systems in that the former operate with instruction signals generated centrally within the cerebral cortex (Figure 9C), whereas the latter are driven primarily by peripheral stimuli. The motor cortex receives instruction signals that designate the content of the movement to be performed. For example, this can be a desired trajectory for an arm movement or even a complex program of movements. The primary motor cortex forms motor command signals and sends them to the brainstem and spinal segmental levels via the corticospinal descending tract. In higher nonhuman primates and humans, part of the primary motor cortex projects corticospinal fibers to directly innervate shoulder, elbow, hand, and finger motoneurons (Rathelot and Strick, 2009). On the other hand, the rostral region of the primary motor cortex represents the “old” primary motor cortex, which sends its descending commands to motoneurons indirectly via segmental circuits (see Lemon, 2008).

158

the cerebellum

What are the sites for generation of the instruction signals to the primary motor cortex? These should arise from the cortical areas devoted to motor planning and preparation, that is, directly from the anterior cingulate gyrus, supplementary motor area, and premotor area and indirectly from the presupplementary motor area via the supplementary motor area. Electrical stimulation of the supplementary motor area in patients has been shown to induce limb movements and to also evoke sensations such as the urge to perform a movement or anticipation that a movement is about to occur (Fried et al., 1991). Stimulation of the ventral bank of the anterior cingulate sulcus evoked an irresistible urge in a patient to grasp something, resulting in exploratory eye movements and a wandering arm (Kremer et al., 2001). These findings are in contrast to the finding that TMS of the primary motor cortex does not induce conscious awareness even when it produces a movement (for example, see Haggard et al., 2002). An fMRI imaging study revealed that self-paced thumb movements caused cerebral activation to spread from the anterior cingulate gyrus through the supplementary motor area and premotor area to the primary motor and sensory cortices. This cascade in the cerebral cortex occurred temporally in parallel to cerebellar activation, which propagated from lateral to medial parts of the cerebellar cortices (Hulsmann et al., 2003). These observations support the view that instruction signals for voluntary movements arise from “higher” cortical regions that include the anterior cingulate gyrus, supplementary motor area, and premotor area.

13-7 Summary Neuronal mechanisms for voluntary movements of the arms, hands, and fingers have been studied using observations on the effects of cerebellar lesions or TMS and brain imaging in monkeys and humans and unit recording from Purkinje cells in monkeys. These studies have been fruitful in revealing some of the basic neuronal mechanisms controlling pointing, reaching, throwing, grasping, and gripping. These are the basis for developing the concept of the internal-model-based control discussed in Chapter 15.

14 Voluntary Eye Movement

14-1 Introduction For securing well-focused foveal vision, the frontal eye field of the cerebral cortex controls three types of voluntary eye movement; smooth pursuit, saccades, and vergence. An additional eye field is located in the dorsomedial frontal cortex, which appears to play roles in integrating voluntary eye movements with nonoculomotor cortical functions. Four distinct areas of the cerebellum are involved in the three types of eye movements: (1) paraflocculus/flocculus, (2) lobulus petrosus of the paraflocculus, (3) vermal lobules VI/VII, and (4) crus I and crus II (corresponding to HV–HVII). These four areas have different roles in tracking objects moving in three-dimensional space.

14-2 The Frontal Eye Field The frontal eye field in monkeys is located in Brodmann’s area 8 (see Figure 2). Brain imaging and lesion data revealed a remarkable consistency regarding the rostro-caudal and dorso-ventral location of the human frontal eye field, but there was a marked variability along the mediolateral axis, which might challenge the commonly held view of the frontal eye field being located in Brodmann’s area 8 (Pause, 1996). Single-unit recording in monkeys and imaging the human brain have shown that the frontal eye field is certainly involved in the execution of eye movements. The field projects to the superior colliculus. The latter’s temporary inactivation (e.g., by the local injection of lidocaine or muscimol) suppressed saccades (Hanes and Wurtz, 2001). Note also that the frontal eye field provides extensive feedback connections to the extrastriate visual cortex, thereby suggesting that neuronal activity in the field can influence neural processing in the extrastriate visual cortex (Schall et al., 1995). 159

160

the cerebellum

Using intracortical microstimulation, subregions of the frontal eye field in monkeys have been mapped for their effects on smooth versus saccadic eye movements. This revealed that the saccadic subregion of the frontal eye field is distinct from the subregion for smooth pursuit, a result that was confirmed by single-unit recording (Tian and Lynch, 1996). In an fMRI study on human subjects, it was shown that frontal eye field activation during smooth-pursuit performance was smaller than that during saccades. This finding is consistent with the relative size of the two subfields found in humans versus monkeys. Another of the fMRI findings was that the mean location of the smooth pursuit-related frontal eye field in humans was more inferior and lateral than the location of the saccade-related field (Petit et al., 1997). Another single-unit recording study in anaesthetized monkeys showed that frontal eye field neurons have patterns of discharge that differed for saccadic versus pursuit movements. Two types of neurons (I and II) were identified. Type I cells fired during saccades in a specific direction and during the saccade-like, fast phase of nystagmus. These neurons were silent during slow-pursuit movements. In contrast, Type II cells showed steady discharges when the eyes were oriented in a specific direction, and they also discharged during smooth-pursuit movements and the slow phase of nystagmus (Bizzi, 1968). In the frontal eye field, certain neurons signal the location of conspicuous and meaningful stimuli that may be the targets for saccades. Other neurons control saccade production as to whether and when the gaze shifts. The presence of two loosely coupled distinct neural processes, one for visual selection and the other for saccade production, appears to be required for flexible, visually guided behavior involving switching between prosaccade (movement to an indicated target) and antisaccade (in the opposite direction) movements (Schall, 2002; see also following sections). The various patterns of frontal eye field discharge have also been found in the medial superior temporal visual area (MST) and the supplementary eye field (Fukushima et al., 2004; Akao et al., 2005a,b; see following sections). Moreover, it has been shown that a region of the frontal cortex located immediately anterior to the saccade-related frontal eye field region is involved in vergence and ocular accommodation (Gamlin and Yoon, 2000). MST and frontal eye field neurons are likely involved conjointly in the initiation of vergence eye movements.

14-3 Smooth Pursuit Humans and monkeys use information from their high-acuity retinal fovea to help control voluntary, well-executed binocular eye movements that can track a small

14 • Voluntary Eye Movement

161

moving target. When trained to pursue a target moving in the horizontal plane in a controlled ramp mode, the monkey shows first an initial slow, smooth pursuit; second, a small catch-up saccade; and finally, a postsaccadic smooth pursuit that almost matches the ramp target’s velocity. Feedback is not available for ~100 milliseconds after the onset of such movements due to a delay in visual processing (Smith et al., 1969). Hence, the pursuit at the initial 100 milliseconds of the target’s movement operates in a feedforward manner. This requires the cerebellum to intervene, just as it does in other types of eye movement like the VOR, OKR, and OFR (Chapter 10, “Ocular Reflexes”). Neuronal circuits for smooth pursuit involve the temporal and frontal lobes of the cerebral cortex and distinct regions of the cerebellum (Lisberger et al. 1987; Keller and Heinen, 1991; Krauzlis, 2004; Thier and Ilg, 2005). The middle temporal (MT) and medial superior temporal (MST) areas in the superior temporal sulcus process visual motion and oculomotor signals that are typically required for pursuit, and these are conveyed to the flocculus/ventral paraflocculus of the cerebellum via the pontine nuclei, primarily through the dorsolateral pontine nucleus. A second cerebro-cerebellar pathway originates in the frontal eye field and continues through the nucleus reticularis tegmenti pontis, which provides outputs exclusively to the cerebellum: in this case lobules VI and VII of the vermis. There is also evidence for the involvement of the cerebellar hemisphere, crus I and crus II (HV–VII), and the lobulus petrosus (see following sections). The cerebellum was shown to be implicated in the adaptation of smooth pursuit, when it was induced by repeating the target’s movement for 100–300 milliseconds immediately after the onset of a catch-up saccade. Injections of NO absorber or NO synthase inhibitor (that blocks conjunctive LTD) into the subdural space above the paraflocculus-flocculus scarcely affected the velocity of smooth-pursuit but these injections markedly depressed an adaptation of smooth pursuit. This suggested that conjunctive LTD underlies adaptation of the velocity of smoothpursuit (Nagao and Kitazawa, 2000). Another role of the cerebellum was revealed when the target was turned off temporarily, while a monkey was following circular trajectories that were clockwise or counterclockwise. The fact that pursuit was maintained in the absence of the target gave sure evidence of the predictive nature of the smooth pursuit (Suh et al., 2000). While a monkey was required to track a target moving along one trajectory selected randomly from four sum-of-sines trajectories (two horizontal, two vertical), Purkinje cells located in the ventral paraflocculus/flocculus discharged simple spikes for 12 milliseconds before eye motion. This time was slightly longer than the 9-millisecond transmission delay

162

the cerebellum

between flocculus stimulation and eye motion. This suggests that these Purkinje cells drove pursuit along predictable sum-of-sines trajectories (Suh et al., 2000). Purkinje cells in regions of the monkey’s ventral paraflocculus/flocculus have been shown to receive mossy fiber inputs of vestibular, visual, and oculomotor origins, as recently reviewed (Lisberger, 2009). These Purkinje cells exhibit simplespike responses during the initiation of smooth-pursuit eye movements. During pursuit of sinusoidal target motion, these Purkinje cells discharged complex spikes modulated out-of-phase with the simple-spike firing rate (Stone and Lisberger, 1990b). For target motion in the preferred direction, Purkinje cells in the same regions showed a large transient increase in simple-spike firing rate at the onset of pursuit, a smaller but sustained increase during the maintenance of pursuit, and a smooth return to baseline firing at the offset of pursuit (Krausliz and Lisberger, 1994). Pukinje cells in the posterior vermis also exhibited smooth-pursuit eyemovement-related activity (Suzuki and Keller, 1988). In another study on monkeys, ablation of the vermal VI/VII lobules induced changes in the dynamic properties of smooth-pursuit eye movements; during the open-loop period (the first 100 milliseconds), in particular. Changes included a decrease in peak eye acceleration and a decrease in the velocity at the end of the open-loop period. When the pursuit was adapted by halving or doubling eye velocity (i.e., by wearing lenses at ×0.5 and ×2 magnification, respectively), the main pattern of change was a decrease in peak acceleration following the former training and an increase in the duration of peak acceleration following the latter. This adaptive capability was impaired by lesioning vermal lobules VI/VII. These results suggest that lobules VI/VII play a critical role in both the immediate online and the adaptive control of smooth pursuit (Takagi et al., 2000). Three lines of evidence suggest the involvement of the lateral cerebellar hemispheres in smooth-pursuit eye movements: (1) in human patients, unilateral cerebellar lesions impaired initiation of ipsilateral smooth-pursuit eye movements (Straube et al., 1997); (2) in monkeys, electrical stimulation of crus I and crus II (HV–VII) and the dentate nuclei evoked smooth-pursuit-like eye movements (Ron and Robinson, 1973); and (3) in monkeys, destruction of the ipsilateral unilateral cerebellar hemisphere impaired initial pursuit movements and decreased the velocity of postsaccadic movement. The pursuit’s adaptation was also impaired (Ohki et al., 2009). In monkey cerebellum, the lobulus petrosus is inserted between the dorsal and ventral paraflocculus. This lobule receives inputs from visual system-related pontine nuclei, and it projects to eye-movement-related cerebellar nuclei. Results obtained in monkeys using tracers have shown that the lobulus petrosus and crus I/II of lobule HVII share some of their mossy and climbing fiber inputs, suggesting

14 • Voluntary Eye Movement

163

that these two areas have similar functional roles in the control of smooth-pursuit movements. In this study, monkeys tracked a moving ramp target using an initially slow eye movement, then a small catch-up saccade, and finally a postsaccadic pursuit that nearly matched the velocity of the moving ramp. After unilateral lesioning of the lobulus petrosus (by local injections of ibotenic acid), no consistent changes were seen in the amplitude of the catch-up saccades, but the velocity of postsaccadic pursuits in the ipsilateral and downward directions decreased by 20%–40%. These deficits lasted for at least one month, after which some recovery was observed (Xiong et al., 2010). These findings suggest the involvement of the lobulus petrosus in the monkey’s control of smooth pursuit. In yet another study, monkeys were trained to pursue a moving target either with their head moving freely in the horizontal plane (with their eyes voluntarily held stationary in their orbits) or with their freely moving eyes (with their head voluntarily held stationary in space). Neurons in the rostral portions of the nucleus reticularis tegmenti pontis were found to encode head-pursuit velocity and headpursuit acceleration. The vast majority of the tested neurons exhibited responses to both head and eye pursuits. As such, the neurons’ discharge contained a provisional gaze-pursuit signal (Suzuki et al., 2009). The frontal eye field projects to the nucleus reticularis tegementi pontis, which, in turn, projects to the cerebellar nuclei. The nucleus reticularis tegmenti pontis also receives inputs from the superior colliculi and pretectal midbrain areas (Gamlin and Clarke, 1995). Therefore, rostral portions of the nucleus reticularis tegmenti pontis are a strong candidate for the source of an active head-pursuit signal that projects to the cerebellum: that is, specifically to the Purkinje cells in vermal lobules VI and VII that have targetvelocity and gaze-velocity-encoding properties. Concerning neuronal mechanisms of smooth pursuit, there is as yet no consensus among researchers. Lisberger (1994) assumed a strong positive feedback loop from the vestibular nucleus to the cerebellar cortex (as in Figure 39B). This structure was thought to play an essential role in reproducing both VOR adaptation and smooth pursuit. On the other hand, Tabata et al. (2002) assumed that the pursuit driving command was generated and maintained in the MST area (as in Figure 39C). There seems to still be much to work out in the attempts to model neuronal mechanisms of smooth pursuit.

14-4 Voluntary Saccades Saccades can be evoked not only reflexively via the superior colliculus and the brainstem circuit (Chapter 10), but also voluntarily via the frontal eye field of the

164

the cerebellum

cerebral neocortex. Lesioning of the frontal eye field causes deficits in (1) generating saccades to briefly presented targets, (2) the production of saccades to two or more sequentially presented targets, and (3) the selection of simultaneously presented targets. Thus, in voluntary saccades, the frontal eye field acts as the controller and the brainstem saccadic system and eyeballs act as the controlled object. In a typical short-term saccadic adaptation protocol, the target moved midflight during the saccade either toward (gain-down) or away (gain-up) from initial fixation, and so caused the saccade to complete with an endpoint error. In the gaindown paradigm, the adapted saccades showed reduced peak velocities, accelerations, and decelerations, and increased durations compared with a control saccade of equal amplitude. (Ethier et al., 2008). The properties required for a model involved in voluntary saccades will be discussed later in Chapter 15, “Internal Models for Voluntary Motor Control.” Purkinje cells whose simple-spike discharge rates were modulated in close correlation with saccadic eye movements were found in fairly restricted areas in the cerebellar hemisphere, mostly in crus II with some in the deep folia of crus I. One study showed that two-thirds of saccade-related Purkinje cells began to change their simple-spike discharge rate 20–100 milliseconds prior to the onset of saccades. The remaining one-third changed their activity at approximately the time of saccade onset. These saccade-related Purkinje cells showed no changes in their activity during smooth-pursuit eye movements (Mano et al., 1991). Thus, a group of Purkinje cells in crus I and II of the cerebellar hemisphere must play a role in the control of voluntary saccadic eye movements.

14-5 Vergence Vergence refers to eye movements that rotate the eyes simultaneously in opposite directions (disconjugate eye movements). In frontal-eyed human and nonhuman primates, whereas the smooth-pursuit system moves both eyes in the same direction to track movements in the frontal plane (frontal pursuit), the vergence system moves the left and right eyes in the opposite direction to track targets moving toward or away from the observer (vergence tracking). Information related to vergence eye movements, namely retinal disparity and the blur signals that elicit them, is coded independently of signals related to frontal pursuit. Fukushima et al. (2002) demonstrated that frontal eye field neurons modulate strongly during both frontal pursuit and vergence tracking, thereby suggesting that they play a role in the control of three-dimensional eye movements. These neocortical neurons may

14 • Voluntary Eye Movement

165

function as part of a system that enables primates to track and manipulate objects moving in a three-dimensional space. Recently, vergence-related neurons have been recorded in the posterior vermis of monkeys (Nitta et al., 2008). Also, vermal lesions have been shown to impair vergence (Takagi et al., 2003). However, there may well be multiple areas of the cerebellum involved in the control of vergence, as is clearly the case for smooth pursuit.

14-6 The Dorsomedial Frontal Cortex In addition to the frontal eye field located in the dorsolateral frontal cortex, another eye-movement-related area resides within the dorsomedial frontal cortex (DMFC). As reviewed by Tehovnik et al. (2000), the DMFC differs from the frontal eye field in several respects. It contains neurons that contribute to the control of both limb and eye movements, whereas the frontal eye field is dedicated to the execution of saccadic and smooth-pursuit eye movements. A study showed that saccades evoked by electrical stimulation of most sites within the DMFC tended to shift the center of gaze to a particular location within craniotopic (head-centered) space, whereas the frontal eye field had a retinotopic (eye-centered) code for saccades (Schlag and Schlag, 1987). The DMFC contains a somatotopic map with the eyes represented rostrally and the hindlimbs caudally. This is in further contrast to the frontal eye field, which has no such map. Furthermore, lesions of the DMFC have a minimal effect on the production of saccadic eye movements and no effect on the execution of smooth-pursuit movements. Imaging in humans and singleunit recording in monkeys have suggested that the DMFC is involved in various motor tasks unrelated to eye movements. Taken together, whereas the frontal eye field acts as the controller of oculomotor neuronal circuits for voluntary saccadic and smooth-pursuit eye movements, the DMFC seems to expand on the functions of the frontal eye field by integrating them with non-oculomotor cortical functions. The DMFC consists of subareas, the supplementary eye field (medial eye field), and the presupplementary eye field. Whereas the frontal eye field has a prominent role in prosaccade initiation, the supplementary eye field is required for antisaccade tasks (Everling and Fischer, 1998). In this task, the subject was instructed to look not at the location of a flashed stimulus as the cue but in the opposite direction, at an equal distance from the central fixation point. The subject was also instructed to glance at the cue. This ability of antisaccade was lost in patients with lesions in the frontal lobe (Guitton et al., 1985). Neuronal discharges in the supplementary eye field were found to be consistently greater before antisaccades than before prosaccades with the same trajectories (Schlag-Rey et al.,

166

the cerebellum

1997). How the brain computes an inverted vector for antisaccades is still an open question.

14-7 Summary The frontal eye field acts as the cortical controller for smooth pursuit, saccades, and vergence. For saccades, the brainstem saccade generator circuits involving the superior colliculus and A-zone-fastigial nucleus microcomplex provide the controlled object (Figure 40 in Chapter 12, “Adaptive Control System Models”). Yet to be identified are the neuronal circuits that provide the analogous controlled objects for smooth pursuit and vergence. Although both the frontal eye field and crus I and crus II of the cerebellar hemisphere are involved in smooth pursuit and saccades, there is no anatomical evidence for a loop connection involving these areas such as those demonstrated in other prefrontal areas (for further discussions, see Chapter 15, Section 4).

15 Internal Models for Voluntary Motor Control

15-1 Introduction After reviewing the data on voluntary movements of the arms, hands, fingers, and eyes in Chapters 13 and 14, we are now ready to address control system models of voluntary movements that have helped define the mechanisms and roles of the cerebellum in the control of voluntary movements. Chapter 1, “Neuronal Circuitry: The Key to Unlocking the Brain,” emphasized that two forms of internal models (forward and inverse) are essential components of the overall control system for such movements. Both address the role of learning in acquiring skilled movements. Internal models also explain how sensory cancellation occurs during learning to obviate the effects of unwanted sensory perturbations, these being inevitable when beginning new voluntary tasks, in general, and those requiring considerable skill, in particular.

15-2 Internal Forward Model In performing voluntary movements such as pointing and reaching (Chapter 13, “Voluntary Motor Control”), the primary motor cortex acts as a controller. It follows instructions received from its higher centers (e.g., supplementary motor cortex, premotor cortex, anterior cingulate gyrus) and, in turn, drives a controlled object composed of lower motor centers in the brainstem and spinal cord and a motor apparatus in the periphery. The primary motor cortex receives somesthetic signals via peripheral nerves as feedback (Asanuma et al., 1979). In addition, the primary motor cortex is incorporated into a unique structure of the brain; the cerebrocerebellar communication loop that functions as an internal forward model. A typical cerebrocerebellar communication loop is formed between the primary motor cortex and the C1/C3 zones of the intermediate part of the cerebellar hemisphere. These two areas are linked via the anterior interpositus nucleus, the 167

168

the cerebellum

ventrolateral thalamic nucleus, and the pontine nucleus/nucleus reticularis tegmenti pontis (Figures 43 and 46). This cerebrocerebellar communication loop has recently been remapped using the transneuronal transport of neurotropic viruses. It has been shown that neurons in the arm area of the primary motor cortex receive inputs from Purkinje cells located primarily in lobules IV–VI of the C1 and C3 zones. In turn, neurons in the arm area of the primary motor cortex project back to the equivalent arm sites in the cerebellum, primarily via the pontine nucleus (Kelly and Strick, 2003) (Figure 42).

Figure 42

Input-output organization of cerebellar loops with the primary motor cortex.

(Left) The distribution of Purkinje cells (small dots) that project to the arm area of the primary motor cortex. These neurons were labeled after retrograde transneuronal transport of rabies virus from injections into the arm area of the primary motor cortex. (Right) The distribution of granule cells (fine lines) that receive input from the arm area of the primary motor cortex. These neurons were labeled after anterograde transneuronal transport of the H129 strain of HSV1 from injections into the arm area of the primary motor cortex. The shaded areas on the flattened surface maps (diagrams on the left side of each panel) are unfolded on the right side of each panel to show the distribution of labeled neurons in the relevant cerebellar cortical fissures. The small icons of scissors in the diagram indicate places where the maps have been cut to facilitate the unfolding process. Scale bars, 15 millimeters. (From Kelly and Strick, 2003.)

15 • Internal Models for Voluntary Motor Control

169

The various roles of cerebrocerebellar loops have been considered. If it represents an internal model simulating the external feedback from a controlled object and if learning makes the simulation exact, the primary motor cortex would be able to perform precise control by referring to the output of this internal model instead of the real controlled object. A model based on a design of a “Smith Predictor” suggested that an internal model replaces long and unavoidable feedback delays (Miall et al., 1993). In developing a computational theory for control of a robot’s arm Kawato et al. (1987) defined a “forward model” as mimicking the input-output relationship of a controlled object. This was in contrast to another inverse model that mimicked a reciprocal relationship of the forward model (see following description). When the microcomplex shown in Figure 43 contains a forward model that simulates the kinematics of the controlled object, the primary motor cortex should be able to perform a precise movement using internal feedback from the forward model instead of external feedback from the real control object. Another closely related idea is that the cerebellum estimates the current state of the motor system. To this end, it may calculate a “state estimate” by combining sensory information about the last known position of the arm with predictions of its future responses to the most recent movement commands, and thereby accurately plan and control a reaching movement (Miall et al., 2007). This hypothesis was supported by the finding that TMS stimulation over the cerebellum caused errors in the later component of the trajectory of an arm reaching for a target (Chapter 13). Learning a voluntary movement by repeated practice can be considered to be a process whereby a forward model is formed and reformed in the cerebellum through modification of the input-output relationship of the involved (relevant) microcomplex. A well-formed forward model would certainly be of use in acquiring motor skills. When we begin executing a novel voluntary movement, we initially require feedback to ensure a progressively improving performance. But when the quality of the movement reaches a certain level, we are then able to perform the task without feedback. For example, one can quickly learn to take a finger to the nose with the eyes closed. This capability is impaired in dysmetric patients with cerebellar damage (Chapter 2, “Traditional Views of the Cerebellum”). It suggests that a forward model in the cerebellum is required to accurately locate the nose relative to the finger in the absence of visual feedback. An interesting case was reported by Sasaki (1985) and discussed by Leiner et al. (1987). It involved a medical doctor who suffered from an infarction in the posterior lobe of the cerebellum on the left side. One day, after he had a short attack, which disturbed his consciousness for about several seconds, he tried the finger-nose test on himself. As his finger approached his nose (or eye, ear, or navel), he noticed that the target disappeared

170

the cerebellum

from his mental image of his body. Instead, he imagined a space with a severalcentimeter radius around the target to be a “sea of clouds.” As a result, he had to point his finger into the middle of this vague space. This conscious experience suggests that a forward model within the cerebellum represents a target to be reached in a mental image of the body’s various parts. When one is learning a novel voluntary movement, the required forward model must be reformulated by climbing fiber signals that represent errors and induce a persistent modification of its input-output relationship. In Figure 43, it is assumed that the inferior olive (IO) is driven by three types of input signal to the primary motor cortex that consist of 1) central instruction, 2) external feedback signals from a sensory system, and 3) internal feedback signals from the forward model. The presence of such triple pathways converging onto the IO is consistent with the three-phase climbing fiber discharges that occur during a monkey’s hand reaching movement (Chapter 13, Section 3).

Figure 43

A forward-model-based control system scheme for voluntary movements.

See the text for further information about this figure. The shade in the upper middle encloses the microcomplex that acts as a forward model. The shade on the right encloses the controlled object. Abbreviations: AIP, anterior interpositus nucleus; C1/C3, two longitudinal zones in the intermediate part of the cerebellum; Co, controlled object; IO, inferior olive; M1, motor cortex; PN, pontine nucleus; PX, possible pathway to IO; VL ventrolateral thalamic nucleus. Symbol: dotted and slanted arrow from the IO’s climbing fiber pathway. (Based on Ito, 2001, 2006.)

15-3 Internal Inverse Model Internal inverse models determine appropriate motor commands from the instruction they receive about the desired motor consequences. An interesting design

15 • Internal Models for Voluntary Motor Control

171

was provided by the Kawato et al. (1987) unique two-degrees-of-freedom adaptive control for voluntary movement (Figure 44). It combined two controllers acting on a common controlled object that was composed of a spinal segmental motor system and its peripheral effector apparatus. The controller was a combination of the primary motor cortex, which mediated classic feedback control, and a microcomplex that provided feedforward control. The microcomplex acquired an inverse model of the controlled object by learning. Before this learning, feedback control by the primary motor cortex was dominant, but as learning proceeded in the microcomplex, feedforward control became predominant.

Figure 44 Inverse-model-based control system scheme for voluntary movements.

See the text for further information on this figure. The abbreviations are the same as those in Figure 43. d indicates descending tract neurons receiving inputs from both M1 and AIP. (Based on Ito, 2001, 2006.)

To act as an inverse model, a microcomplex must be connected with the primary motor cortex in a manner different to a forward model (compare Figures 43 and 44). Neurons in the supplementary motor cortex and the premotor cortex project to both the primary motor cortex (Fang et al., 2005) and the cerebellum via the pontine nucleus (Wiesendanger et al., 1979). Furthermore, the outputs of the primary motor cortex and the cerebellar interpositus nucleus converge on some descending tract neurons (d in Figure 44). For example, the rubrospinal tract neurons receive input from the anterior interpositus nucleus (Toyama et al., 1970) and also from the primary motor cortex (Tsukahara and Kosaka, 1968). Therefore, there seems to be an anatomical basis for postulating an inverse model for helping the primary motor cortex, but these connections need more detailed investigation.

172

the cerebellum

For their two-degrees-of-freedom control system, Kawato et al. (1987) postulated that the most plausible possibility was that errors are derived from the outputs of the primary motor cortex, which perform feedback control to generate motor commands (feedback-error learning) (Figure 44). A computer simulation of this for a reaching movement reproduced Purkinje cell complex-spike discharges in two phases: an early response locked to movement onset (corresponding to the first phase; see above), which was always present, and a later response (corresponding to the third phase) that disappeared after learning, albeit the learning was stable and maintained (Schweighofer et al., 1998). Therefore, the temporal patterns of complex-spike discharges during movements should vary depending on which model (forward or inverse) is adopted in voluntary motor control (see Section 15-4). In conformity with the feedback error-learning hypothesis, the primary motor cortex sends motor error signals to the IO via collaterals of the corticospinal tract and/or via the parvocellular red nucleus (Chapter 6, “Pre- and Post-Cerebellar Cortex Neurons”). Yttri et al. (2006) injected muscimol into the left parvocellular red nucleus of a monkey that had been trained to perform a right arm reaching task with and without 20 diopter Fresnel prisms that shifted the gaze 11.5° to the right. Parvocellular red nucleus inactivation did not affect the control (“no-prism”) gazereach calibration, but it impaired the learned calibration for right-shifting prisms. Moreover, the monkey was unable to adapt to stronger 30 diopter novel prisms that shifted the gaze 17.2° to the right. The parvocellular red nucleus appears to be necessary for the storage of error information from previous trials and the use of this error information to update motor programs in order to prevent an error in subsequent trials. Because this nucleus also receives inputs from the supplementary motor cortex, cingulate gyrus, frontal eye field, and posterior parietal area (Ralston 1994; Burman et al., 2000a, b), it may well mediate error learning in the cerebellum for both voluntary movements and motor actions (Chapter 16, “Motor Actions and Tool Use”).

15-4 Sensory/Motor Signals and Forward/Inverse Models The basic structures of forward- and inverse-model-based control systems are shown schematically in Figures 8, 43, and 44. To discuss how they differ in their operation, one must realize that instruction and sensory signals represent movement in spatial coordinates—that is, the description of motion of the body, or its various parts, in terms of position, velocity, acceleration, and direction (termed

15 • Internal Models for Voluntary Motor Control

173

“kinematics”). In contrast, motor signals commanding movements are represented in bodily coordinates—that is, the description of motion that includes force as the cause of motion (termed “dynamics” or “kinetics”). In this volume, we prefer “dynamics” as it is used widely in internal-model-based control of a robot manipulator (An et al., 1988). In simple control systems, these two types of coordinates can be approximately identical; for example, for the VOR, the spatial coordinates determined by the semicircular canals parallel the bodily coordinates determined by the activation of extraocular muscles (Ezure and Graaf, 1984a; Chapter 10, “Ocular Reflexes”). In complex control systems, however, such as those required for multijoint arm/hand/finger movements, the bodily coordinates are complex and clearly distinct from spatial coordinates. Therefore, sensory and motor signals need to be represented in their spatial and bodily coordinates, respectively, but not in their admixture. Here, we address the question of what rules determine the sensory/motor nature of mossy and climbing fiber information, and the nuclear output of a microcomplex. It follows that a forward model receives motor signals from a controller and generates sensory signals that provide information on the kinematics of the produced movement. In contrast, an inverse model receives instruction signals in spatial coordinates and generates motor commands like a controller. Therefore, the forward/inverse nature of a microcomplex can be distinguished by determining whether the output of this microcomplex represents kinematics or dynamics. It has been shown that Purkinje cells in lobules HIV–HVI of the intermediate and lateral parts of the cerebellar hemisphere encode the position of an arm and the direction and speed of its movements, thereby favoring the forward model (Fu et al., 1997: Roitman et al., 2005). Another similar example is grip force-load force coordination in holding a load stationary in space (Chapter 13). An fMRI study revealed the brain activity related to this coordination and demonstrated that the cerebellum was the most likely site for forward models to be stored (Kawato et al., 2003). After reviewing the data in this field, Ebner and Pasalar (2008) concluded that the simple-spike discharges of Purkinje cells do not have dynamics-related signals in keeping with an inverse dynamics model. However, Yamamoto et al. (2007) detected muscle dynamics representation in a group of Purkinje cells in lobules HIV–HVI of the intermediate parts of the cerebellar hemisphere. These Purkinje cells formed a group separate from other Purkinje cells that represented movement kinematics. These findings suggested that both forward and inverse models are utilized in the control of arm movements. However, it is desirable to test more cases of voluntary

174

the cerebellum

motor control to clarify how these two types of internal model are differentially utilized for voluntary motor control. It is also important to test the kinematic/dynamic nature of the activity of cerebellar/vestibular nuclear neurons, but not Purkinje cells, because the former provide the real output signals of internal models (as discussed in Chapter 12, “Adaptive Control System Models”). The forward and inverse models can also be distinguished by the difference in the anatomical design of control systems incorporating a microcomplex. If one wants to provide a forward model, it is essential to have a microcomplex within a cerebrocerebellar communication loop. Microcomplexes involving C1 and C3 zones are indeed linked in a loop with the primary motor cortex and hence are able to provide forward models (Figure 43). In contrast, microcomplexes located in the cerebellar hemisphere and involved in voluntary eye movements have no connections to the frontal eye field (Kelly and Strick, 2002). Hence, the cerebellar hemisphere cannot provide forward models to controllers in the frontal eye field. Possibly, the cerebellar hemisphere provides inverse models to the frontal eye field in the manner similar to that shown in Figure 44 for M1. This possibility is in accordance with a functional feature of eye movement, which, in particular for the saccades, is so fast that it cannot be controlled by internal feedback through a forward model. These considerations suggest an interesting dichotomy that voluntary limb movements are controlled with forward models, whereas inverse models control voluntary eye movements. This suggestion is worth testing by designing experiments that involve further recording from the cerebellum.

15-5 Climbing Fiber Signals The mixed sensory/motor nature of climbing fiber signals has been observed in the VOR and OFR (Chapter 10). In voluntary movement, climbing fibers convey either sensory or motor signals depending on the nature of the internal model involved: climbing fibers convey sensory signals to a forward model (Figure 43) and motor signals to an inverse model (Figure 44). In the decorrelation control algorithm proposed by Dean et al. (2002) and Porrill et al. (2004) a combination of motor signals in mossy fibers and sensory signals in climbing fibers produced efficient learning. If a positive correlation induces LTD and a negative correlation produces LTP, a cerebellar microcomplex can learn effectively even without monitoring motor errors. This situation applies to the forward model (Figure 43), but not to the inverse model (Figure 44). For the latter, Gomi and Kawato (1993) applied a feedback error-learning rule (see previous discussion). These two rules

15 • Internal Models for Voluntary Motor Control

175

are incompatible with each other because the former involves the sensory nature of climbing fiber activity, whereas the latter emphasizes its motor nature. What signals do climbing fibers convey in an actual voluntary movement? In a well-known study (Kitazawa et al., 1998), a monkey performed short-lasting reaching by moving the hand and arm (without seeing them) to touch a visual target that appeared at a random location on a screen. Complex Purkinje cell spikes occurred in three successive response phases. The spikes in the third phase ceased at the end of the reaching movements, suggesting that they represented visually perceived deviations between the instructed target and the reached finger position (consequence errors). The second phase appeared near the moment of reaching but too early to reflect the consequence of reaching. These spikes presumably reflected deviations between the instructed target and the target position predicted by the forward model (i.e., internal errors). This second phase is in keeping with forward model control equipped with an internal feedback (Figure 43), but it has no basis in terms of inverse model control without an internal feedback (Figure 44). The first phase of the preceding discharges occurred when the monkey initiated the reaching movement. At that moment, the cerebral cortex was being driven by instruction about the need for the reaching movement toward the visual target. The difference between the position of the ready-to-go hand and the instructed target position can be considered to be an “initial set error.” Alternatively, this error might represent the discrepancy between the target position actually viewed and that which the monkey anticipated from preceding experiences. This would be the case if the monkey had learned previously to reach a target that appeared repeatedly at the same position. In the experiment under consideration, however, the target’s position was randomized from trial to trial to prevent this type of learning. Therefore, the earliest complex-spike discharges in the Kitazawa et al. (1998) study were more likely to have reflected the first alternative: that is, the difference between the positions of the hand and target immediately prior to the reaching movement.

15-6 Combination of Forward and Inverse Models Kawato and his colleagues developed a new model with complex architecture: the “modular selection and identification for control (MOSAIC).” It was designed for motor learning and control as based on multiple pairs of forward (predictor) and inverse (controller) models. The MOSAIC learns simultaneously the multiple inverse models necessary for control as well as how to select a set of inverse

176

the cerebellum

models that is appropriate for a given task. It combines feedforward and feedback sensorimotor information so that controllers can be selected both prior to a movement and subsequently during this movement. The MOSAIC can operate in a number of ways. For example, it can learn to manipulate multiple objects and switch appropriately between them. After such learning, the model can generalize novel objects whose dynamics are within the range of already learned ones. In this manner, the MOSAIC can learn the dynamics required for a given movement task and select the controller before executing the task (Haruno et al., 2001).

15-7 Transition from Reflex Control to Voluntary Control We now know that a microcomplex is combined with a control system in two distinct ways. One way is defined in Chapter 12, Section 8 in terms of V- and C-types, in which a microcomplex is embedded in a control system in such a way that the nuclear component of the microcomplex serves both as a controller and as a part of an adaptive mechanism (Figures 28, 32, 33, 35–38, 40, 41). The other way is that a microcomplex forms an internal model to assist the cortical controller (Figures 43, 44). Note that in the latter cases the microcomplex is a separate unit from the controller, not part of it as in the former cases. How these two forms of microcomplexes switch as a function of the evolution of the cerebellum is an interesting question addressed in the following discussion. As mentioned previously, the internal forward model was conceived in connection with the cerebrocerebellar communication loop as assisting the primary motor cortex in voluntary motor control (Figure 43). However, this form of internal model may be provided whenever there is a communication loop attached to a controller. For example, the rubrospinal tract is known to extrude axon collaterals that project to the lateral reticular nucleus (Robinson et al., 1987), which is a source of mossy fiber afferents to the cerebellum (Chapter 6). It is therefore possible that a rubrocerebellar communication loop is formed (Figure 45), and it may enable a forward-model-based control of a long-looped reflex mediated by the magnocellular red nucleus (such as grasping). A microcomplex involving the C1/C3 zones and the anterior interpositus nucleus is also attached, possibly as an internal forward model, to the primary motor cortex (Figure 46). When we move further to the lateral cerebellar hemisphere, a microcomplex involving D1 or D2 zones and dentate nucleus forms the cerebrocerebellar communication loop (Figure 49, Chapters 16 and 17). It may thus appear that whereas the adaptive control form of microcomplexes (N- and C-types) prevails in the evolutionarily old cerebellum, it changes to the internal forward-model-based form located in the intermediate to the lateral cerebellar hemisphere.

15 • Internal Models for Voluntary Motor Control

177

Figure 45 Forward-model-based control by the magnocellular red nucleus.

This wiring diagram focuses on the recurrent projection from the rubrospinal tract to the cerebellum via the lateral reticular nucleus (LRN), which may form a rubrocerebral communication loop as indicated by the circle with arrows. Other abbreviations: IO, inferior olive; RNm, magnocellular red nucleus: RST, rubrospinal tract; SOT, spinoolivary tract. Symbol: filled bulbs at the end of projection lines, excitatory synapses. Other abbreviations and symbols are as in Figure 33. Reflexive grasping might well be subserved by this system.

An interesting aspect of control system structures is that anterior interpositus neurons project branches of the same axons to both the magnocellular red nucleus and the primary motor cortex (the latter via the ventrolateral thalamic nucleus) (Toyama et al., 1970). With these projections, the same microcomplex involving the C1/C3 zone and the anterior interpositus nucleus is linked to both the magnocellular red nucleus (Figure 45) and the primary motor cortex (Figure 46). How can the same microcomplex serve as a common forward model for two separate controllers? This is indeed possible because for both controllers, the microcomplex is to be tuned to simulate the common controlled object at segmental levels of the spinal cord. This is illustrated in Figure 47 for the case of a feedforward model. Thus, a microcomplex tuned during reflex activity can also be effective during voluntary activity, and vice versa. Such a hybrid control system (Figure 9C) is worthy of testing because it answers the long-standing question of how the same anterior interpositus neurons can control two descending systems at the same time.

178

the cerebellum

Figure 46 Forward-model-based control by the primary motor cortex.

This wiring diagram is like that in Figure 45. It focuses on the recurrent projection from the corticospinal tract (CST) to the cerebellum via the pontine nucleus (PN), which forms a cerebrocerebellar communication loop (indicated by the circle with arrows). Abbreviations: SOT, spinoolivary tract; VL, ventrolateral thalamic nucleus. Other abbreviations and symbols are as shown in Figure 45. This system serves for voluntary movement control of a limb by the primary motor cortex as the controller.

15-8 Sensory Cancellation Because internal feedback predicts the sensory consequences of a movement, it can be used to cancel the actual effects of this movement, which otherwise would induce sensations that might disrupt the movement’s execution. A clear example of sensory cancellation has been observed in a fish “cerebellum-like” neuronal circuit. It has been shown to generate discharge that cancels the disruptive signals of lateral line organs caused by swimming (Bell et al., 1997). This mode of forward model control corresponds to the efference copy mechanism proposed earlier by

15 • Internal Models for Voluntary Motor Control

179

Figure 47 Hybrid control of a reflex and a voluntary movement.

A candidate example of the hybrid controller proposed in Figure 9C can be implied in the connections to both the magnocellular red nucleus (RNm) and the primary motor cortex (M1) from a microcomplex consisting of the C 1/C 3 zones and the anterior interpositus nucleus (AIP). Other abbreviations are similar to those in Figures 43 and 44.

von Holst (1954). A unique example of sensory cancellation is that we perceive a self-generated tactile stimulus to be less ticklish than the same stimulus when it is applied externally by another means. In an fMRI experiment, Blakemore et al. (1998) revealed significantly less activity in bilateral secondary somatosensory cortices, the anterior lobe of the right cerebellum, and the anterior cingulate gyrus, for a tactile stimulus that was self- versus externally produced. Sensory cancellation may have functional implications that are much broader than usually considered for tickling in humans and the sonar system of whales and bats, which may be located in their paraflocculus (recall Chapter 2). The reason is that the requirement of sensory cancellation is common to any voluntary movement that inevitably provokes disturbances created by the surroundings. How sensory cancellation occurs in neuronal circuits involving the cerebellum, however, is not yet known. On the basis of current information obtained in neuronal circuit analyses, however, I would like to suggest the possibility that a key role is played by the pathway sequentially connecting the primary motor cortex, pontine nucleus, C2 zone, posterior interpositus nucleus, nucleus ventralis posterolateralis oralis (VPLo) of the thalamus, and then back to the primary motor cortex (Figure 48). A study using HRP tracing and single-cell recording showed that the VPLo relays sensory inputs from peripheral receptors to cells in the primary motor cortex (Horne and Tracey, 1979; Tracey et al., 1980). Sensory signals generated by tickling, for example, might be fed back via the VPLo to the primary motor cortex, whereas a microcomplex

180

the cerebellum

Figure 48 A possible device for sensory cancellation.

The proposed device is a microcomplex consisting of the C 2 zone and the posterior interpositus nucleus (PIP). When the primary motor cortex (M1) responds to a central instruction (Ci) by commanding a voluntary movement, the command signals are also sent to the microcomplex via pontine nuclei (PN) and return from the microcomplex to M1 via the ventrolateral posterior oralis thalamic nucleus (VPLo), which receives somatic sensory signals from the periphery. A tentative hypothesis entertained in the text is that this circuit serves for sensory cancellation of the sensory perturbation induced by a voluntary movement.

involving the C2 zone and posterior interpositus nucleus would provide internal feedback. Signals of the two pathways would meet at the VPLo and cancel each other. This hypothesis is well worth testing because it suggests that the C2 zone and posterior interpositus nucleus contribute not directly by modifying operation of the controller, but by participating in the sensory cancellation process that removes perturbations caused by voluntary movements. Caution is needed, however, because it is not yet known how the events in VPLo influence our perception.

15-9 Summary Analyses of neuronal circuit suggest that the intermediate part of the cerebellar hemisphere involves three basic functional mechanisms: (1) an embedded adaptive mechanism, (2) an internal-model-based mechanism, and (3) a sensory cancellation mechanism. How we acquire skill in the execution of a novel voluntary movement can now be explained on the basis of combinations of these three control mechanisms.

16 Motor Actions and Tool Use

16-1 Introduction In Chapters 10–15, we examined individual reflexes and voluntary limb and eye movements. Now, we are ready to consider a higher level of control of motor actions, in which a series of complex movements involving many body parts and even a tool are combined to achieve a certain goal. Here, the model-based control concepts proposed for the cerebellum in terms of forward and inverse models are extended to the cerebral cortex in terms of body schema and motor schema. Cooperation between these cerebral- and cerebellar-model-based controls seems to be the mechanism that enables us to perform complex motor actions such as piano playing and gymnastics. The understanding of these issues is still based largely on conceptual modeling, but major experimental advances might well occur in the near future.

16-2 Action Controllers in the Premotor Cortex As discussed in Chapter 1, “Neuronal Circuitry: The Key to Unlocking the Brain,” a controller of a motor action (popularly now called an “action controller”) is located in the premotor cortex. It receives instructions from the anterior cingulate gyrus and supplementary motor cortex and, in turn, acts on a controlled object, which nests the primary motor cortex and the segmental motor system including its peripheral effectors (Figure 12C). Anatomically, the premotor cortex is linked to the cerebellar hemisphere and dentate nucleus forming a cerebrocerebellar loop (Figure 49). It is also linked to the parietal association cortex (see Figure 50 and following text). In an fMRI study on normal subjects, a sequential key-press task of increasing length and complexity was imposed, without changing the frequency,

181

182

the cerebellum

force, and number of single finger movements. This task involved activation of a subset of cortical areas, notably the contralateral ventral and dorsal premotor cortices, in addition to the bilateral superior parietal cortices, left inferior frontal gyrus/Broca’s area, right dentate nucleus, and left visual association cortex (Haslinger et al., 2002). The results of this study suggested the importance of premotor-cerebellar-parietal circuits for the studied movements. Another fMRI study showed that during musical performance, musically naïve control subjects displayed stronger activation than did pianists in the anterior cingulate cortex, right dorsal premotor cortex, both cerebellar hemispheres, and the right basal ganglia. Also while playing “parallel” and “mirror” bimanual movements (five fingers of two hands were paired in-phase and anti-phase, respectively), control subjects exhibited stronger signal increases than pianists within the supplementary motor cortex, bilateral cerebellar hemispheres and vermis, bilateral prefrontal cortex, left ventral premotor cortex, right anterior insula, and the right basal ganglia (Haslinger et al., 2004). These findings suggested an increased efficiency in musicians of neuronal processing within the cortical and subcortical systems controlling bimanual pianoplaying movements. An fMRI study during cyclical hand tasks of spatiotemporal complexity and varying frequency also demonstrated that the dorsal premotor cortex and the cerebellum are critical sites for bimanual coordination. Subjects performed four such tasks of increasing complexity (unimanual left-right hand-, bimanual in-phase-, bimanual anti-phase-, and bimanual 90° out-of-phase movements) at four frequencies (0.9, 1.2, 1.5, 1.8 Hz). Activation in the supplementary motor area and superior parietal cortex were correlated mainly with increasing spatiotemporal complexity of the limb movements (Debaerea et al., 2004). The results are in accordance with the idea that the premotor cortex acts as the action controller for several types of highly skilled movements.

16 • Motor Actions and Tool Use

183

Figure 49 Wiring diagram for the D1 zone-dentate system.

The new abbreviations in this control system are DDN (dorsal portion of the dentate nucleus), PM (primary motor cortex), and VLx (subarea x of the ventrolateral thalamic nucleus). See Figure 46 for the other abbreviations and symbols.

16-3 Body Schema and Motor Schema We postulated in Chapter 1 that our cerebral neocortex acquires an “action schema” in the temporoparietal cortex, which helps the premotor cortex to control motor actions. Here, we propose that an action schema involves both “body schema” and “motor schema.” A body schema is a set of neural representations of the body and bodily functions (Stamenov, 2005). It may be encoded in a neuronal circuit similar to the “neuromatrix,” which was originally proposed to represent pain as a multidimensional experience produced by characteristic activity patterns of nerve impulses generated by a widely distributed neural network in the brain (Melzack, 1990). The related concept, motor schema, provides controllers that can be coordinated to bring about a wide variety of actions (Arbib, 2005). The body and motor schemata are heuristic concepts, which have no rigorous definition in the precise parlance of neuroscience, but their formalistic positions in the action control system resemble the cerebellar forward and inverse models proposed for voluntary motor control systems. We therefore assume that a cerebral body schema

184

the cerebellum

and a cerebellar forward model are connected to the premotor cortex in parallel (Figure 50). Likewise a cerebral motor schema and a cerebellar inverse model may be connected in parallel.

Figure 50 Model-based control of motor actions.

This block diagram shows the control system operation of the premotor cortex with the aid of a body schema and a cerebellar forward model. It is assumed that the body schema is formed during conscious effort to practice an action and that a cerebellar forward model then copies the body schema. By these means, the action can be performed automatically.

Let us consider the steps required to learn a complex motor action. First, the premotor cortex, as an action controller, acts on the controlled object, including the primary motor cortex, to perform the motor action while relying on external sensory feedback. Second, while this process continues, the parietal cortex acquires a body schema and a motor schema by a cortical mechanism known as associative network learning (Doya, 1999). These schemata provide models of the learned motor action that the premotor cortex is going to perform. This initial step of learning proceeds consciously and involves the premotor and primary motor cortices and the temporo-parietal cortex. The further step of learning involves the cerebellum (see the following section).

16 • Motor Actions and Tool Use

185

16-4 Roles of the Cerebellum in Motor Actions In Chapter 15, “Internal Models for Voluntary Motor Control,” we assumed that the cerebellum uses error learning to copy the kinematics and dynamics of the controlled objects including the motor apparatus. As an analogy of this proposal, we now assume that the cerebellum copies the body schema to a forward model (Figure 50). An anatomical finding supporting the assumption of the forward model is that the premotor cortex and the lateral cerebellum are reciprocally linked to form a cerebrocerebellar communication loop, which parallels the loop between the primary motor cortex and the intermediate part of the cerebellum for voluntary motor control. Indeed, the premotor cortex receives a projection from the cerebellar D1-zone via the dorsal portion of the dentate nucleus and area X of the ventrolateral thalamic nucleus (Dum and Strick, 2003) (Figure 51). In turn, the premotor cortex projects to the lateral cerebellum via the pontine nucleus (Allen et al., 1978; Wiesendanger et al., 1979). The premotor cortex should thus be capable of controlling a motor action by relying on cerebellar forward models of the body schemata. On the other hand, a motor scheme may be copied in an internal inverse model by feedback error learning, which would replace the motor schema. When this late phase of learning proceeds, a motor action can be performed automatically. In summary, we explain the entire brain mechanism for learning motor actions by assuming a sequential combination of (1) an initial feedback-control phase during which a body/motor schema is being formed, (2) a conscious cerebral phase relying on the body/motor schema, and finally (3) an unconscious cerebellar phase wherein a forward and inverse internal model, which copied the body/motor schemata, are in operation. In accordance with the preceding model, brain-imaging studies have demonstrated a co-activation of the parietal association cortex and the cerebellum during certain motor actions. For example, when fMRI brain images were compared for execution of a hand movement paced at 0.5 Hz by an auditory cue versus imagination (mental simulation) of the same movement at the same rate, it was shown that specific cortico-subcortical areas were more engaged in the mental simulation case, with the areas activated including the bilateral premotor, prefrontal, supplementary motor, and left posterior parietal areas and the caudate nuclei (Gerardin et al., 2000). In this study, the left parietal area was a likely candidate for representation of the body/motor schemata needed for the auditory-cued hand movements. The three-staged learning mechanism proposed previously for motor actions may explain a common observation in patients afflicted with Alzheimer’s disease (AD). They exhibit serious impairments of memory and intellect, yet they retain

186

the cerebellum

Figure 51 The cerebellothalamocortical circuit.

(A) A map of monkey dentate nucleus is unfolded with the dorsal surface upward and the caudal surface to the right side. (B) Selected cortical targets of cerebellothalamocortical circuits. Shading in this diagram indicates the cortical regions (lateral hemisphere only) that project to the cerebellum via the pons. (C) Origin of selected cortical projections from the ventrolateral thalamus. The cortical regions so indicated receive inputs from regions of the ventrolateral thalamus that lie within the termination zone of cerebellar efferents. The thalamus is turned upside down to indicate the match between its topography and that of the dentate nucleus. Abbreviations: 7b, 9L, 46d, Brodmann’s areas (see Figure 2); X, area X; ArS, arcuate sulcus; C, caudal; CM/Re, nucleus centrum medianum/nucleus reuniens; CS, central sulcus; D, dorsal; IPS, intraparietal sulcus; M, medial; M1, primary motor cortex: MD, nucleus medialis dorsalis; PMv, ventral premotor area; PS, principal sulcus; VLcc, caudal portion of the nucleus ventralis lateralis, pars caudalis; VPI, nucleus ventralis posterior inferior; VPLo, nucleus ventralis posterior lateralis, pars oralis. (From Dum and Strick, 2003.)

nurtured patterns of daily life such as eating, dressing, and bathing. This suggests that in AD patients, whereas body/motor schemas in the cerebral cortex are impaired, internal models in the cerebellum are preserved. Indeed, AD patients are characterized by the regional impairment of cerebral glucose metabolism in

16 • Motor Actions and Tool Use

187

neocortical association areas, including the posterior cingulate, temporoparietal, and frontal association cortices, whereas the primary visual and sensorimotor cortices, basal ganglia, and cerebellum are relatively well preserved (Herholz, 2003). From a modelistic viewpoint, it would be difficult for a single controller to skillfully perform a complex motor action with the aid of but one internal model. Probably, multiple controllers and associated internal models are involved simultaneously, and there is a mechanism for selecting an appropriate combination for a specific condition (Wolpert et al., 2003). The MOSAIC model, using combinations of forward- and inverse-model-based controls, can simulate such mechanisms (Chapter 15, Section 6).

16-5 Mirror Neurons The premotor cortex of monkeys contains “mirror neurons.” Their discharge patterns are quite similar when a particular object-directed action (such as grasping, tearing, manipulating, holding, and bringing an object to the mouth) is taken and when the same action is observed while being undertaken by another individual (another monkey or a human experimenter) (Rizzolatti and Craighero, 2004). Correspondingly, in an fMRI study on human subjects, motor actions performed using different effectors (mouth, hand, foot) induced a somatotopically organized activation of the premotor cortex (Buccino et al., 2001). Mirror neurons are located in the ventral premotor cortex (area F5; Color PlateXVI) of the monkey, where many other cells are also located that code for visually guided actions rather than observed actions. Mirror neurons do not respond to the object alone, but they discharge during a reach to an object placed out of sight, as long as the intention of the reach and grasp action is clear. Thus, these neurons are not driven simply by visual input, but rather, they respond to an object-directed action. Cerebral cortical areas associated with mirror neurons have been mapped in monkeys and humans, and neurons responding similarly to F5 mirror neurons were found in area PF of the posterior parietal cortex (Iacoboni et al., 1999) (Color Plate XVI). They coded more specifically for the kinesthetic and somatosensory components of an action. Another relevant area has been located in the superior temporal sulcus (Iacoboni et al., 2001; Carr et al., 2003). It was activated during both hand action observation and imitation even in the absence of direct vision of the imitator’s hand. Motor-related activity was greater during imitation than during control motor tasks. The observed actions and the copies of these actions as made by the imitator may interact in this area (Iacoboni et al., 2001).

188

the cerebellum

Many conceptual models have been proposed to reproduce the possible functions of mirror neurons, in particular, the ability to imitate an object-directed action (Miall, 2003; Oztop et al., 2006; Iriki, 2006). Here, we consider a simple plausible control system design for this function. Assume that F5 mirror neurons operate as part of the premotor cortical action controller for the controlled object nesting a corticospinal and a segmental motor system. Assume also that the PF area represents kinesthetic and somatosensory components of an action (see previous discussion) as a body schema. Assume further that the premotor cortex receives feedback information about self-generated actions via a somesthetic pathway, and that it also receives visual information about an action performed by another individual. The most crucial assumption is then that the preceding two pathways leading to F5 mirror neurons are commonly mediated by the superior temporal sulcus region (Figure 52).

Figure 52 A possible control system structure for mirror neurons.

Mirror neurons are assumed to be a component of premotor cortical neurons, which serve as a controller of motor actions. To behave as mirror neurons, they should be equipped with a comparator of the self-generated and observed individual’s movements, which is likely located in the superior temporal sulcus (STS). Other abbreviations: PF, parietal cortex defined in Color Plate XVI; t, broken arrow indicating close connections between PF and STS.

In this postulated system, a group of F5 mirror neurons serves as an action controller, whereas PF mirror neurons help these F5 mirror neurons control skillfully an action by providing a body schema. During repeated trials of a motor action, a cerebellar forward model copies the body schema from the PF region so

16 • Motor Actions and Tool Use

189

that the skillful motor action can be performed more automatically. F5 mirror neurons are driven to discharge by instruction signals from a higher center, and they are also driven by feedback from self-generated actions via the superior temporal sulcus area. This area also receives information about the observed “other-individual’s” motor actions (Figure 52). The superior temporal sulcus area is anatomically interconnected with the PF area, and together, they may share the body schema for generating information about performed actions. Of essential importance is the postulated neuronal mechanism of the superior temporal sulcus area, where information about an observed action replaces that of a performed action, provided the former imitates the latter. The major role of mirror neurons is generally considered to be in recognizing the actions of others (action recognition). Recently, however, a human fMRI study showed that premotor mirror neurons exhibited greater activity when an action was observed in the presence of a specific context, such as the background scenery. This suggested that such neurons were interpreting the intention of the observed action (intention understanding) (Iacobini et al., 2005). Mirror neurons have also been considered to be responsible for imitation learning—that is, learning to perform an action by seeing it done. However, because monkeys rarely learn by imitation, it does not seem to be a primary function of mirror neurons, which are commonly located in the cerebral cortex of humans and monkeys (Rizzolatti and Craigher, 2004). Nevertheless, fMRI activity observed in both humans and monkeys during their responding to an observed action was greater than that during responding to a non-movement-relative scene. The greater imitation-activated activity in the cerebral cortex was observed in the superior temporal sulcus area (Iacoboni et al., 1999, 2001). In summary, the data to this point suggest that the imitation capacity involves use of the mirror neuron system.

16-6 Self-Monitoring of Motor Actions The idea of sensory cancellation (Chapter 15) also applies to problems of motor actions. Schizophrenic patients often exhibit the delusion of alien control, which makes them misattribute self-generated motor actions to externally generated motor actions by others. Blakemore et al. (2003) explained this phenomenon as arising from the inefficient or erroneous operation of a forward model. That is, if an internal feedback through a forward model does not cancel the sensations induced by self-generated actions, then the latter will reach the level of consciousness in a fashion similar to that induced by external stimuli generated by others. In

190

the cerebellum

this situation, the patient cannot distinguish self-generated actions from those generated externally by others. The alien control seen in subjects under hypnosis may likewise be explained as arising from abnormal operation of a forward model. A brain imaging study revealed that the active movement attributed to another individual resulted in significantly higher activations in both the parietal cortex and the cerebellum than an identical movement correctly attributed to the self (Blakemore et al., 2003). Blakemore and Sirigu (2003) interpreted these observations as indicating that both the parietal cortex and the cerebellum provide internal models of a different nature. They speculated that whereas a cerebellar internal model makes rapid predictions about the sensory consequences of a self-generated movement commanded and controlled automatically at a low level of the brain, a model in the parietal cortex addresses more cognitive aspects of the same movement and this processing reaches the level of conscious awareness. This view is close to the previously postulated learning of motor actions except that here we assume that the parietal cortex first forms a series of body schemata, which are then copied by the cerebellum. Another phenomenon suggesting the involvement of the parietal cortex in motor action control is the sensation of a phantom limb—that is, an amputated limb being still present and sensed on occasion to be in motion (Frith et al., 2000). This sensation is represented in the parietal cortex (Ramachandran and Hirsten, 1998), and it may arise from motor schemata for the limb that are retained after loss of the limb. An internal model in the cerebellum may also be retained, but it does not contribute to the phantom limb sensation because cerebellar events do not reach self-awareness.

16-7 Tool Use Various types of sport and dancing demonstrate the ability of some humans to achieve superb motor control. In one type of motor action, the controlled objects are mainly body parts (e.g., in swimming, running, gymnastics). In another type, various tools are used (e.g., balls, bat, gloves). Throughout training, tools become incorporated into the motor control system as if they are newly added parts of the body. Such extended motor capability requires changes in the neural representation of the body schema, that is, updated maps of body shape and posture incorporated into the cerebral cortex (Maravita and Iriki, 2004). Lesion studies in right-handed patients have revealed that the left cerebral hemisphere is specialized for representing tool-use skills, but there is evidence that the left cerebral

16 • Motor Actions and Tool Use

191

hemisphere is also specialized for representing these skills in left-handed individuals (Johnson-Frey, 2004). Together, these findings indicate that behaviors associated with complex tool use arise from functionally specialized networks involving temporal, parietal, and frontal areas within the left cerebral hemisphere. I believe that the resultant body schema will then be copied by an internal model in the cerebellum (Chapter 15). Semantic knowledge about familiar tools and their uses is stored also in the cerebral hemisphere, but it is uncertain whether or not this cognitive memory is relevant to the cerebellum. Neural events underlying tool use have been studied in monkeys and humans. In one experiment, a monkey was trained to manipulate a stick to shift a cursor on a screen from a starting box to a target box. When the target box was suddenly repositioned, the monkey had to modify the ongoing movement to place the cursor within an also repositioned target box. This induced a significant increase in the complex-spike discharges of Purkinje cells located in the ipsilateral hemisphere and intermediate zone of the cerebellum. This finding was interpreted to mean that climbing fiber signals occurred when the motor state changed and/or when errors were occurring in the motor performance (Wang et al., 1987). In another study, monkeys performed a multijoint arm-reaching task, in which movement direction and distance were varied systematically. The complex-spike activity of relevant Purkinje cells was often correlated significantly with movement distance, usually in one direction. It was therefore suggested that the complex-spike discharge of Purkinje cells was spatially tuned and related strongly to the movement’s kinematics (Fu et al., 1997b). In a study on humans, the subject manipulated a computer mouse to follow a moving small square target with a small cross-hair cursor on a screen. During continuous following, the position of the cursor was suddenly shifted by rotating it 120° around the center of the screen to provide a novel mouse condition. In the first training session, large regions of the cerebellum were activated significantly, but the extent of activation decreased during repeated test trials in parallel with a reduction in the rate of tracking errors. Eventually, certain subregions (near the posterior superior fissure) continued to be activated. This remaining activity may have represented an internal model being formed during the repeated test trials. This would imply a novel relationship between the movements of the cursor and the mouse. Because the subject could switch quickly between the old and the novel mouse, it seemed likely that there must be a neural mechanism for selecting an appropriate model out of many internal models of the various mouse situations (Imamizu et al., 2000).

192

the cerebellum

The final study mentioned here was one in which monkeys were trained to use a rake to obtain food rewards. All three monkeys tested (monkeys E, N, and F) learned to use the rake within the 14-day training period. Monkeys E and F used their right hand to rake and their left hand to retrieve the food reward. Monkey N used both hands to both rake and retrieve the food. They were examined repeatedly in a 4-T scanner for a six-week period: two weeks of habituation to the task, two weeks of intensive daily training, and a two-week post-training period. As task performance improved, structural MRI scans revealed significant increases in the signal intensity of gray matter in the cerebral cortex. These increases were the most significant in the right superior temporal sulcus, right second somatosensory area, and right intraparietal sulcus, with less significant increases in these same regions of the left hemisphere. Interestingly, signal increases were also observed in the white matter of the bilateral cerebellar hemispheres in lobule V (Quallo et al., 2009). An observed enhancement of structural MRI signals indicated an increase in the volume of gray and/or white matter. What really underlies the just-described cerebellar enhancement is still unknown, but it is in keeping with long-lasting plasticity in cerebellar circuits that play a role in cerebellar learning.

16-8 Summary Neuronal mechanisms for motor actions have been studied by unit recording of Purkinje cell discharge in monkeys and brain imaging in monkeys and humans. Mirror neurons in the premotor cortex appear to be a part of the neuronal systems controlling motor actions. To advance further, it will be particularly important to find neural substrates of the psychological concepts of body schema and motor schema. Complex motor actions such as an actor’s performance on the stage typically integrate movements with perceptual and conceptual behavior. This indicates that neural mechanisms for motor actions must involve both motor and cognitive domains, as discussed in the next chapter.

17 Cognitive Functions

17-1 Introduction In Chapter 16, “Motor Actions and Tool Use,” we examined the roles of the cerebellum in the control of complex motor actions in primates. We are now ready to extend our consideration to an even higher level of control; that for mental activity in humans. There is an increasing body of evidence (both experimental and theoretical) that supports the presumed unique roles of the cerebellum in our thought processes in which we manipulate ideas and concepts instead of moving body parts.

17-2 Neural Systems for Thought Here, we conceive a neural system for mental activities as an analogy of that for voluntary motor control and action control (Chapters 13–16). First, the prefrontal cortex is postulated to be the controller. The purpose is to orchestrate thought and action in accordance with (1) internal goals (Miller and Cohen, 2001), (2) the executive function required for the conscious control of thought and action (Happaney et al., 2004), and (3) the supervisory activating system for working memory (see following text). Second, controlled objects are represented in the temporoparietal cortex, which will be defined later in terms of mental models and schemata. Third, the neocerebellum provides cerebellar internal models (forward and inverse) of the controlled objects represented in the temporoparietal cortex. Thus, the prefrontal cortex, temporoparietal cortex, and neocerebellum conjointly constitute a control system for mental activities. In support of these views, the co-activation of these three structures corresponding to the controller, controlled object, and internal model, respectively, has been demonstrated in an increasing number of cases. For example, in a verb-to-noun conversion task, the prefrontal cortex and Wernicke’s area in the temporoparietal cortex (both of the left side) and the cerebellum (on the right side) have been shown to be co-activated (Fiez et al., 1996). 193

194

the cerebellum

Kelly and Strick (2003) provided anatomical substrates for involvement of the cerebellum in cognitive function using viral tracing techniques in nonhuman primates (for review, see Ramnani, 2006; Strick et al., 2009). They revealed a loop connection between the prefrontal cortex (area 46) and the cerebellar hemispheres (crus I and crus II of the ansiform lobule) (Figure 53). In human subjects, Krienen and Buckner (2009) used functional connectivity MRI (fcMRI) to identify four topographically distinct frontocerebellar circuits that targeted the (1) motor cortex, (2) dorsolateral prefrontal cortex, (3) medial prefrontal cortex, and (4) anterior prefrontal cortex. A direct comparison between the right- versus left-side frontal regions revealed contralateral lateralization in the cerebellum for each of the segregated circuits. Overall, the extent of the cerebellum associated with the prefrontal cortex included a large portion of the posterior hemispheres associated with non-motor functions.

Figure 53

Input-output organization of the cerebellar loops with cortical area 46.

(Left) Distribution of Purkinje cells (small dots) that project to area 46. (Right) Distribution of granule cells (fine lines) that receive input from area 46. The shaded areas on the flattened surface maps (diagrams on the left in each panel) are unfolded on the right side of each panel to show the distribution of labeled neurons in the relevant cerebellar cortical fissures. Scale bars, 15 millimeters, (ignore the numbers on the X axes, which indicate the positions of two sections illustrated in another figure). (From Kelly and Strick, 2003.)

17 • Cognitive Functions

195

Anatomical connections have also been shown to exist between the temporoparietal cortex and the cerebellum. In nonhuman primates, the posterior parietal areas of the parietal cortex were shown to receive inputs from the ventral portions of the dentate nucleus in the cerebellar hemisphere (Dum and Strick, 2003) and the anterior intraparietal area of the posterior parietal cortex received inputs from a broad area of the dentate nucleus (Clower et al., 2005). In a human fcMRI study, connections were shown between the inferior temporal cortex and regions in the cerebellum (Krienen and Buchner, 2009). Because the inferior temporal cortex receives few anatomical connections from the pons, these latter connections must be mediated indirectly.

17-3 Cerebral Cortical Model for Thoughts Sensory information about a stimulus is first represented in the primary sensory area of the cerebral cortex. Then, while being transferred through sensory association cortices, its mental representation is formed and eventually reaches the level of conscious awareness. For example, “binocular rivalry” occurs in the anterior part of the infratemporal cortex where neurons can behave in accordance with conscious experiences, but not with what is actually sensed by the retina (Sheinberg and Logothetis, 1997). Wilder G. Penfield (1891–1976) demonstrated that electrical stimulation of certain areas of the human cerebral cortex invokes various mental representations, even an episode that occurred in the past (Penfield and Rasmussen, 1950; Penfield and Perot, 1963). A mental representation of complex integrated information may constitute a “mental model” that Craik (1943) defined as a psychological substrate of a mental representation of both real and imaginary situations. A mental model is a smallscale version of reality, which the mind constructs and uses to reason, explain, and anticipate a future event. In other words, mental models are representations of images, concepts, and ideas. No information is yet available, however, on just how mental models are expressed as specific activity in a cortical circuit. The concept of “schema” introduced by Jean Piaget (1951) is also close to that of mental models (recall Chapter 1, “Neuronal Circuitry: The Key to Unlocking the Brain”). His idea was that a child develops a schema that helps in interpreting and understanding the world. It includes both a category of knowledge and the process of obtaining that knowledge. An assembly of mental models constitutes an internal world that each individual’s conscious awareness can access. It seems likely that a mental model is encoded in a neuronal circuit of the cerebral association cortex. Neurophysiological mapping in monkeys revealed activities related to the perception of various objects (e.g., dolls and artificial fruits) in the

196

the cerebellum

temporoparietal association areas (Tanaka, 1996; Tsunoda et al., 2001). Neuroimaging revealed that neuronal activity associated with motor imagery required to simulate a movement involves activity occurring simultaneously in the premotor cortex, parietal association area, and cerebellum (Jeannerod and Frank, 1999). Such activity in the parietal cortex may represent a mental model of the movement. However, a more complex way of representing perception and behavior has been suggested (Hesslow, 2002). The brain may simulate motor behavior by activating motor structures, as during an overt action but suppressing its actual execution. Similarly, the brain may simulate perception by the internal activation of the cerebral sensory cortex, as during the normal perception of external stimuli. These considerations suggest that mental models elicit the perceptual simulation of the normal consequences of overt actions. Here, we consider that Craik’s mental model and Piaget’s schema are useful explanatory concepts, giving a psychological counterpart of the cerebral cortical models we assume to operate in the control system mechanisms for cognitive functions.

17-4 Cerebellar Internal Model for Thought We assume that during repeated thoughts the cerebellum forms an internal model of the cerebral cortical mental model of these thoughts. A cerebellar forward and inverse model may copy essential properties of such a mental model. However, in contrast to a movement in the physical domain that can be computationally represented in an internal model, it remains unknown how a mental model having an abstract conceptual nature is represented in a neuronal circuit of the cerebral cortex and is copied by a circuit in the cerebellum. Mechanisms have been discussed for forming a computable mental model residing in the cerebral cortex (JohnsonRaird, 1983). Indeed, an episodic-like (event) memory or semantic-like (fact) memory is encoded in a neuronal circuit of the medial temporal lobe (see Miyashita, 2004). After it is encoded in a cortical network, it may be transferred by error learning to a network of the cerebellar cortex. To determine just how this is done is a great challenge in contemporary computational neuroscience. Wolpert et al. (2003) explored the computational parallels between the processes that occur in motor control and in social interaction. They examined how models of motor control, such as the HMOSAIC (consisting of several layers of MOSAIC; Haruno et al., 2003), could be used for action observation, imitation, social interaction, and Theory of Mind (see Section 7). They suggested that models of motor control provide an efficient mechanism for performing the computations needed in social interaction.

17 • Cognitive Functions

197

17-5 Explicit and Implicit Thoughts In accordance with our daily life experiences, we assume that thought is performed through two parallel processes—one explicit and the other implicit. When we try to solve a novel problem that requires mentation, we initially devote strenuous conscious effort to the issue and continue to think about its solution. Later, we may recall the problem from time to time, but we are otherwise unaware that we are still trying to solve the problem. Yet, the thought of how this is to be done appears to proceed implicitly because the sought solution may be derived suddenly without obvious conscious effort to obtain it. We further assume that the implicit process for the manipulation of our thoughts proceeds in the cerebellum, whereas the explicit process is undertaken in the cerebral cortex. Various forms of motor learning involving the cerebellum are indeed executed implicitly. Whereas electric stimulation of the cerebral cortex evokes conscious experiences in humans (Penfield and Perot, 1963), that of the cerebellar cortex causes no more than increased alertness and reduced depression and anxiety (Riklan et al., 1976). TMS of the cerebellum does not evoke conscious experience (Koch et al., 2006). These observations support the assumption that thought processes in the cerebellum do not reach the level of conscious awareness. We do not assume, however, that all events processed in the cerebral cortex reach conscious awareness because brain imaging has shown repeatedly that self-unrecognized activities take place in many cerebral cortical areas. A structure for the thought system may consist of several so-far-proposed neural systems. For example, the working memory system maintains and stores information in the short term. Its core component is the supervisory activating system that is an attentionally limited controller (Norman and Schallice, 1986). This core forms the central executive of the working memory system that is located in the dorsolateral prefrontal cortex (Smith and Jonides, 1997; Stuss and Knight, 2002; Baddeley, 2003). It has been proposed that when a stimulus is presented in the realm of thoughts and ideas, it is maintained in the working memory, where it is compared with mental models stored in the temporoparietal cortex. Comparison is also made to cerebellar internal models, which have been copied from mental models and stored in the cerebellum. If a received stimulus matches a mental model or an internal model, the stimulus is considered at the level of conscious awareness to be familiar and the thought process is completed. If the stimulus is truly novel, however, existing models will not recognize the stimulus. Error signals are then generated that (1) suppress at the level of conscious awareness acceptance of the idea that the mental model is familiar, (2) activate the attentional system, which in turn

198

the cerebellum

activates the working memory system, which will then strengthen the search for stored mental and internal models, and (3) modify internal models in the cerebellum if errors continue to occur. Eventually, the stimulus will find a matching counterpart in the pools of mental models and internal models. Then the suppression imposed in the first step will be removed in a manner analogous to the SchmajukLam-Gray (SLG) model that simulated the mechanism of classic conditioning (Schmajuk et al., 1996). The final process is an “aha” at the level of conscious awareness; that is, expression of the fact that a successful conversion has taken place. In other words, the original stimulus is no longer novel. The preceding processes and mechanisms may well be the basis for creativity and innovation (Ito, 2007; Vandervert et al., 2007).

17-6 Cognitive Activity in the Cerebellum Numerous studies have now shown cognitive activity in the cerebellum. For example, in an fMRI study, when the subject unexpectedly received painful heat stimulation, the hippocampus, superior frontal gyrus, superior parietal gyrus, and an extreme part of lateral part of the cerebellum were activated conjointly (Ploghaus et al., 2000) (Figure 54, prediction error). This observation is consistent with previous considerations about explicit and implicit thought. In another fMRI study, subjects pressed a button at a comfortable self-determined pace versus during an attention task, which tested the ability to selectively attend and respond to a variety of visual targets. The cerebellum exhibited significant activation during the attention versus self-generated tasks (Allen and Courchesne, 2003) (Figure 59, attention). The activation of a thought system as postulated above seems also to occur in various language tasks. Neuroimaging and lesion studies in humans suggest that the right posterolateral cerebellar hemisphere is involved in certain aspects of language performance. In a task that required changing nouns into verbs, learning was impaired by a large cerebellar infarction (Fiez et al., 1992). In normal subjects, this task co-activated the left prefrontal cortex and the left parietal cortex together with the right posterolateral portion of the cerebellum (Fiez et al., 1996). A similar coactivation of the left prefrontal cortex, left dorsolateral cortex, and right cerebellum was induced during silent performance of a verbal fluency task. The subjects were instructed to think silently about as many words that they knew to begin with a specified letter (Schlosser et al., 1998). In another study, the tasks used included antonym generation, noun (category member) generation, verb selection, and a lexical decision. The subjects compared were a normal (control) group and groups

17 • Cognitive Functions

199

with focal right- or left-side posterolateral cerebellar lesions. The results showed that the subjects with a right cerebellar lesion were impaired only in their performance of the antonym generation task. The deficit was not due solely to deficits in “mental movement” coupled with a verb. Rather, the faulty internal generation of a word seemed to be a key factor for eliciting the deficit (Gebhart et al., 2002). The Wisconsin card-sorting test has been used to examine functions of the prefrontal cortex. Participants were given cards that could be sorted by color, shape, or name, and their mental task was to deduce the correct sorting criterion. After several consecutive correct responses, the correct sorting criterion was changed without warning. PET imaging revealed that the test co-activated the left (or bilateral) dorsolateral prefrontal cortex, bilateral inferior parietal cortex, left superior occipital gyrus, and the left neocerebellum (Nagahama et al., 1996).

Figure 54 Mental activities in the cerebellum.

Neuroimaging studies have provided evidence of a role for the cerebellum in various mental activities. The figure shows a coronal section of a human cerebellum, on which circles indicate the sites of observed neuronal activity. The tasks that elicited these activities include those for attention, future vision, and verbal working memory. Activity observed when a subject received an unexpected painful heat stimulus on the back of the left hand is also indicated as responses to a prediction error. Abbreviations: A–B, subdivisions of hemispheric areas; H II–VIII, areas of the cerebellar hemispheres; Inf, inferior; Sup, superior. (From Ito, 2008.)

200

the cerebellum

In testing verbal working memory, participants were asked to remember six letters or one letter presented visually. These letters were presented visually only once to the subjects for 1.5 seconds. After a 5-second delay, one lowercase consonant letter was presented as a probe stimulus. For the high-load condition, this probe matched one of the six letters in the array for half of the trials. For the low-load condition, the probe matched the single letter in parentheses in half of the trials. One trial, 8.0 seconds in duration, was repeated four times in each block, and 12 such blocks were presented for 6.4 minutes. The high-load test, but not the low-load one, increased fMRI activity in certain regions of the bilateral superior cerebellar hemispheres (left superior HVIIA and right HVI), the right cerebellar hemisphere (HVIIB), and portions of the posterior vermis (VI and superior VIIA) (Desmond et al., 1997) (Figure 54). In another study, one task required producing as many words as possible (excluding proper names), beginning with a specific letter such as F, A, and S (letter task), and another task required producing as many different words as possible belonging to the semantic categories for “birds” and “furniture” (semantic task). A 60-second period was granted for each letter and semantic task. It was found that cerebellar patients were deficient in their phonemic rule performance but quite capable in semantic rule performance (Leggio et al., 2000). Another interesting observation was an fMRI study in which the left lateral premotor cortex, left precuneus (part of the superior parietal lobule hidden in the medial longitudinal fissure), and right posterior cerebellum were shown to be more active while envisioning the future than while recollecting the past (Figure 54). These regions were similar to those that became more active while imagining (simulating) bodily movements. Another set of brain regions including the bilateral posterior cingulate, bilateral parahippocampal gyrus, and left occipital cortex displayed similar activity during both future and past mentation tasks. This similarity was attributed to the reactivation of previously experienced visual-spatial contexts into which the subjects placed their future scenarios (Szpunar et al., 2007). Chess players require the capability to predict what occurs consequent to a lengthy alternating series of moves made by them and their opponents. Neuroimaging studies of novice chess players revealed bilateral activation in the premotor area, parietal cortex, and occipital lobe, and unilateral activation in the left cerebellar hemisphere (Atherton et al., 2003). Similar results were reported for the Chinese board game “Go” (Chen et al., 2003). Quite recently, we compared neuroimaging data for two groups of Japanese chess (Shogi) players: one group of well-trained professional players and the other less-well-trained amateur players (Fujii et al., 2009). We imposed quick checkmate tasks on these players. They were instructed to pay attention to a chess board showing a stage of a game, and

17 • Cognitive Functions

201

their task was to determine whether a checkmate could or could not be made. The number of correct answers was high (near 100%) for professional players and much lower (~50%) for amateur players. fMRI activity during these tasks are now under study.

17-7 Mental Disorders Associated with Cerebellar Dysfunction The cognitive sequela that follows damage to the dorsolateral prefrontal cortex is similar to that following damage of the neocerebellum (Diamond, 2000). A patient with paraneoplastic cerebellar degeneration was shown to exhibit selective frontal-executive disturbance, psychomotor slowing, and affective change, despite the finding that there was no apparent extracerebellar involvement (Collinson et al., 2006). These findings underline the importance of the cerebellum in regulating cognitive function. Cerebellar mutism is another indication of the cerebellum’s role in language acquisition and performance. This syndrome typically affects children and, in rare cases, young adults. They become mute one to two days after the surgical removal of a tumor in the posterior cranial fossa. The surgery requires a certain degree of damage to the cerebellar hemispheres. The syndrome, which persists for one to four months, is not accompanied by disturbances of consciousness and language comprehension. It has been suggested that the obligatory damage to the cerebellar hemispheres is the most important factor in development of the mutism (Janssen et al., 1998). In a variety of brain diseases that cause mental disorders, patients often display motor disorders caused by cerebellar dysfunction. This occurs when pathological changes occur in cerebellar areas devoted to movement and other areas implicated in mental activities. One study showed that over 90% of autistic patients examined at autopsy had well-defined anatomical abnormalities in the cerebellum (Carper and Courchesne, 2000). Other studies on autism showed that the size of Purkinje cells was reduced in these subjects (Fatami et al., 2002) and nicotinic receptor abnormalities have also been observed (Lee et al., 2002). A genetic deficit in autism has been found (Sadakata et al., 2007a,b) and hyperserotonemia was recognized in autistic individuals and their relatives (Lee et al., 2002). Allen and Courchesne (2003) examined fMRI activation within anatomically defined cerebellar regions in autistic patients versus matched control healthy subjects. For a motor task, subjects pressed a button at a comfortable pace, and CNS activation was compared to the rest condition. For an attention task, visual stimuli were presented one at a time on a screen, and subjects pressed a button each time a target appeared. Activation was again compared to the visual rest condition. While performing these

202

the cerebellum

tasks, autistic individuals showed significantly greater cerebellar motor activation and significantly less cerebellar attention activation. Autistic patients are characterized by the defective “Theory of Mind,” that is, loss of the ability to understand the thoughts and motivational processes of other individuals, with consequent actions that are socially aberrant (Deuel, 2002). This deficit may involve an autistic subject’s inability to guess what another person believes or desires because the subject cannot simulate an internal model of the other person. Defective central coherence is manifested as an inability to mold diverse and detailed internal information into relevant higher-order concepts that guide behavior over the long term. This again may reflect the inability of forming an internal model that embodies highorder concepts. Developmental dyslexia is a syndrome that includes difficulty in separating words into discrete segments (such as phonemes). In turn, this leads to difficulty in learning spelling-sound correspondences. It has been suggested that the underlying deficit is the inability of the brain to filter irrelevant data such as perceptual noise (Sperling et al., 2005). The focus of this dysfunction has been placed in the frontal lobe and the temporal lobe (Galaburda et al., 1985), a magnocellular component of the visual pathway (Stein and Walsh, 1997), and the cerebellum (Nicolson et al., 2002). The cerebellar theory of dyslexia originated from the observation that dyslexic children perform less well than control children in certain motor (including balancing) tasks. One study showed that when the speed and accuracy of pointing were combined, dyslexic participants performed less effectively than control subjects. Furthermore, there was a significant relationship between performance of the pointing task and literacy skills: that is, a regression analysis showed that the error and speed of pointing contributed significantly to the variance in literacy skill (Stoodley et al., 2006). Schizophrenic patients were shown to be significantly more impaired than normal control subjects in adapting to prism distortion. They also had significantly greater difficulties in the reorientation that followed removal of the prisms (Bigelow et al., 2006). It seems likely that the cerebellar disorders of these patients affected both their voluntary movements (tested with prism-adaptation tests) and their mental processes, the latter being manifested as symptoms of schizophrenia. Hallucination and “passivity” are unique symptoms of schizophrenia, which can be explained as arising from the failure of a mental model and its copy in a cerebellar internal model to provide appropriate internal feedback to the prefrontal cortex (Blakemore et al., 2000). Hallucination includes the hearing of unreal spoken speech, which may result in the patient not recognizing that the unreal speech was produced internally. Passivity is a patient’s feeling that her/his will is being replaced by that of some other force or agent. This feeling might arise as a result of a lack of

17 • Cognitive Functions

203

awareness of intended actions, which might be due to impairments of mental models and their copies in cerebellar internal models. This would result in a loss of prediction of the possible outcome of an intended action. Hence, an aberrant self-monitoring mechanism involving the cerebellum, as discussed in Chapter 16 in connection with alien control, appears to apply also to mechanisms underlying hallucination and passivity in schizophrenia.

17-8 Summary Neuroimaging data suggesting cognitive involvement of the cerebellum ever increase, but one should be aware of a caution such that the observed activities might be contaminated by eye movements that often accompany cognitive tasks (Glickstein et al., 2009). It is also important to realize that the internal model-based control of mental activities presented above is still a conceptual hypothesis based on psychological aspects of Craik’s mental model and Piaget’s schema. Neuroimaging has made plausible that (1) activation of certain regions of the temporoparietal cortex is the neural substrate of a mental model or schema, and (2) activation of the cerebellum is an indication of the involvement of cerebellar internal models. This hypothesis is also supported to some degree by other experimental and clinicopathological data. It must be admitted, however, that further testing of the hypothesis is challenged by difficulties arising from current technical limitations in human experimentation and the lack of relevant computational support.

18 Concluding Thoughts

18-1 What Is the Cerebellum in a Nutshell? Neuronal circuits in the cerebellum have now been decomposed to the extent that their various neuronal components are defined and synaptic plasticity (LTD/LTP) is as memory elements in neuronal networks; that is, at least for the early phase of memory processes (Chapters 3–7). Detailed signal flow charts underlying synaptic plasticity have been revealed, as have been some genetic and pharmacological means to manipulate this plasticity. Reconstruction of the cerebellar machine has also advanced markedly (Chapter 9, “Network Models”). Three-layered neuronal networks like a simple perceptron, which is based on the design of neuronal circuits, have been proposed as a model of the cerebellum that is capable of spatial pattern recognition. In addition, adaptive filter models are capable of temporal pattern recognition (Chapter 9). Real cerebellar networks integrate these two capabilities to major extent, and the networks have been simulated using liquid state machine models. Circuits have also been modeled for error learning during repeated attempts to execute a task. Experimentally-based knowledge about neuronal networks in the cerebellum has also become increasingly more detailed. For example, Lugaro cells and small inhibitory neurons have been added to the neuronal wiring diagram of the cerebellar cortical network, albeit their functional roles are still unclear. Also, many different neuropeptides and amines conveyed by beaded fibers to the cerebellum from the hypothalamus have been identified, and their functional roles are now being considered in terms of neuromodulation. Recently, it has become clear that the cerebellar and vestibular nuclei act as a memory site and as such, are a complement to the memory role of the cerebellar cortex. It will be intriguing to see how this ever-emerging new knowledge will modify and expand our present models of cerebellar function.

204

18 • Concluding Thoughts

205

A microcomplex has been conceived as a modular unit of cerebellar neuronal circuits. It incorporates a microzone of the cerebellar cortex and a small group of cerebellar and/or vestibular nuclear neurons that are attached to a small group of inferior olive (IO) neurons and that of parvocellular red nuclear neurons (Chapter 9). Such a microcomplex is equipped with two types of memory (cortical and nuclear), which have complementary roles in learning. Microcomplex modules in the cerebellum seem capable of mimicking the input-output or output-input relationship of other neuronal systems in the CNS by use of an error learning mechanism. Modeling has suggested that discrepancies between the outputs of a microcomplex and the circuit it is copying can be obviated progressively throughout repeated trials. This is thought to be accomplished in main part by a change in the properties of the microcomplex brought on by conjunctive LTD in its Purkinje cells and associated LTP/LTD at mossy fiber-nuclear neuron synapses. In this way, a cerebellar internal model can be expected to simulate a controlled object or its inverse. This internal model concept is now being tested by unit recording from Purkinje cells (and hopefully also from nuclear neurons) during various forms of voluntary movement, and it is also being applied successfully in the field of robotics. Its application to complex motor actions and cognitive functions is still a matter of conceptual modeling, but it certainly points the way to future research.

18-2 Seven Questions Despite the above progress, a number of crucial questions, specific and general, and conceptual and technical, remain unanswered at all levels of analysis: that is, at the molecular, cellular, circuitry, and behavioral levels. The major purpose of this book is to collect so-far-uncovered facts and answered and unanswered questions in research on the cerebellum and then assort them according to systems control principles. Now, at the end of this exercise, it behooves me to ask seven questions, the answers to which are my opinion of what should guide the next phase of research on the cerebellum. 1. Conjunctive LTD as synaptic plasticity. The signal transduction properties of this synaptic plasticity have been analyzed extensively. Nevertheless, there could well be more as-yet-unknown molecules and processes to be discovered and their functional role determined. For example, we have shown recently that prostaglandin D2 and E2 play a crucial role at the stage of LTD induction when PKCα phosphorylates AMPA receptors. These receptors are then severed from the cytoskeleton and removed from the stable synaptic pool of AMPA receptors (Figure 20). However, just how prostaglandins D2/E2 act on AMPA receptors is not yet known (Le et al., 2010).

206

the cerebellum

The most intriguing question about conjunctive LTD is the mechanism(s) by which its manifestation leads to more persistent memory-like processes. Synapses undergoing conjunctive LTD do not cause an immediate change in the shape of the dendritic spines on Purkinje cells, but the functional state of these synapses seems to be transformed sometimes into silence or even further, the synapses disappear (Chapter 8, “Multiplicity and Persistency of Synaptic Plasticity,” Section 6). An open question is just how LTD links to such synapse liability. A relevant enduring issue is the length of time that LTD can be maintained (Chapter 7, “Conjunctive Long-Term Depression (LTD)”). Studies on the VOR and OKR have shown that short-term adaptation recovers in 24 hours. This duration seems to be unduly short for a memory formation in the usual sense. Nonetheless, there has been a tendency to generalize this duration to conjunctive LTD. It seems possible, however, that this 24-hour value for VOR/OKR adaptation is due, at least in part, to a relearning process, the goal of which is to reach the pre-adaptation level (e.g., like extinction of eye-blink conditioning). Climbing fiber discharge, which indeed occurs while animals are resting in the dark, could drive such a relearning mechanism (Chapter 10, “Ocular Reflexes,” Section 3). If such complications are avoided in some way, the VOR/OKR may continue to be a useful test paradigm for learning and memory mechanisms of the cerebellum. 2. Conjunctive LTD versus motor learning. A popular view is that the appearance of conjunctive LTD implies the formation of a memory trace, which is acquired during motor learning. This idea is usually based on the experimental finding of a coincident failure of conjunctive LTD and motor learning in response to genetic and/or pharmacological manipulations. LTD is usually observed in intracellular recording undertaken in tissue culture and slice preparations, or in vivo but under general anesthesia, and using synchronize electric shock stimuli. In contrast, motor learning is usually tested for by observing changes in an animal’s behavior under natural stimulating conditions. Welsh et al. (2005) challenged this experimental dichotomy in coincidence testing. In a study of anesthetized in vivo rat preparations, they confirmed that LTD detected in extracellular field recordings was blocked when a pharmacological agent (T-588, a blocker of calcium release from intracellular stores, Kimura et al., 2005) was continuously infused intravenously into rats. They also reported that awake rats administered perorally with T-588 exhibited normal eye blink conditioning even though the concentration of T-588 in the brain reached the level that was attained by intravenous infusion. This might suggest that LTD was not needed for motor learning but rather, its possible role was to protect the animal from excitotoxicity. In the above study,

18 • Concluding Thoughts

207

however, there was a discrepancy between the experimental conditions for the manifestation of LTD and motor learning. The presence of the former was shown in anesthetized rats while they were receiving a continuous intravenous infusion of T-588, with the LTD evoked by electric pulse stimuli. In contrast the manifestation of motor learning was shown in awake rats, who received orally administered T-588 during their reactions to natural stimuli. In the latter case, there was no definitive proof that LTD was really blocked when the animals were exhibiting eye blink conditioning. Similar reservations apply to a recent report that three mutant mice with an internalization of AMPA receptors lacked conjunctive LTD, even though VOR adaptation and eye blink conditioning occurred in a seemingly normal fashion (Schonewille et al., 2011). The above cited observations may compel us to reconsider the present LTDbased hypothesis of motor learning and attempt to formulate a new hypothesis. Before doing so, however, it deserves emphasis that none of the present studies have shown the lack or presence of LTD when the tested animals were undergoing motor learning. An unequivocal test would be to monitor LTD in behaving animals. To date, such an inference has been made (Gilbert and Thach, 1977; Ojakangas and Ebner, 1992; Medina and Lisberger, 2008) but not with the definitive observation of conjunctive LTD appearing during the learning process. 3. Function of the IO. Climbing fibers originating from the IO are a unique structure of the cerebellum and have been assigned the role of supervising teacher for the learning process in the cerebellar cortex (Chapter 3, “The Cerebellum as a Neuronal Machine”). The intricate network structure interconnected via electrical synapses in the IO generates highly regular rhythmic discharge under certain conditions, but stochastic, low-rate discharge under other conditions. This appears to imply a subtle mechanism for IO neurons to generate error signals consistent with the conditions to which the animals are subjected (Chapter 6, “Pre- and Post-Cerebellar Cortex Neurons,” Section 3). However, a contrasting interpretation has been put forward that the olivocerebellar system is a sort of clock for generating temporal patterns (Yarom and Cohen, 2002). These authors speculated that the olivocerebellar system generates a large repertoire of temporal patterns. When a specific temporal pattern is required, a group of IO neurons are assembled by pausing the activity in the cerebellar nuclear neurons and thereby removing the inhibition from appropriate gap junctions between IO neurons. The so-disinhibited IO ensemble begins oscillatory discharge, and thereby generates a temporal pattern that is uniquely suited to the task at hand. This hypothesis needs further testing, however.

208

the cerebellum

4. Nuclear memory. Although a memory process is now known to be formed in vestibular and interpositus nuclear neurons, such a finding has not yet been reported for dentate neurons, which receive collaterals of pontocerebellar mossy fibers. Because these collaterals are so meager (Chapter 6), however, it is far from certain that an equally effective memory site could be located in the dentate nucleus. Dentate neurons also receive collaterals from mossy fibers originating from the nucleus reticularis tegmenti pontis. How these two pathways to dentate neurons contribute to nuclear memory is an interesting question for future studies. 5. Internal models. Another current issue is determining how cerebellar forward and inverse models share their roles in the “real” cerebellum. There are only a small number of microcomplexes for which model properties have been defined. On the other hand, consideration of anatomical connections points to the possibility that the primary motor cortex (area 4) and the prefrontal cortex (areas 9 and 46) operate with a forward model, whereas the frontal eye field (area 8) seems to operate with an inverse model. Close comparison of these two systems should prove to be fruitful. 6. Evolution in cerebellar circuits and their functional roles. The large cerebellar mass contains numerous microcomplexes, which are in general designed uniformly throughout the cerebellum. However, there are also certain regional differences in the fundamental wiring of microcomplexes. These seem to be closely related to evolution. That is, the V/C-type of microcomplexes prevails in the evolutionarily old part of the cerebellum (Chapter 12, “Adaptive Control System Models,” Section 8). In contrast, the evolutionarily newer lateral zones of the cerebellum are characterized by the cerebrocerebellar communication loop, which provides an anatomical basis for the internal forward model (Figure 8A). The D1-zone, which is prominent in monkeys, is linked to the premotor cortex via the dorsal part of the dentate nucleus and the thalamus (Figure 49). These microzones appear to provide internal models of the body schema for the control of motor actions (Chapter 16, “Motor Actions and Tool Use”). The D2-zone, which is developed particularly well in humans, is linked to the prefrontal cortex via the ventral part of the dentate nucleus and it likely forms internal models of a mental model or Piaget’s schema in the control of cognitive functions (Chapter 17, “Cognitive Functions”). Lying between the old and new areas of the cerebellum, the intermediate hemisphere (C1–C3-zones) contains two types of microcomplex; an adaptive controller for reflexes and an internal model for voluntary movement (Figures 33 and 46). This region might also contain a microcomplex that provides an internal forward model for reflexes (Figure 45).

18 • Concluding Thoughts

209

Further research on structural-functional similarities and differences between the different zones will provide valuable information for refining our knowledge of the evolutionary features of the cerebellum. 7. Beyond movements. It is now clear that the classic view of the cerebellum as a motor center applies to the evolutionarily old flocculonodular lobe and the vermis (zones-A and B). This view applies also to the intermediate part of the cerebellar hemisphere, which is largely devoted to voluntary motor control. However, the intermediate part is involved also in sensory cancellation, which helps movements by removal of sensory perturbation, rather than by modulating the movements themselves. The unique circuit organization for the C2zone suggests such function (Figure 48). In this regard, recall that whales have an enormously developed C2 zone associated with a large posterior interpositus nucleus (Oelschläger and Oelschläger, 2009). It is an interesting possibility that whales’ particularly developed C2-zone-posterior-interpositus nucleus system is used for sensory cancelation and utilized for echolocation and/or acoustic communication across the ocean (Chapter 2, “Traditional Views of the Cerebellum,” Section 2). The lateral cerebellar hemisphere (D1- and D2-zones, underlain by the dentate nucleus) is even further remote from the sites for the modulation of movements. Shambes et al. (1978) showed that tactile sensory signals reach in abundance the rat cerebellar hemisphere. Gao et al. (1996) and Parsons et al. (1997) recognized that the human dentate nucleus is activated in association with sensation, but not with movements. The dentate nucleus can be activated by cutaneous stimuli, even when there are no accompanying overt finger movements. On the other hand, finger movements not associated with tactile sensory discrimination produced no dentate activation. These findings suggest that the primary function of the dentate nucleus is to process sensory information involved in motor, perceptual, and cognitive tasks. This interpretation is not incompatible with the present hypothesis that the dentate nucleus is the final output of the microcomplexes, which represent body schema in the D1-zone associated with the premotor cortex (Figure 50). Body schemata possess a continually updating map of the self’s body shape and postures (Chapter 1, “Neuronal Circuitry: The Key to Unlocking the Brain,” Section 8). A body schema is a set of neural representations of the body and bodily functions, based on sensory experience (Chapter 16, Section 3). One may therefore speculate that the sensory activation of the dentate nucleus is a reflection of the body schema representation in the temporoparietal cortex. A similar consideration may also apply to the control of mental activity, for which mental models or Piaget’s schema are formulated in the temporopariteal cortex

210

the cerebellum

and then copied by microcomplexs in the D2-zone as internal models. Such internal models should consist largely of sensory experiences. Modeling the cerebellar control of complex motor actions and nonmotor cognitive functions is an endeavor that currently features the use of psychological concepts of body schema, motor schema, Piaget’s schema, and mental models. We must realize that these valuable approaches are all in the realm of cognitive science and as yet, are not fully integrated into the field of neuroscience.

18-3 The Cerebellum and the Basal Ganglia The basal ganglia form a massive network in the deep interior of the cerebrum. With the cerebellum, it constitutes the two largest subcortical stations of motor system, with both operating implicitly. From the well-known characteristic symptoms of lesions in the basal ganglia, such as akinesia in Parkinson’s disease and chorea in Huntington’s disease, the primary function of the basal ganglia may appear to augment stabilization of complex activities in the CNS. “Stabilization augmentation” involves the selection of an activity that best fits the behavioral situation and context, and the suppression of other ongoing CNS activity that would interfere with the desired behavior (Mink, 1996). This notion is supported by the recent fMRI study of brain activity of professional and amateur players in a board game named Shogi: activations specific to professionals occurred in the caudate nucleus of the basal ganglia during quick generation of the best next move (Wan et al., 2011). This stabilization augmentation should not be confused with the “control augmentation” provided by the cerebellum. These two processes may not be achieved by the same strategy. It is important to realize that augmented stability may make a system less controllable or vice versa. Here we see the demand of the CNS to develop separate devices; the cerebellum for control augmentation and the basal ganglia for stabilization augmentation (Ito, 1986). A mechanism that might underlie the basal ganglia’s selection process is inhibiting—the inhibition that the substantial nigra pars reticulata exerts on a target system of the thalamus and/or other brainstem areas. This disinhibtion is exerted by the caudate nucleus that inhibits the substantial nigra pars reticulata. Selection occurs within the caudate nucleus where various cortical inputs “compete” and certain neurons “win.” The winning caudate neurons are thought to then suppress surrounding caudate neurons in a center (excitation)-surround (inhibition) manner. The result is that only a limited number of the substantia nigra pars reticulata neurons would be inhibited and their target system thereby released from inhibition. This disinhibitory pathway is presumably associated with a side path involving the subthalamic nucleus and the external segment of the globus pallidus, which may

18 • Concluding Thoughts

211

enhance inhibition in the substantia nigra pars reticulata. The interaction of dopaminergic projections from the substantia nigra pars compacta with cortical input in the caudate nucleus is considered to be the primary mechanism underlying this particular learning process. Such a selection-based stabilization mechanism has been investigated in studies on saccadic eye movements (see Hikosaka et al., 2000). The unique symptoms of basal ganglia disorders suggest that its postulated stabilization-by-selection mechanism does indeed regulate the multifaceted movements that occur during posture and locomotion. The mechanism has also been demonstrated for the control of saccadic eye movements. These movements are inhibited normally but, when appropriate, released from inhibition by a cortical mechanism (Hikosaka and Wurz, 1985). Although the basal ganglia receive inputs from the entire neocortex, their outputs are directed largely to the prefrontal cortex via the thalamus. This suggests that the basal ganglia contribute to the orderly operation of the prefrontal cortex as a controller by exerting the stabilization-byselection mechanism against numerous simultaneous, competing, and even conflicting inputs received from the entire neocortex (Hikosaka et al., 2000). Based on remarkable progresses that revealed unique neuronal circuit structures/functions and mechanisms of learning in the basal ganglia, most investigators now accept that the basal ganglia and cerebellum are both anatomically and functionally distinct (Graybiel, 2005). Nevertheless, a long-standing question still continues to ask whether the basal ganglia and cerebellum interact directly with each other. Two anatomical pathways have been identified that might come into play in their interactions. (1) The dentate nucleus has a disynaptic projection to the striatum, which is known to operate as an input stage in basal ganglia processing (Hoshi et al., 2005). (2) The subthalamic nucleus has a substantial disynaptic projection to the cerebellar cortex (Bostan et al., 2010). It remains to be shown just how these pathways function. A useful thought at this stage for the design of future experiments may be that the basal ganglia select a certain repertoire out of many prepared by the cerebellum in the form of differently tuned microcomplexes.

18-4 How Might Cerebellar Research Now Develop? In this monograph, I have emphasized that the cerebellum is on the forefront of brain structures where there has been a convergence of neurobiological analysis and computer modeling and simulation. Needless to say, the cerebellum is also on the forefront of a vast field of brain diseases where neurological, genetic, and molecular analyses have converged (Manto and Pandolfo, 2002). Efforts in these two major directions will help each other. In Chapter 1, I emphasized that cyclic decomposition-reconstruction is a fundamental methodology for studying a

212

the cerebellum

complex system like the brain. Admittedly, it is difficult if not impossible to apply this with instant success. Rather, the initial model provided by this approach is usually imperfect and it fails to represent the properties of the original system. Examples are now available, however, that through repeated trials using improved technologies, theories, and modeling, the reconstructed model can indeed come closer and closer to the complex system in its full essence. It is obvious that improved neuroscientific techniques and modeling hardware/software are on the immediate horizon. These advantages co-exist with the caution that there are still deficiencies in our basic knowledge of the cerebellum’s neuronal circuits. Sustained neurobiological efforts are required to clarify such issues at both the cellular/molecular level for analyzing single neurons and the systems level for advancing understanding of how neuronal circuits truly work. Further efforts are also required to proceed along the hierarchical levels of neural control, from reflexes to brainstem/spinal pattern generation, automatic and voluntary movements, motor actions, and cognitive functions. This is a logical and easily understood way for research on the cerebellum to progress and contribute to the advancement of overall brain research. Research on the cerebellum is also advantageous for considering implicit aspects of brain function. The cerebellum controls adaptively numerous reflexes and rhythmical movements, such as breathing and locomotion, at the unconscious level and also enables us to gain skill in the execution of voluntary movements and motor actions without being aware of just how the skill actually improves. Current research on the cerebellum is rapidly advancing our understanding of how this remarkable structure can govern unconsciously the implicit component of our mental activity. In this overall endeavor, we come to face the crucial question of how knowledge and idealization are represented and processed in neuronal circuits. This issue, like the creation of artificial intelligence, is one of the key challenges for research throughout the remainder of this century.

References Abe H, Shima HM, Sekiguchi H, Guo M, Nagao M, Tamura S, Kondo H (1994) Localization of mRNA for protein phosphatase 2A in the brain of adult rats. Mol Brain Res 22:139–143. Adrian ED (1935) Discharge frequencies in the cerebral and cerebellar cortex. J Physiol (London) 83:33P. Agnati LF, Zoli M, Strömberg I, Fuxe K (1995) Intercellular communication in the brain: wiring versus volume transmission. Neuroscience 69:711-726. Aguado F, Sanchez-Franco F, Cacidedo L, Fernandez T, Rodrigo J, Martinez-Murillo R (1992) Subcellular localization of insulin-like growth factor I (IGF-I) in Purkinje cells of the adult rat: an immunocytochemical study. Neurosci Let 135:171–174. Aguado F, Sanchez-Franco F, Rodrigo J, Cacicedo L, Martinez-Murillo R (1994) Insulin-like growth factor I-immunoreactive peptide in adult human cerebellar Purkinje cells: co-localization with lowaffinity nerve growth factor receptor. Neuroscience 59:641–650. Ahn S, Ginty DD, Linden DJ (1999) A late phase of cerebellar long-term depression requires activation of CaMKIV and CREB. Neuron 23:559–568. Airaksinen M, Panula P (1988) The histaminergic system in the guinea pig central nervous system: an immunocytochemical mapping study using an antiserum against histamine. J Comp Neurol 273:163-186. Aizenman CD, Linden DJ (1999) Regulation of the rebound depolarization and spontaneous firing patterns of deep nuclear neurons in slices of rat cerebellum. J Neurophysiol 82:1697–1709. Akaike T, Fanardjian VV, Ito M, Nakajima H (1973) Cerebellar control of the vestibulospinal tract cells in rabbit. Exp Brain Res 18:446–463. Akao T, Kurkin S, Fukushima J, Fukushima K (2005a) Visual and vergence eye movement related responses of pursuit neurons in the caudal frontal eye fields to motion-in-depth stimuli. Exp Brain Res 164:92–108. Akao T, Mustari MJ, Fukushima J, Kurkin S, Fukushima K (2005b) Discharge characteristics of pursuit neurons in MST during vergence eye movements. J Neurophysiol 93:2415–2434. Akazawa K, Aldridget JW, Steevest JD, Stein RB (1982) Modulation of stretch reflexes during locomotion in the mesencephalic cat. J Physiol (London) 329:553–567. Akazawa C, Shigemoto R, Bessho Y, Nakanishi S, Mizuno N (1994) Differential expression of five N-methyl-D-aspartate receptor subunit mRNAs in the cerebellum of developing and adult rats. J Comp Neurol 347:150–160. Albus JS (1971) Theory of cerebellar function. Mathem Biosci 10:25–61. 213

214

the cerebellum

Albus JS (1975) New Approach to Manipulator Control: The Cerebellar Model Articulation Controller (CMAC). J Dynamic Systems, Measurement, and Control September 220-227. Albus JS. Meystel AM (2001) Engineering of mind. An introduction to the science of intelligent systems. New York: John Wiley & Sons. p. 411. Allen G, Courchesne E (2003) Differential effects of developmental cerebellar abnormality on cognitive and motor functions in the cerebellum: an fMRI study of autism. Am J Psychiatry 160:262–273. Allen GI, Gilbert PFC, Yin TCT (1978) Convergence of cerebral inputs onto dentate neurons in monkey. Exp Brain Res 32:151–170. Alstermark B, Lan N, Pettersson L-G (2007) Building a realistic neuronal model that simulates multi-joint arm and hand movements in 3D space. HFSP Journal 1:209–214. Alstermark B, Ogawa J, Isa T (2004) Lack of monosynaptic corticomotoneuronal EPSPs in rats: disynaptic EPSPs mediated via reticulospinal neurons and polysynaptic EPSPs via segmental interneurons. J Neurophysiol 91:1832–1839. Alviña K, Walter JT, Kohn A, Ellis-Davies G, Khodakhah K (2008) Questioning the role of rebound firing in the cerebellum. Nat Neurosci 11:1256–1258. Amat J (1983) Interaction between signals from vestibular and forelimb receptors in Purkinje cells of the frog vestibulocerebellum. Brain Res 278:287–290. An CH, Atkeson CG, Hollerbach JM (1988) Model-Based Control of a Robot Manipulator. The MIT Press, Cambridge Massachusetts, London England. Andersson G, Armstrong DM (1987) Complex spikes in Purkinje cells in the lateral vermis (B zone) of the cat cerebellum during locomotion. J Physiol (London) 385:107–134. Andersson G, Oscarsson O (1978) Projections to lateral vestibular nucleus from cerebellar climbing fiber zones. Exp Brain Res 32:4549-4563. Anzai M, Kitazawa H, Naga S (2010) Effects of reversible pharmacological shutdown of cerebellar flocculus on the memory of long-term horizontal vestibulo-ocular reflex adaptation in monkeys. Neurosci Res 68:191–198. Aoki E, Semba R, Kashiwamata S (1986) New candidates for GABAergic neurons in the rat cerebellum: an immunocytochemical study with antiGABA antibody. Neurosci Lett 68:267–271. Apps R, Garwicz M (2005) Anatomical and physiological foundations of cerebellar information processing. Nat Rev Neurosci 6:297-311. Apps R, Hawkes R (2009) Cerebellar cortical organization: a one-map hypothesis. Nat Rev Neurosci 10:670–681. Arata A, Fujii M, Ito M (2009) Respiratory activity involved in the cerebellar circuit formation. 36th IUPS Congress 2009 Kyoto P2PM-11-7. Arata A, Ito M (2004) Purkinje cell functions in the in vitro cerebellum isolated from neonatal rats in a block with the pons and medulla. Neurosci Res 50:361–367. Arbib MA (1971) The metaphorical brain. An introduction to cybernetics as artificial intelligence and brain theory. New York: John Wiley & Sons. Arbib MA (2006) A sentence is to speech as what is to action? Cortex 42:507-514. Arenz A, Silver RA, Schaefer AT, Margrie TW (2008) the contribution of single synapses to sensory representation in vivo. Science 321:977–980.

References

215

Armano S, Rossi P, Taglietti V, D’Angelo E (2000) Long-term potentiation of intrinsic excitability at the mossy fiber-granule cell synapse of rat cerebellum. J Neurosci 20:5208–5216. Arnold DB, Robinson DA (1997) The oculomotor integrator: testing of a neural network model. Exp Brain Res 113:57–74. Arshavsky YI, Gelfand IM, Orlovsky GN, Pavlova GA (1978) Messages conveyed by spinocerebellar pathways during scratching in the cat. II. Activity of neurons of the ventral spinocerebellar tract. Brain Res 151:493–506. Arshavsky YI, Orlovsky GN, Perret C (1988) Activity of rubrospinal neurons during locomotion and scratching in the cat. Behav Brain Res 28:193–199. Asanuma H, Larsen K, Yumiya H (1997) Peripheral input pathways to the monkey motor cortex. Exp Brain Res 38:349–355. Atherton M, Zhuang J, Bart WM, Hu X, He S (2003) A functional MRI study of high-level cognition. I. The game of chess. Cogn Brain Res 16:26–31. Attwell PJ, Ivarsson M, Millar L, Yeo CH (2002) Cerebellar mechanisms in eyeblink conditioning. Ann NY Acad Sci 978:79–92. Attwell PJE, Rahman S, Yeo CH (2001) Acquisition of eyeblink conditioning is critically dependent on normal function in cerebellar cortical lobule HVI. J Neurosci 21:5715–5722. Audinat E, Knoepfel T, Gahwiler BH (1990) Responses to excitatory amino acids of Purkinje cells’ and neurons of the deep nuclei in cerebellar slice cultures. J Physiol (London) 430:297–313. Auger C, Attwell D (2000) Fast removal of synaptic glutamate by postsynaptic transporters. Neuron 28:547–558. Azizi SA, Burne RA, Woodward DJ (1985) The auditory corticopontocerebellar projection in the rat: inputs to the paraflocculus and midvermis. An anatomical and physiological study. Exp Brain Res 59:36–49. Baddeley A (2003) Working memory: looking back and looking forward. Nat Rev Neurosci 4:829–839. Baev, KV (1999) Biological neural networks: hierarchical concept of brain function. Boston, Basel, Berlin: Birkhäeuser. Bagnall MW, Zingg B, Sakatos A, Moghadam SH, Zeilhofer HU, du Lac S (2009) Glycinergic projection neurons of the cerebellum. J Neurosci 29:10104–10110. Baker AKR, Simpson JI (1975) Afferents to the vestibulo-cerebellum and the origin of the visual climbing fibers in the rabbit. Brain Res 98:582–589. Balaban CD, Freilino M, Romero GG (1999) Protein kinase C inhibition blocks the early appearance of vestibular compensation. Brain Res 845:97–101. Balaban CD, Ito M, Watanabe E (1981) Demonstration of zonal projections from the cerebellar flocculus to vestibular nuclei in monkeys (Macaca fuscata). Neurosci Lett 27:101–105. Barash S, Melikyan A, Sivakov A, Zhang M, Glickstein M, Thier P (1999) Saccadic dysmetria and adaptation after lesions of the cerebellar cortex. J Neurosci 19:10931–10939. Barbour B (1993) Synaptic currents evoked in Purkinje cells by stimulating individual granule cells. Neuron 11:759–769. Barbour B, Häusser M (1997) Intersynaptic diffusion of neurotransmitter. Trends Neurosci 20:377–384.

216

the cerebellum

Barlow JS (2002) The cerebellum and adaptive control. Cambridge University Press. Barmack NH, Baughman RW, Eckenstein FP (1992a) Cholinergic innervation of the cerebellum of rat, rabbit, cat, and monkey as revealed by choline acetyltransferase activity and immunohistochemistry. J Comp Neurol 317:233–249. Barmack NH, Baughman RW, Eckenstein FP, Shojaku H (1992b) Secondary vestibular cholinergic projection to the cerebellum of rabbit and rat as revealed by choline acetyltransferase immunohistochemistry, retrograde and orthograde tracers. J Comp Neurol 317:250–270. Barmack NH, Yakhnitsa V (2008) Functions of interneurons in mouse cerebellum. J Neurosci 28:1140-1152. Batchelor AM, Garthwaite J (1993) Novel synaptic potentials in cerebellar Purkinje cells: probable mediation by metabotropic glutamate receptors. Neuropharmacology 32:11–20. Batini C, Ito M, Kado RT, Jastreboff PJ, Miyashita Y (1979) Interaction between the horizontal vestibulo-ocular reflex and optokinetic response in rabbits. Exp Brain Res 37:1–15. Batini C, Moruzzi G, Pompeiano O (1957) Cerebellar release phenomena. Arch Ital Biol 95:71–95. Baude A, Molnar E, Latawiec D, Mcilhinney RA, Somogyi P. (1994) Synaptic and nonsynaptic localization of the GluR1 subunit of the AMPA-type excitatory amino acid receptor in the rat cerebellum. Neuroscience 14:2830–2843. Bell CC, Han VZ, Sugawara Y, Grant K (1997) Synaptic plasticity in a cerebellum-like structure depends on temporal order. Nature 387:278–281. Bellamy TC, Ogden D (2005) Short-term plasticity of Bergmann glial cell extrasynaptic currents during parallel fiber stimulation in rat cerebellum. Glia 52:325–335. Belton T, Suh M, Simpson JI (2002) The nonvisual complex spike signal in the flocculus responds to challenges to the vestibulo-ocular reflex gain. Ann NY Acad Sci 978:503–504. Bengtsson F, H Jörntell (2009a) Climbing fiber coupling between adjacent Purkinje cell dendrites in vivo. Front Cell Neurosci 3: article 7, 1–9. Bengtsson F, Jörntell H (2009b) Sensory transmission in cerebellar granule cells relies on similarly coded mossy fiber inputs. PNAS USA 106:2389–2394. Bengtsson F, Svensson P, Hesslow G (2004) Feedback control of Purkinje cell activity by the cerebello-olivary pathway. Eur J Neurosci 20:2999–3005. Bergles DE, Jahr CE (1997) Synaptic Activation of Glutamate Transporters in Hippocampal Astrocytes. Neuron 19:1297–1308 Bezprozvanny I, Watras J. Ehrlich BE (1991) Bell-shaped calcium-response curves of Ins (1,4,5) P3and calcium-gated channels from endoplasmic reticulum of cerebellum. Nature 351:751–754. Bigelow NO, Turner BM, Andreasen NC, Paulsen JS, O’Leary DS, B-C Ho (2006) Prism adaptation in schizophrenia. Brain Cogn 61:235–242. Billups D, Liu YB, Birnstiel S, Slater NT (2002) NMDA receptor- mediated currents in rat cerebellar granule and unipolar brush cells. J Neurophysiol 87:1948–1959. Bizzi E (1968) Discharge of frontal eye field neurons during saccadic and following eye movements in unanesthetized monkeys. Exp Brain Res 6:69–80. Blakemore SJ, Oakley DA, Frith CD (2003) Delusions of alien control in the normal brain. Neuropsychologia 41:1058–1067. Blakemore SJ, Sirigu A (2003) Action prediction in the cerebellum and in the parietal lobe. Exp Brain Res 153:239–245.

References

217

Blakemore SJ, Smith J, Steel R, Johnstone EC, Frith CD (2000) The perception of self-produced sensory stimuli in patients with auditory hallucinations and passivity experiences: evidence for a breakdown in self-monitoring. Psychol Med 30:1131–1139. Blakemore SJ, Wolpert DM, Frith CD (1998) Central cancellation of self-produced tickle sensation. Nat Neurosci 1:635–640. Blazquez PM, Hirata Y, Heiney SA, Green AM, Highstein SM (2003) Cerebellar signatures of vestibulo-ocular reflex motor learning. J Neurosci 23:9742–9751. Blazquez PM, Hirata Y, Highstein SM (2006) Chronic Changes in Inputs to Dorsal Y Neurons Accompany VOR Motor Learning. J Neurophysiol 95:1812-1825. Blazquez PM, Hirata Y, Highstein SM (2004) The vestibulo-ocular reflex as a model system for motor learning: what is the role of the cerebellum? Cerebellum 3:188–192. Bliss TVP, Lomo T (1973) Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J Physiol (London) 232:357–374. Bolk L (1906) Das Cerebellum der Säugetiere. G. Fisher. Bosco G, Eian J, Poppele RE (2006) Phase-specific sensory representations in spinocerebellar activity during stepping: evidence for a hybrid kinematic/kinetic framework. Exp Brain Res 175:83–96. Bostan AC, Dum RP, Strick PL (2010) The basal ganglia communicate with the cerebellum. PNAS USA 107:8452–8456. Bourinet E, Soong TW, Sutton K, Slaymaker S, Mathews E, Monteil A, Zamponi GW, Nargeot J, Snutch TP (1999) Splicing of a1A subunit gene generates phenotypic variants of P- and Q- type calcium channels. Nat Neurosci 2:407–415. Boxall AR, Garthwaite J (1996) Long-term depression in rat cerebellum requires both NO synthase and NO-sensitive guanylyl cyclase. Eur J Neurosci 8:2209–2212. Bracha V, Irwin KB, Webster ML, Wunderlich DA, Stachowiak MK, Bloedel JR (1998) Microinjections of anisomycin into the intermediate cerebellum during learning affect the acquisition of classically conditioned responses in the rabbit. Brain Res 788:169–178. Braitenberg V, Atwood RP (1958) Morphological observations on the cerebellar cortex. J Comp Neurol 109:1–27. Brebner JT, Welford AT (1980) Introduction: an historical background sketch. In: Welford AT, editor. Reaction Times. New York: Academic Press. pp. 1–23. Bridgeman B (1995) A review of the role of efference copy in sensory and oculomotor control systems. An Biomed Engineer 23:409–422. Brindley GS (1964) The use made by the cerebellum of the information that it receives from sense organs. IBRO Bull 3:80. Brock LG, Coombs JS, Eccles JC (1952) The recording of potentials from motoneurons with an intracellular electrode. J Physiol (London) 117:431–460. Brodal A (1972) Vestibulocerebellar input in the cat: anatomy. In: Brodal A, Pompeiano O, editors. Basic aspects of central vestibular mechanisms. Prog Brain Res 37:315–328. Brodal P (1980) The cortical projection to the nucleus reticularis tegmenti pontis in the rhesus monkey. Exp Brain Res 38:19–27. Brodal P (1982a) The cerebropontocerebellar pathway: salient features of its organization. In: The cerebellum—new vistas. Exp Brain Res Suppl. 6 Springer-Verlag, Berlin-Heidelberg, 108–132.

218

the cerebellum

Brodal P (1982b) Further observations on the cerebellar projections from the pontine nuclei and the nucleus reticularis tegmenti pontis in the rhesus monkey. J Comp Neurol 204:44–55. Brodal P, Bjaalle JG (1992) Organization of the pontine nuclei. Neurosci Res 13:83–118. Brodmann K (1909) Vergleichende Lokalisationslehre der Grosshimrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues. Leipzig: Barth. Broussard DM, Kassardjian CD (2004) Learning in a simple motor system. Learn Mem 11:127–136. Brown SP, Safo PK, Regehr WG (2004) Endocannabinoids inhibit transmission at granule cell to Purkinje cell synapses by modulating three types of presynaptic calcium channels. J Neurosci 24:5623–5631. Brunel N, Hakim V, Isope P, Nodal J-P, Barbour B (2004) Optimal information storage and the distribution of synaptic weights: perceptron versus Purkinje cells. Neuron 43:745–757. Buccino EG, Binkofski F, Fink GR, Fadiga L, Fogassi L, Gallese V, Seitz RJ, Zilles K, Rizzolatti G, Freund H-J (2001) Action observation activates premotor and parietal areas in a somatotopic manner: an fMRI study. J Neurosci 13:400–404. Buonomano DV, Mauk MD (1994) Neural network model of the cerebellum: temporal discrimination and the timing of motor responses. Neur Comput 6:38–55. Burke D, Gandevia SC, McKeon B (1984) Monosynaptic and oligosynaptic contributions to human ankle jerk and H-reflex. J Neurophysiol 52:435–448. Burman K, Darian-Smith C, Darian-Smith I (2000a) Macaque red nucleus: origins of spinal and olivary projections and terminations of cortical inputs. J Comp Neurol 423:178–196. Burman K, Darian-Smith C, Darian-Smith I (2000b) Geometry of rubrospinal, rubroolivary, and local circuit neurons in the macaque red nucleus. J Comp Neurol 423:197–219. Cajal Ramon y S (1911) Histologie du système nerveux de l’homme et des vertebrés. II. (French edition by L. Azoulay, 1955). Madrid: Inst. Ramon y Cajal. Carper RA, Courchesne E (2000) Inverse correlation between frontal lobe and cerebellum sizes in children with autism. Brain 123:836–844. Carr L, Iacoboni M, Dubeau M-C, Mazziotta JC, Lenzi GL (2003) Neural mechanisms of empathy in humans: a relay from neural systems for imitation to limbic areas. PNAS USA 100:5497–5502. Carroll RC, Beattie EC, von Zastrow M, Malenka RC (2001) Role of AMPA receptor endocytosis in synaptic plasticity. Nat Rev Neurosci 2:315–324. Carter AG, Regehr WG (2000) Prolonged synaptic current and glutamate spillover at the parallel fiber to stellate cell synapse. J Neurosci 20:4423–4434. Cerminara NL, Apps R, Marple-Horvat DE (2008) An internal model of a moving visual target in the lateral cerebellum. J Physiol (London) 587:429–442. Chadderton P, Margrie TW, Hausser M (2004) Integration of quanta in cerebellar granule cells during sensory processing. Nature 428:856–860. Chan-Palay V (1977) Cerebellar dentate nucleus. Organization, cytology and transmitters. Berlin, Heidelberg, New York: Springer-Verlag. Chaudhry FA, Lehre KP, van Lookeren Campagne M, Ottersen OP, Danbolt NC, Storm-Mathisen J (1995) Glutamate transporters in glial plasma membranes: highly differentiated localizations revealed by quantitative ultrastructural immunocytochemistry. Neuron 15:711–720.

References

219

Chen CC, Brücke C, Kempf F, Kupsch A, Lu CS, Lee ST, Tisch S, Limousin P, Hariz M, Brown P (2006) Deep brain stimulation of the subthalamic nucleus: a two-edged sword. Current Biology 16:R952–R953. Chen S, Hillman DE (1993) Colocalization of neurotransmitters in the deep cerebellar nuclei. J Neurocytol 22:81–91. Chen X, Zhang D, Zhang X, Li Z, Meng X (2003) A functional MRI study of high-level cognition. II. The game of GO. Cogn Brain Res 16:32–37. Chen XY, Wolpaw JR (2002) Probable corticospinal tract control of spinal cord plasticity in the rat. J Neurophysiol 87:645–652. Chen XY, Wolpaw JR (2005) Operant conditioning of H-reflex in freely moving rats. J Neurophysiol 73:411–415. Christian KM, Thompson RF (2003) Neural substrates of eyeblink conditioning: acquisition and retention. Learn Mem 10:427–455. Clark BA, Barbour B (1997) Currents evoked in Bergmann glial cells by parallel fibre stimulation in rat cerebellar slices. J Physiol (London) 502:335–350. Clendenin M, Ekerot C-F, Oscarsson O, Rosén I (1974) The lateral reticular nucleus in the cat I. Mossy fibre distribution in cerebellar cortex. Exp Brain Res 21:473–486. Clower DM, Dum RP, Strick PL (2005) Basal ganglia and cerebellar inputs to ‘AIP’. Cereb Cortex 15:913–920. Cohen P (1989) The structure and regulation of protein phosphatases. Ann Rev Biochem 58:453–508. Colin F, Manil J, Desclin JC (1980) The olivocerebellar system. I. Delayed and slow inhibitory effects: an overlooked salient feature of cerebellar climbing fibers. Brain Res. 187: 3-27. Collewijn H (1969) Optokinetic eye movements in the rabbit: input-output relations. Vision Res 9:117–132. Collewijn H, Grootendorst AF (1979) Adaptation of optokinetic and vestibulo-ocular reflexes to modified visual input in the rabbit. Prog Brain Res 50:771–781. Collinson SL, Anthonisz B, Courtenay D, Winter C (2006) Frontal executive impairment associated with paraneoplastic cerebellar degeneration: a case study. Neurocase 12:350–354. Condorelli DF, Parenti R, Spinella F, Trovato-Salinaro A, Belluardo N, Cardile V, Cicirata F (2001) Cloning of a new gap junction gene (Cx36) highly expressed in mammalian brain neurons. Eur J Neurosci 10:1202–1208. Connor S, Bloomfield J, LeBoutillier JC, Thompson RF, Petit TL, Weeks AC (2009) Eyeblink conditioning leads to fewer synapses in the rabbit cerebellar cortex. Behav Neurosci 123:856–862. Conquet F, Bashir ZI, Davies CH, Daniel H, Ferraguti F, Bordi F, Franz-Bacon K, Reggiani A, Matarese V, Conde F, Collingridge GL, Crepel F (1994) Motor deficit and impairment of synaptic plasticity in mice lacking mGluR1. Nature 372:237–243. Cooke JD, Larson B, Oscarsson O, Sjölund B (1971) Origin and termination of cuneocerebellar tract. Exp Brain Res 13:339–358. Courjon JH, Flandrin JM, Jeannerod M, Schmid R (1982) The role of the flocculus in vestibular compensation after hemilabyrinthectomy. Brain Res 239:251–257. Courtemanche R, Lamarre Y (2005) Local field potential oscillations in primate cerebellar cortex: synchronization with cerebral cortex during active and passive expectancy. J Neurophysiol 93:2039–2052.

220

the cerebellum

Courtemanche R, Pellerin JP, Lamarre Y (2002) Local field potential oscillations in primate cerebellar cortex: modulation during active and passive expectancy. J Neurophysiol 88:771–782. Coutinho V, Mutoh H, Knöpfel T (2004) Functional topology of the mossy fibre-granule cell-Purkinje cell system revealed by imaging of intrinsic fluorescence in mouse cerebellum. Eur J Neurosci 20:740–748. Cowey A, Walsh V (2001) Tickling the brain: studying visual sensation, perception and cognition by transcranial magnetic stimulation. Prog Brain Res 134:411–425. Craik KJW (1943) The nature of explanation. Cambridge UK: Cambridge University Press. Crepel F, Audinat E, Daniel H, Hemart N, Jaillard D, Rossier J, Lambolez B (1994) Cellular locus of the nitric oxide-synthase involved in cerebellar long-term depression induced by high external potassium concentration. Neuropharmacology 33:1399–1405. Crepel F, Jaillard D (1990) Protein kinases, nitric oxide and long-term depression of synapses in the cerebellum. NeuroReport 1:133–136. Crepel F, Jaillard D (1991) Pairing of pre- and postsynaptic activities in cerebellar Purkinje cells induces long-term changes in synaptic efficacy in vitro. J Physiol (London) 432:123–141. Crepel F, Krupa M (1988) Activation of protein kinase C induces a long-term depression of glutamate sensitivity of cerebellar Purkinje cells. An in vitro study. Brain Res 458:397–401. Czubayko U, Sultan F, Thier P, Schwarz C (2001) Two types of neurons in the rat cerebellar nuclei as distinguished by membrane potentials and intracellular fillings. J Neurophysiol 85:2017–2029. D’Angelo E (2008) The critical role of Golgi cells in regulating spatio-temporal integration and plasticity at the cerebellum input stage. Front Neurosci 2:35–46. D’Angelo E, Rossi P, Armano S, Taglietti V (1999) Evidence for NMDA and mGlu receptordependent long-term potentiation of mossy fiber-granule cell transmission in rat cerebellum. J Neurophysiol 81:277–287. Daniel H, Hemart N, Jaillard D, Crepel F (1993) Long-term depression requires nitric oxide and guanosine 3'5' cyclic monophosphate production in rat cerebellar Purkinje cells. Eur J Neurosci 5:1079–1082. Daniel H, Levenes C, Crepel F (1998) Cellular mechanisms of cerebellar LTD. Trends Neurosci 21:401–407. De Gruijl JR, van der Smagt P, de Zeeuw CI (2009) Anticipatory grip force control using a cerebellar model. Neuroscience 162:777–786. De Schutter E (1995) Cerebellar long-term depression might normalize excitation of Purkinje cells: a hypothesis. Trends Neurosci 18:291–295. De Schutter E (1999) Using realistic models to study synaptic integration in cerebellar Purkinje cells. Rev Neurosci 10:233–245. De Schutter E, Bower JM (1994a). An active membrane model of the cerebellar Purkinje cell: I. Simulation of current clamps in slice. J Neurophysiol 71:375–400. De Schutter E, Bower JM (1994b) An active membrane model of the cerebellar Purkinje cell. II. Simulation of synaptic responses. J Neurosci 71:401–419. De Zeeuw CI, Berrebi AS (1995) Postsynaptic targets of Purkinje cell terminals in the cerebellar and vestibular nuclei of the rat. Eur J Neurosci 7:2322–2333.

References

221

De Zeeuw CI, Hansel C, Bian F, Koekkoek SK, van Alphen AM, Linden DJ, Oberdick J (1998) Expression of a protein kinase C inhibitor in Purkinje cells blocks cerebellar LTD and adaptation of the vestibulo-ocular reflex. Neuron 20:495–508. De Zeeuw CI, Holstege JC, Calkoen F, Ruigrok TJH, Voogd J (1988) A new combination of WGAHRP anterograde tracing and GABA immunocytochemistry applied to afferents of the cat inferior olive at the ultrastructural level. Brain Res 447:369–375. De Zeeuw CI, Holstege JC, Ruigrok TJH, Voogd J (1989) Ultrastructural study of the GABAergic, cerebellar, and mesodiencephalic innervation of the cat medial accessory olive: anterograde tracing combined with immunocytochemistry. J Comp Neurol 284:12–35. De Zeeuw CI, Wylie DR, Digiorgi PL, Simpson JI (2004) Projections of individual Purkinje cells of identified zones in the flocculus to the vestibular and cerebellar nuclei in the rabbit. J Comp Neurol 3:428–447. Dean I, Robertson SJ, Edwards FA (2003) Serotonin drives a novel GABAergic synaptic current recorded in rat cerebellar Purkinje cells: a Lugaro cell to Purkinje cell synapse. J Neurosci 23:4457–4469. Dean P, Porrill J, Ekerot C-F, Jörntell H (2010) The cerebellar microcircuit as an adaptive filter: experimental and computational evidence. Nat Rev Neurosci 11:30–43. Debaerea F, Wenderotha N, Sunaert S, Van Hecke P, Swinnen SP (2004) Cerebellar and premotor function in bimanual coordination: parametric neural responses to spatiotemporal complexity and cycling frequency. NeuroImage 21:1416–1427. Deiters OFK (1865) Untersuchungen über Gehirn und Rückenmark des Menschen und der Säugetiere. Braunschweig: Vieweg. Descartes R (1649) Discourse on the Method of Rightly Conducting the Reason, and Seeking Truth in the Sciences. Project Gutenberg, etext93/dcart10.txt (1996/12/20, 141533 bytes) [cited 1 Jan 2011]. Desmond JE, Gabrieli JDE, Wagner AD, Ginier BL, Glove GH (1997) Lobular patterns of cerebellar activation in verbal working memory and finger-tapping tasks as revealed by functional MRI. J Neurosci 17:9675–9685. Detre JA, Nairn AC, Aswad DW, Greengard P (1984) Localization in mammalian brain of G-substrate, a specific substrate for guanosine 3'5'-cyclic monophosphate-dependent protein kinase. J Neurosci 4:2843–2849. Deuel RK (2002) Autism: a cognitive developmental riddle. Pediatr Neurol 26:349–357. Devor A, Yarom Y (2002) Electrotonic coupling in the inferior olivary nucleus revealed by simultaneous double patch recordings. J Neurophysiol 87:3048–3058. Diamond A (2000) Close interrelation of motor development and cognitive development and the cerebellum and prefrontal cortex. Child Develop 71:44–56. Dieudonne S (1998) Submillisecond kinetics and low efficacy of parallel fibre-Golgi cell synaptic currents in the rat cerebellum. J Physiol (London) 510:845–866. Dieudonne S, Dumoulin A (2000) Serotonin-driven long-range inhibitory connections in the cerebellar cortex. J Neurosci 20:1837–1838. DiGregorio DA, Nusser Z, Silver RA (2002) Spillover of glutamate onto synaptic AMPA receptors enhances fast transmission at a cerebellar synapse. Neuron 35:521–533. (Comment in: Neuron 2002. 35, 412–414. Erratum in: Neuron 2002. 36, p. 187).

222

the cerebellum

Diño MR, Perachio AA, Mugnaini E (1999) Distribution of unipolar brush cells and other calretinin immunoreactive components in the mammalian cerebellar cortex. J Neurocytol 28:99–123. Diño MR, Perachio AA, Mugnaini E (2001) Cerebellar unipolar brush cells are targets of primary vestibular afferents: an experimental study in the gerbil. Exp Brain Res 140:162–170. Doi T, Kuroda S, Michikawa T, Kawato M (2005) Inositol 1,4,5- trisphosphate-dependent Ca2+ threshold dynamics detect spike timing in cerebellar Purkinje cells. J Neurosci 25:950–961. Dow RS, Moruzzi G (1958) The physiology and pathology of the cerebellum. Minneapolis: University of Minnesota Press. Doya K (1999) What are the computations of the cerebellum, the basal ganglia, and the cerebral cortex. Neural Netw 12:961–974. Dufossé M, Ito M, Jastreboff PJ, Miyashita Y (1978) A neuronal correlate in rabbit’s cerebellum to adaptive modification of the vestibulo-ocular reflex. Brain Res 150:611–616. Dugué GP, Dumoulin A, Triller A, Dieudonné S (2005) Target-dependent use of coreleased inhibitory transmitters at central synapses. J Neurosci 25:6490-6498. Dum RP, Strick PL (2003) An unfolded map of the cerebellar dentate nucleus and its projections to the cerebral cortex. J Neurophysiol 89:634–639. Dumoulin A, Triller A, Dieudonné S (2001) IPSC kinetics at identified GABAergic and mixed GABAergic and glycinergic synapses onto cerebellar Golgi cells. J Neurosci 21:6045–6057. Dzubay JA, Jahr CE (1999) The concentration of synaptically released glutamate outside of the climbing fiber-Purkinje cell synaptic cleft. J Neurosci 19:5265–5274. Ebner TJ, Pasalar S (2008) Cerebellum predicts the future motor state. Cerebellum 7:583–588. Eccles JC, Fatt P, Landgren S, Winsbury GJ (1954) Spinal cord potentials generated by volleys in the large muscle afferents. J Physiol (London) I25:590–606. Eccles JC, Ito M, Szentagothai J (1967) The cerebellum as a neuronal machine. New York: Springer-Verlag. Eccles JC, Llinás R, Sasaki K (1966a) Intracellularly recorded responses of the cerebellar Purkinje cells. Exp Brain Res 1:161–183. Eccles JC, Llinás R, Sasaki K (1966b) The excitatory synaptic action of climbing fibres on the Purkinje cells of the cerebellum. J Physiol (London) 182:268–296. Edamura M, Yang JF, Stein RB (1991) Factors that determine the magnitude and time course of human H-reflexes in locomotion. J Neurosci 11:420–427. Eiraku M, Tohgo A, Ono K, Kaneko M, Fujishima K, Hirano T, Kengaku M (2005) DNER acts as a neuron-specific notch ligand during Bergmann glial development. Nat Neurosci 8:873–880. Ekerot CF, Jörntell H (2001a) Parallel fibre receptive fields: a key to understanding cerebellar operation and learning. Cerebellum 2:101–109. Ekerot CF, Jörntell H (2001b) Parallel fibre receptive fields of Purkinje cells and interneurons are climbing fibre-specific. Eur J Neurosci 13:1303–1310. Ekerot CF, Jörntell H, Garwicz M (1995) Functional relationship between corticonuclear input and movements evoked on microstimulation in cerebellar nucleus interpositus anterior in the cat. Exp Brain Res 106:365–376. Ekerot C-F, Kano M (1985) Long-term depression of parallel fibre synapses following stimulation of climbing fibres. Brain Res 342:357–360.

References

223

Endo S, Nairn AC, Greengard P, Ito M (2003) Thr123 of rat G-substrate contributes to its action as a protein phosphatase inhibitor. Neurosci Res 45:79–89. Endo S, Shutoh F, Le TD, Okamoto T, Ikeda T, Suzuki M, Kawahara S, Yanagihara D, Sato Y, Yamada K, Sakamoto T, Kirino Y, Hartell NA, Yamaguchi K, Itohara S, Nairn AC, Greengard P, Nagao S, Ito M (2009) Dual involvement of G-substrate in motor learning revealed by gene deletion. PNAS USA 106:3525–3530. Endo S, Suzuki M, Sumi M, Nairn AC, Morita R, Yamakawa K, Greengard P, Ito M (1999) Molecular identification of human G-substrate, a possible downstream component of the cGMP-dependent protein kinase cascade in cerebellar Purkinje cells. PNAS USA 96:2467–2472. Englund C, Kowalczyk T, Daza RAM, Dagan A, Lau C, Rose MF, Hevner RF (2006) Unipolar brush cells of the cerebellum are produced in the rhombic lip and migrate through developing white matter. J Neurosci 26:9184 -9195. Ethier V, DS Zee, Shadmehr R (2008) Changes in control of saccades during gain adaptation. J Neurosci 28:13929 –13937. Everling S, Fischer B (1998) The antisaccade: a review of basic research and clinical studies. Neuropsychologia 36:885–899. Ezure K, Graf W (1984a) A quantitative analysis of the spatial organization of the vestibulo-ocular reflexes in lateral- and frontal-eyed animals—I. Orientation of semicircular canals and extraocular muscles. Neuroscience 12:85–93. Ezure K, Graf W (1984b) A quantitative analysis of the spatial organization of the vestibulo-ocular reflexes in lateral- and frontal-eyed animals—II. Neuronal networks underlying vestibulo-oculomotor coordination. Neuroscience 12:95–109. Fadi X, Frazier DT (2000) Respiratory motor output by cerebellar deep nuclei in the rat. J Appl Physiol 89:996–1004. Fang PC, Stepniewska I, Kaas J, Ispilateral H (2005) Cortical connections of motor, premotor, frontal eye, and posterior parietal fields in a prosimian primate, Otalemur garnetti. J Comp Neurol 490:305–333. Fatemi SH, Halt AR, Realmuto G, Earle J, Kist DA, Thuras P, Merz A (2002) Purkinje cell size is reduced in cerebellum of patients with autism. Cell Mol Neurobiol. 22:171–175. Feil R, Hartmann J, Luo C, Wolfsgruber W, Schilling K, Feil S, Barski JJ, Meyer M, Konnerth A, De Zeeuw CI, Hofmann F (2003) Impairment of LTD and cerebellar learning by Purkinje cell-specific ablation of cGMP-dependent protein kinase I. J Cell Biol 163:295–302. Feldman AG (1986) Once more on the equilibrium-point hypothesis (lambda model) for motor control. J Mot Behav 18:17–54. Fellows SJ, Ernst J, Schwarz M, Töpper R, Noth J (2001) Precision grip deficits in cerebellar disorders in man. Clin Neurophysiol 112:1793–802. Fiez JA, Petersen SE, Cheney MK, Raichle ME (1992) Impaired non-motor learning and error detection associated with cerebellar damage. A single case study. Brain 115:155–178. Fiez JA, Raicle ME, Balota DA, Tallal P, Petersen SE (1996) PET activation of posterior temporal regions during auditory word presentation and verb generation. Cereb Cortex 6:1–10. Finch EA, Augustine GJ (1998) Local calcium signalling by inositol-1,4,5-trisphosphate in Purkinje cell dendrites. Nature 396:753–756. Finch EA, Turner TJ, Goldin SM (1991) Calcium as a coagonist of inositol 1,4,5-trisphosphateinduced calcium release. Science 252:443–446.

224

the cerebellum

Flament D, Hore J (1988) Comparison of cerebellar intention tremor under isotonic and isometric conditions. Brain Res 439:179–186. Flourens J-P (1822) cited from Dow RS, Moruzzi G (1958) The physiology and pathology of the cerebellum. Flourens J-P (1842) cited from Dow RS, Moruzzi G (1958) The physiology and pathology of the cerebellum. Forti L, Cesana E, Mapelli J, D’Angelo E (2006) Ionic mechanisms of autorhythmic firing in rat cerebellar Golgi cells. J Physiol (London) 574:711–729. Fox CA, Bernard JW (1957) A quantitative study of the Purkinje cell dendritic branches and their relationship to afferent fibres. J Anat 91:299–313. Freeman JH, Scharengerg AM, Olds JL, Schreurs BG (1998) Classical conditioning increases membrane-bound protein kinase C in rabbit cerebellum. NeuroReport 9:2669–2673. Fried I, Katz A, McCarthy G, Sass KJ, Williamson P, Spencer SS, Spencer DD (1991) Functional organization of human supplementary motor cortex studied by electrical stimulation. J Neurosci 11:3656–3666. Frith SD, Blakemore SJ, Wolpert DM (2000) Abnormalities in the awareness and control of action. Phil Trans R Soc London B 355:1771–1788. Fu QG, Flament D, Coltz JD, Ebner TJ (1997) Relationship of cerebellar Purkinje cell simple spike discharge to movement kinematics in the monkey. J Neurophysiol 78:478–491. Fujii M, Ueno K, Asamizuya T, Nakatani H, Wan X, Cheng K, Ito M (2009) cerebellar activation during Check-mate task execution. Neurosci Res. 65:S240. Fujita M (1982a) Adaptive filter model of the cerebellum. Biol Cybern 45:195–206. Fujita M (1982b) Simulation of adaptive modification of the vestibulo-ocular reflex with an adaptive filter model of the cerebellum. Biol Cybern 45:207–214. Fujita M (2005) Feed-forward associative learning for volitional movement control. Neurosci Res 52:153–165. Fukuda J, Highstein SM, Ito M (1972) Cerebellar inhibitory control of the vestibulo-ocular reflex investigated in rabbit’s IIIrd nucleus. Exp Brain Res 14:511–526. Fukushima K, Kaneko CRS, Fuchs AF (1992) The neuronal substrate of integration in the oculomotor system. Prog Neurobiol 39:609–639. Fukushima K, Yamanobe T, Shinmei Y, Fukushima J, Kurkin S (2004) Role of the frontal eye fields in smooth-gaze tracking. Prog Brain Res 143:391–401. Fukushima K, Yamanobe T, Shinmei Y, Fukushima J, Kurkin S, Peterson BW (2002) Coding of smooth eye movements in three-dimensional space by frontal cortex. Nature 419:157–162. Fuster JM (1997) The prefrontal cortex: anatomy, physiology, and neuropsychology of the frontal lobe, 2nd ed. Lippincott, Williams & Wilkins. Galaburda AM, Sherman GF, Rosen GD, Aboitiz F, Geschwind N (1985) Developmental dyslexia: four consecutive patients with cortical anomalies. Ann Neurol 18:222–233. Gamlin PD, Clarke RJ (1995) Single-unit activity in the primate nucleus reticularis tegmenti pontis related to vergence and ocular accommodation. J Neurophysiol 73:2115–2119. Gamlin PD, Yoon K (2000) An area for vergence eye movement in primate frontal cortex. Nature 407:1003–1007.

References

225

Gao JH, Parsons LM, Bower JM, Xiong J, Li J, Fox PT (1996) Cerebellum implicated in sensory acquisition and discrimination rather than motor control. Science 272:545–547. Garcia KS, Mauk MD (1998) Pharmacological analysis of cerebellar contributions to the timing and expression of conditioned eyelid responses. Neuropharmacology 37:471–480. Garcia KS, Steele PM, Mauk MD (1999) Cerebellar cortex lesions prevent acquisition of conditioned eyelid responses. J Neurosci 19:10940–10947. Garey LJ (1994) Broadmann’s ‘Localization in the cerebral cortex.’ Translated and edited by Laurence J. Garey. London: Smith-Gorden. Garwicz M, Jorntell H, Ekerot CF (1998) Cutaneous receptive fields and topography of mossy fibres and climbing fibres projecting to cat cerebellar C3 zone. J Physiol (London) 512:277–293. Gauck V, Jaeger D (2000) The control of rate and timing of spikes in the deep cerebellar nuclei by inhibition. J Neurosci 20:3006–3016. Gebhart AL, Petersen SE, Thach WT (2002) Role of the posterolateral cerebellum in language. Ann NY Acad Sci 978:318–333. Gerardin E, Sirigu A, Lehericy S. Poline JB, Gaymard B, Marsault C, Agid Y, Le Bihan D (2000) Partially overlapping neural networks for real and imagined hand movements. Cereb Cortex 10:1093–1104. Gerrits NM, Voogd J (1989) The topographical organization of climbing and mossy fiber afferents in the flocculus and ventral paraflocculus in rabbit, cat and monkey. Exp Brain Res Suppl 17:26–29. Gerrits, NM, Epema AH, van Linge A, Dalm E (1989) The primary vestibulocerebellar projection in the rabbit: absence of primary afferents in the flocculus. Neurosci Let 105:27–33. Geurts FJ, Timmermans J, Shigemoto R, De Schutter E (2001) Morphological and neurochemical differentiation of large granular layer interneurons in the adult rat cerebellum. Neuroscience 104:499–512. Ghelarducci B, Ito M, Yagi N (1975) Impulse discharges from floccular Purkinje cells of alert rabbits during visual stimulation combined with horizontal head rotation. Brain Res 87:66–72. Gilbert PF, Thach WT (1977) Purkinje cell activity during motor learning. Brain Res 128:309–328. Glickstein M (1994) Cerebellar agensis. Brain 117:1209–1212. Glickstein M, Gerrits N, Kralj-Hans I, Mercier B, Stein J, Voogd J (1994) Visual pontocerebellar projections in the macaque. J Comp Neurol 349:51–72. Glickstein M, Strata P, Voogd J (2009) Cerebellum: history. Neuroscience 162:549–559. Glickstein M, Sultan F, Voogd J (2009) Functional localization in the cerebellum. Cortex 47:59–80. Glickstein M, Voogd J (1995) Lodewijk Bolk and the comparative anatomy of the cerebellum. Trend Neurosci 18:206–210. Gómez-Beldarrain M, García-Moncó, JC, Rubio B, Pascual-Leone A (1998) Effect of focal cerebellar lesions on procedural learning in the serial reaction time task. Exp Brain Res 120:25–30. Gomi H, Shidara M, Takemura A, Inoue Y, Kawano K, Kawato M (1998) Temporal firing patterns of Purkinje cells in the cerebellar ventral paraflocculus during ocular following responses in monkeys. I. Simple spikes. J Neurophysiol 80:818–831. Gomi H, Sun W, Finch CE, Itohara S, Yoshimi K, Thompson RF (1999) Learning induces a CDC2related protein kinase, KKIAMRE. J Neurosci 19:9530–9537.

226

the cerebellum

Gonshor A, Melvill-Jones GM (1974) Extreme vestibulo-ocular adaptation induced by prolonged optical reversal of vision. J Physiol (London) 256:381–414. Goto MM, Romero GG, Balaban CD (1997) Transient changes in flocculonodular lobe protein kinase C expression during vestibular compensation. J Neurosci 17:4367–4381. Graf W, Gerrits N, Yatim-Dhiba N, Ugolini G (2002) Mapping the oculomotor system: the power of transneuronal labelling with rabies virus. Eur J Neurosci 15:1557–1562. Graham BP, Dutia MB (2001) Cellular basis of vestibular compensation: analysis and modelling of the role of the commissural inhibitory system. Exp Brain Res 137:387–396. Granit R (1975) The functional role of the muscle spindles—facts and hypotheses. Brain 98:531–556. Granit R, Phillips CG (1956) Excitatory and inhibitory processes acting upon individual Purkinje cells of the cerebellum in cats. J Physiol (London) 133:520–547.5. Graybiel AM (2005) The basal ganglia: learning new tricks and loving it. Cur Opinion Neurobiol 15:638–644. Grey MJ, Ladouceur M, Andersen JB, Nielsen JB, Sinkjær T (2001) Group II muscle afferents probably contribute to the medium latency soleus stretch reflex during walking in humans. J Physiol (London) 534:925–933. Grillner S, Hongo T, Lund S (1971) Convergent effects on a-motoneurons from the vestibulospinal tract and a pathway descending in the medial longitudinal fasciculus. Exp Brain Res 12:457–479. Grillner S, Jessell TM (2009) Measured motion: searching for simplicity in spinal locomotor networks. Cur Opinion Neurobiol 19:572–586. Grillner S, Kozlov A, Dario P, Stefanini C, Menciassi A, Lansner A, Kotaleski JH (2007) Modeling a vertebrate motor system: pattern generation, steering and control of body orientation. Prog Brain Res 165:221–234. Grillner S, Wallen P, Brodin L (1991) Neuronal network generating locomotor behavior in Lamprey: circuitry, transmitter, membrane properties, and simulation. Annu Rev Neurosci 14:169–199. Groenewegen HJ, Voogd J (1977) The parasagittal zonation within the olivocerebellar projection. I. Climbing fiber distribution in the vermis of cat cerebellum. J Comp Neurol 174:417–488. Groenewegen HJ, Voogd J, Freedman SL (1979) The parasagittal zonation within the olivocerebellar projection. II. Climbing fiber distribution in the intermediate and hemispheric parts of cat cerebellum. J Comp Neurol 183:551–601. Guitton D, Buchtel HA, Douglas RM (1985) Frontal lobe lesions in man cause difficulties in suppressing reflexive glances and in generating goal-directed saccades. Exp Brain Res 58:455–472. Gundappa-Sulur G, De Schutter E, Bower JM (1999) Ascending granule cell axon: an important component of cerebellar cortical circuitry. J Comp Neurol 408:580–596. Haggard P, Clark S, Kalogeras J (2002) Voluntary action and conscious awareness. Nat Neurosci 5:382–385. Haines DE, Dietrichs E (1984) An HRP study of hypothalmo-cerebellar and cerebello-hypothalamic connections in squirrel monkey (Saimiri sciureus). J Comp Neurol 229:559-575. Haines DE, Dietrichs E, Sowa TE (1984) Hypothalamo-cerebellar and cerebello-hypothalamic pathways: a review and hypothesis concerning cerebellar circuits which may influence autonomic centers affective behavior. Brain Behav Evol 24:198-220.

References

227

Hall KU, Collins SP, Gamm DM, Massa E, DePaoli-Roach AA, Uhler MD (1999) Phosphorylationdependent inhibition of protein phosphatase-1 by G-substrate. A Purkinje cell substrate of the cyclic GMP-dependent protein kinase. J Biol Chem 274:3485–3495. Hanes DP, Wurtz RH (2001) Interaction of the frontal eye field and superior colliculus for saccade generation. J Neurophysiol 85:804–815. Hansel C, Linden DJ, D’Angelo E (2001) Beyond parallel fiber LTD: the diversity of synaptic and non-synaptic plasticity in the cerebellum. Nat Neurosci 4:467–475. Hantman AW, Jessell TM (2010) Clarke’s column neurons as the focus of a corticospinal corollary circuit. Nat Neurosci 13:1233–1240. Happaney K, Zelazo PD, Stuss DT (2004) Development of orbitofrontal function: current theme and future directions. Brain Cogn 55:1–10. Hartell NA (1994) Induction of cerebellar long-term depression requires activation of glutamate metabotropic receptors. NeuroReport 5:913–916. Hartell NA (1996) Strong activation of parallel fibers produces localized calcium transients and a form of LTD that spreads to distant synapses. Neuron 16:601–610. Hartman MJ, Bower JM (1998) Oscillatory activity in the cerebellar hemispheres of unrestrained rats. J Neurophysiol 80:1598–1604. Haruno M, Wolpert DM, Kawato M (2001) MOSAIC model for sensorimotor learning and control. Neural Computation 13:2201–2220. Harvey RJ, Napper RM (1988) Quantitative study of granule and Purkinje cells in the cerebellar cortex of the rat. J Comp Neurol 274:151–157. Hashikawa T, Nakazawa K, Mikawa S, Shima H, Nagao M (1995) Immunohistochemical localization of protein phosphatase isoforms in the rat cerebellum. Neurosci Res 22:133–136. Hashimoto K, Kano M (2003) Functional differentiation of multiple climbing fiber inputs during synapse elimination in the developing cerebellum. Neuron 38:785–796. Hashimoto M, Mikoshiba K (2004) Neuronal birthdate-specific gene transfer with adenoviral vectors. J Neurosci 24:286–296. Hashimotodani Y, Ohno-Shosaku T, Kano M (2007) Ca2+-assisted receptor-driven endocannabinoid release: mechanism relevant to associate pre- and postsynaptic activities. Cur Opinion Neurobiol 17:360–365. Hashimoto K, Yoshida T, Sakimura K, Mishina M, Watanabe M, Kano M (2009) Influence of parallel fiber–Purkinje cell synapse formation on postnatal development of climbing fiber–Purkinje cell synapses in the cerebellum. Neuroscience 162:601-611. Haslinger B, Erhard P, Altenmüller E, Hennenlotter A, Schwaiger M, Gräfin von Einsiedel H, Rummeny E, Conrad B, Ceballos-Baumann AO (2004) Reduced recruitment of motor association areas during bimanual coordination in concert pianists. Human Brain Map 22:206–215. Haslinger B, Erhard P, Weilke F, Ceballos-Baumann AO, Bartenstein P, Gräfin von Einsiedel H, Schwaiger M, Conrad B, Boecker H (2002) The role of lateral premotor-cerebellar-parietal circuits in motor sequence control: a parametric fMRI study. Brain Res Cogn Brain Res 13:159–168. Hausser M, Clark BA (1997) Tonic synaptic inhibition modulates neuronal output pattern and spatiotemporal synaptic integration. Neuron 19:665–678. Hausser M, Mel B (2003) Dendrites: bug or feature? Cur Opinion Neurobiol 13:372–383.

228

the cerebellum

Hawkes R, Leclerc N (1989) Purkinje cell axon collateral distributions reflect the chemical compartmentation of the rat cerebellar cortex. Brain Res 476:279–290. Hebb O (1949) The organization of behavior. New York: Wiley. Heffner R, Masterton B (1975) Variation in form of the pyramidal tract and its relationship to digital dexterity. Brain Behav Evol 12:161–200. Hemart N, Daniel H, Jaillard D, Crepel F (1995) Receptors and second messengers involved in long-term depression in rat cerebellar slices in vitro: a reappraisal. Eur J Neurosci 7:45–53. Herholz K (2003) PET studies in dementia. Ann Nucl Med 17:79–89. Hesslow G (2002) Conscious thought as simulation of behaviour and perception. Trends Cogn Sci 6:242–247. Higgins DC (1987) The cerebellum and initiation of movement: the stretch reflex. Yale J Biol Med 60:123–131. Highstein SM, Goldberg JM, Moschovakis AK, Fernandez C (1987) Inputs from regularly and irregularly discharging vestibular nerve afferents to secondary neurons in the vestibular nuclei of the squirrel monkey. II. Correlation with output pathways of secondary neurons. J Neurophysiol 58:4719-4738. Hikosaka O & Wurz RH (1985) Modification of saccadic eye movements by GABA-related substances. II. Effects of muscimol in monkey substantia nigra pars reticulata. J Neurophysiol 53:292–308. Hikosaka O, Takikawa Y, Kawagoe R (2000) Role of the basal ganglia in the control of purposive saccadic eye movements. Physiol Rev 80:953–978. Hirai H (2000) Clustering of d glutamate receptors is regulated by the actin cytoskeleton in the dendritic spines of cultured rat Purkinje cells. Eur J Neurosci 12:563–570. Hirai H, Matsuda S (1999) Interaction of the C-terminal domain of d glutamate receptor with spectrin in the dendritic spines of cultured Purkinje cells. Neurosci Res 34:281–287. Hirai H, Launey T, Mikawa S, Torashima T, Yanagihara D, Kasaura T, Miyamoto A, Yuzaki M (2003) New role of ∂2-glutamate receptors in AMPA receptor trafficking and cerebellar function. Nat Neurosci 6:869-876. Hirata Y, Highstein SM (2001) Acute adaptation of the vestibulo ocular reflex: signal processing by floccular and ventral parafloccular Purkinje cells. J Neurophysiol 85:2267–2288. Hirono M, Sugiyama T, Kishimoto Y, Sakai I, Miyazawa T, Kishio M, Inoue H, Nakao K, Ikeda M, Kawahara S, Kirino Y, Katsuki M, Horie H, Ishikawa Y, Yoshioka T (2001a) Phospholipase Cbeta4 and protein kinase Calpha and/or protein kinase CbetaI are involved in the induction of long term depression in cerebellar Purkinje cells. J Biol Chem 276:45236–45242. Hirono M, Yoshioka T, Konishi S (2001b) GABAB receptor activation enhances mGluR-mediated responses at cerebellar excitatory synapses. Nat Neurosci 4:1207–1216. Hochner B, Shomrat T, Fiorito G (2006) The octopus: a model for a comparative analysis of the evolution of learning and memory mechanisms. Biol Bull 210:308–317. Hollerbach JM (1982) Computers, brains and the control of movement. Trends Neurosci 5:189–192. Hongo T, Jankowska E, Lundberg A (1969) The rubrospinal tract. I. Effects on alpha-motoneurons innervating hindlimb muscles in cats. Exp Brain Res 7:44–364. Hongo T, Okada Y (1967) Cortically evoked pre- and postsynaptic inhibition of impulse transmission to the dorsal spinocerebellar tract. Exp Brain Res 3:163–177.

References

229

Hongo T, Okada Y, Sato M (1967) Corticofugal influences on transmission to the dorsal spinocerebellar tract from hindlimb primary afferents. Exp Brain Res 3:135–149. Horikawa J, Suga N (1986) Biosonar signals and cerebellar auditory neurons of the mustached bat. J Neurophysiol 55:1247–1267. Horne MK, Tracey DJ (1979) The afferents and projections of the ventroposterolateral thalamus in the monkey. Exp Brain Res 36:129–141. Hoshi E, Tremblay L, Féger J, Carras PL, Strick PL (2005) The cerebellum communicates with the basal ganglia. Nat Neurosci 8:1491–1493. Houk JC (1979) Regulation of stiffness by skeletomotor reflexes. Annul Rev Physiol 41:99–114. Huang PL, Dawson TM, Bredt DS, Snyder SH, Fishman MC (1993) Targeted disruption of the neuronal nitric oxide synthase gene. Cell 75:1273–1286. Huang YH, Bergles DE (2004) Glutamate transporters bring competition to the synapse. Cur Opinion Neurobiol 14:346-352. Hubel DH, Wiesel TN (1962) Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex. J Physiol (London) 160:106–154. Hulsmann E, Erb M, Grodd W (2003) From will to action: sequential cerebellar contributions to voluntary movement. Neuroimage 20:1485–1492. Hultborn H (2001) State-dependent modulation of sensory feedback. J Physiol (London) 533:5–13. Iacoboni M, Koski LM, Brass M, Bekkering H, Woods RP, Dubeau M-C, Mazziotta JC, Rizzolatti G (2001) Reafferent copies of imitated actions in the right superior temporal cortex. PNAS USA 98:13995–13999. Iacoboni M, Molnar-Szakacs I, Gallese V, Buccino G, Mazziotta JC, Rizzolatti G (2005) Grasping the intentions of others with one’s own mirror neuron system. PLoS Biol 3(3):e79. Iacoboni M, Woods RP, Brass M, Bekkering H, Mazziotta JC, Rizzolatti G (1999) Cortical mechanisms of human imitation. Science 286:2526–2528. Ichise T, Kano M, Hashimoto K, Yanagihara D, Nakao K, Shigemoto R, Katsuki M, Aiba A (2000) mGluR1 in cerebellar Purkinje cells essential for long-term depression, synapse elimination, and motor coordination. Science 288:1832–1835. Iino M (1990) Biphasic Ca2+ dependence of inositol 1,4,5-trisphosphate-induced Ca release in smooth muscle cells of the guinea pig taenia caeci. J Gen Physiol 95:1103–1122. Ilg W, Giese MA, Gizewski ER, Schoch B, Timmann D (2008) The influence of focal cerebellar lesions on the control and adaptation of gait. Brain 131:2913-2927. Illert M, Lundberg A (1978) Collateral connections to the lateral reticular nucleus from cervical propriospinal neurons projecting to forelimb motoneurons in the cat. Neurosci Lett 7:167–172. Illert M, Lundberg A, Padel Y, Tanaka R (1978) Integration in descending motor pathways controlling the forelimb in the cat. 5. Properties of and monosynaptic excitatory convergence on C3–C4 propriospinal neurons. Exp Brain Res 33:101–130. Illert M, Lundberg A, Tanaka R (1977) Integration in descending motor pathways controlling the forelimb in the cat. 3. Convergence on propriospinal neurons transmitting disynaptic excitation from the corticospinal tract and other descending tracts. Exp Brain Res 29:323–346. Imamizu H, Kuroda T, Miyauchi S, Yoshioka T, Kawato M (2003) Modular organization of internal models of tools in the human cerebellum. PNAS USA 100:5461–5466.

230

the cerebellum

Imamizu H, Miyauchi S, Tamada T, Sasaki Y, Takino R, Putz B, Yoshioka T, Kawato M (2000) Human cerebellar activity reflecting an acquired internal model of a new tool. Nature 403:192–195. Inoue T, Kato K, Kohda K, Mikoshiba K (1998) Type 1 inositol 1,4,5-trisphosphate receptor is required for induction of long-term depression in cerebellar Purkinje neurons. J Neurosci 18:53665373. Iriki A (2006) The neural origins and implications of imitation, mirror neurons and tool use. Cur Opinion Neurobiol 16:660-667. Isope P, Barbour B (2002) Properties of unitary granule cell Purkinje cell synapses in adult rat cerebellar slices. J Neurosci 22:9668–9678. Isope P, Dieudonné S, Barbour B (2004) Temporal organization of activity in the cerebellar cortex: a manifesto for synchrony. Ann NY Acad Sci 976:164-174. Isu N, Graf W, Sato H, Kushiro K, Zakir M, Imagawa M, Uchino Y (2000) Sacculo-ocular reflex connectivity in cats. Exp Brain Res 131:262–268. Ito M (1970) Neurophysiological basis of the cerebellar motor control system. Intern J Neurol 7:162–176. Ito M (1972) Neural design of cerebellar motor control system. Brain Res 40:81–84. Ito M (1974) The control mechanisms of cerebellar motor system In: Schmidt D, Worden FG, editors. The neuroscsience, IIIrd Study Program, MIT Press, pp. 293–303. Maasachusetts USA. Ito M (1982) Cerebellar control of the vestibulo-ocular reflex—around the flocculus hypothesis. Annu Rev Neurosci 5:275–296. Ito M (1984) The cerebellum and neural control. New York: Raven Press. Ito M (1986) Neural systems controlling movement. Trends Neurosci 9:515-518. Ito M (1989) Long-term depression. Annu Rev Neurosci 12:85–102. Ito M (1990) A new physiological concept of cerebellum. Rev Neurol (Paris) 146: 564-569. Ito M (1993a) Cerebellar flocculus hypothesis. Nature 363:24–25. Ito M (1993b) Movement and thought: identical control mechanisms by the cerebellum. Trends Neurosci 16:448–450. Ito M (1997) Obituary: John C. Eccles (1903-97) — Neuroscientist who discovered synaptic inhibition. Nature 387:664. Ito M (1998) Cerebellar learning in the vestibulo-ocular reflex. Trends Cog Sci 2:313–321. Ito, M (2000) My years with Sir John Eccles in memoriam: - a tireless Warrior for dualism. H. Eccles and H.-J. Biersack, editors. Ecomed, Landsberg, Germany. pp 29-36. Ito M (2001) Cerebellar long-term depression—characterization, signal transduction and functional roles. Physiol Rev 81:1143–1195. Ito M (2002) The molecular organization of cerebellar long-term depression. Nat Rev Neurosci 3:896 –902. Ito M (2006) Cerebellar circuitry as a neuronal machine. Prog Neurobiol 78:272–303. Ito M (2007) On “How working memory and the cerebellum collaborate to produce creativity and innovation” by L. R. Vandervert, P. H. Schimpf, and H. Liu. Creativity Res J 19:35–38. Ito M (2008) Control of mental activities by internal models in the cerebellum. Nat Rev Neurosci 9:304–313.

References

231

Ito M (2009) Functional roles of neuropeptides in cerebellar circuits. Neuroscience 162:666–672. Ito M, Kano M (1982) Long-lasting depression of parallel fiber-Purkinje cell transmission induced by conjunctive stimulation of parallel fibers and climbing fibers in the cerebellar cortex. Neurosci Lett 33:253–258. Ito M, Kawai N, Udo M, Sato N (1968) Cerebellar-induced disinhibition in dorsal Deiters neurons. Exp Brain Res 6: 247–264. Ito M, Kawai N, Udo M, Sato N (1969) Axon reflex activation of Deiters neurons from the cerebellar cortex through collaterals of the cerebellar afferents. Exp Brain Res 8: 249–268. Ito M, Miyashita Y (1975) The effect of chronic destruction of the inferior olive upon visual modification of the horizontal vestibulo-ocular reflex of rabbits. Proc Jpn Acad 51:716–720. Ito M, Nisimaru N, Yamamoto M (1977) Specific patterns of neuronal connexons involved in the control of the rabbit’s vestibulo-ocular reflexes by the cerebellar flocculus. J Physiol (London) 265:833–854. Ito M, Nisimaru N, Shibuki K (1979) Destruction of inferior olive induces rapid depression in synaptic action of cerebellar Purkinje cells. Nature 277:568-569. Ito M, Sakurai M, Tongroach P (1981) Evidence for modifiability of parallel fiber-Purkinje cell synapses. In: Szentagothai J, Hamori J, Palkovitz M, editors. Adv Physiol Sci Vol 2 Regulatory Functions of the CNS. Subsystems. pp. 97–105. Ito M, Sakurai M, Tongroach P (1982) Climbing fibre induced depression of both mossy fibre responsiveness and glutamate sensitivity of cerebellar Purkinje cells. J Physiol (London) 324:113–134. Ito M, Shiida N, Yagi N, Yamamoto M (1974) The cerebellar modification of rabbit’s horizontal vestibulo-ocular reflex induced by sustained head rotation combined with visual stimulation. Proc Jpn Acad 50:85–89. Ito M, Udo M, Mano N, Kawai N (1970) Synaptic action of the fastigio-bulbar impulses upon neurons in the medullary reticular formation and vestibular nuclei. Exp Brain Res 11:29–47. Ito M, Yoshida M (1964) The cerebellar-evoked monosynaptic inhibition of Deiters’ neurons. Experientia 20:515–516. Classics (2007) Cerebellum 6:103–104. Ito M, Yoshida M (1966) The origin of cerebellar-induced inhibition of Deiters neurons I. Monosynaptic initiation of the inhibitory postsynaptic potentials. Exp Brain Res 2:330–349. Ito M, Yoshida M, Obata K (1964) Monosynaptic inhibition of the intracerebellar nuclei induced from the cerebellar cortex. Experientia 20:575–576. Ito S, Yuasa H, Luo ZW, Ito M, Yanagihara D (1999) Adaptive locomotion to periodic perturbation, adaptation mechanism with coupling of oscillator and link dynamics. Artif Life Robotics 3:97–101. Ivry RB, Keele SW (1989) Timing functions of the cerebellum. J Cognit Neurosci 1:136–152. Jaarsma D, Diño MR, Ohishi H, Shigemoto R, Mugnaini E (1998) Metabotropic glutamate receptors are associated with non-synaptic appendages of unipolar brush cells in rat cerebellar cortex and cochlear nuclear complex. J Neurocytol 27:303–327. Jaarsma D, Ruigrok TJ, Caffe R, Cozzari C, Levey AI, Mugnaini E, Voogd J (1997) Cholinergic innervation and receptors in the cerebellum. Prog Brain Res 114:67–96. Jaarsma D, Wenthold RJ, Mugnaini E (1995) Glutamate receptor subunits at mossy fiber-unipolar brush cell synapses: light and electron microscopic immunocytochemical study in cerebellar cortex of rat and cat. J Comp Neurol 357:145–160.

232

the cerebellum

Jaeger CB, Kapoor R, Llinás R (1988) Cytology and organization of rat cerebellar organ cultures. Neuroscience 26:509–538. Jaeger D, Bower JM (1999) Synaptic control of spiking in cerebellar Purkinje cells: dynamic current clamp based on model conductances. J Neurosci 16:6090–6101. Jakab RL, Hámori J (1988) Quantitative morphology and synaptology of cerebellar glomeruli in the rat. Anatomy and Embryology. 179:81-88. Jankowska E, Jukes MGM, Lund S, Lundberg A (1967a) The effect of DOPA on the spinal cord. 5. Reciprocal organization of pathways transmitting excitatory action to alpha motoneurons of flexors and extensors. Acta Physiol Scand 70:369–388. Jankowska E, Jukes MGM, Lund S, Lundberg A (1967b) The effect of DOPA on the spinal cord. 6. Half-centre organization of interneurons transmitting effects from the flexor reflex afferents. Acta Physiol Scand 70:389–402. Jankowska E, Krutki P, Hammar I (2010) Collateral actions of premotor interneurons on ventral spinocerebellar tract neurons in the cat. J Neurophysiol 104:1872-1883. Jansen J, Brodal A (1954) Aspects of cerebellar anatomy. Oslo: Forlagt Johan Grundt Tanum. Janssen G, Messing-Junger AM, Engelbrecht V, Gobel U, Bock WJ, Lenard HG (1998) Cerebellar mutism syndrome. Klin Padiatr 210:243–247. Jeannerod M (1994) The representing brain: neural correlates of motor intention and imagery. Behav Brain Sci 17:187–202. Jeannerod M, Frank V (1999) Mental imaging of motor activity in humans. Cur Opinion in Neurobiol 9:735–739. Jeromin A, Huganir RL, Linden DJ (1996) Suppression of the glutamate receptor delta 2 subunit produces a specific impairment in cerebellar long-term depression. J Neurophysiol 76:3578-3583. Jirenhed D-A, Bengtsson F, Hesslow G (2007) Acquisition, Extinction, and Reacquisition of a Cerebellar Cortical Memory Trace. J Neurosci 27:2493-2502. Johansson RS, Westling G (1987) Signals in tactile afferents from the fingers eliciting adaptive motor responses during precision grip. Exp Brain Res 66:141–154. Johnson-Frey SH (2004) The neural bases of complex tool use in humans. Trends Cog Sci 8:71–78. Johnson-Laird PN (1983) Mental models. Cambridge University Press Cambridge UK. Jörntell H, Ekerot CF (2002) Reciprocal bidirectional plasticity of parallel fiber receptive fields in cerebellar Purkinje cells and their afferent interneurons. Neuron 34:797–806. Jörntell H, Ekerot CF (2003) Receptive field plasticity profoundly alters the cutaneous parallel fiber synaptic input to cerebellar interneurons in vivo. J Neurosci 22:9620–9631. Jörntell H, Ekerot CF (2006) Properties of somatosensory synaptic integration in cerebellar granule cells in vivo. J Neurosci 26:11786-11797. Kakegawa W, Yuzaki M (2005) A mechanism underlying AMPA receptor trafficking during cerebellar long-term potentiation. PNAS USA 102:17846–17851. Kalinichenko SG, Okhotin VE (2005) Unipolar brush cells—a new type of excitatory interneuron in the cerebellar cortex and cochlear nuclei of the brainstem. Neurosci Behav Physiol 35:21–36. Kandel ER (2009) The Biology of Memory: A Forty-Year Perspective. J Neurosci 29:12748-12756 Kaneko CRS (2006) Saccade-related, long-lead burst neurons in the monkey rostral pons. J Neurophysiol 95:979–994.

References

233

Kano M, Hashimoto K, Kurihara H, Watanabe M, Inoue Y, Aiba A, Tonegawa S. (1997) Persistent multiple climbing fiber innervation of cerebellar Purkinje cells in mice lacking mGluR1. Neuron 18:71–79. Kano M, Iino K, Maekawa K, Kano M-S (1991) Optokinetic response of cells in the nucleus reticularis tegmenti pontis of the pigmented rabbit. Exp Brain Res 87:239–244. Kano M, Kato M (1987) Quisqualate receptors are specifically involved in cerebellar synaptic plasticity. Nature 325:276–279. Kano M, Rexhausen U, Dreessen J, Konnerth A (1992) Synaptic excitation produces a long-lasting rebound potentiation of inhibitory synaptic signals in cerebellar Purkinje cells. Nature 356:601–604. Karachot L, Kado RT, Ito M (1994) Stimulus parameters for induction of long-term depression in in vitro slices of rat cerebellum. Neurosci Res 21:161–168. Karachot L, Shirai Y, Vigot R, Yamamori T, Ito M (2001) Induction of long-term depression in cerebellar Purkinje cells requires a quickly turned over protein. J Neurophysiol 86:280–289. Karakossian MH, Otis TS (2004) Excitation of cerebellar interneurons by group I metabotropic glutamate receptors. J Neurophysiol 92:1558–1565. Kardamakis AA, Moschovakis AK (2009) Optimal control of gaze shifts. J Neurosci 29: 7723-7730. Kasono K, Hirano T (1995) Involvement of inositol trisphosphate in cerebellar long-term depression. NeuroReport 6:569-572. Kassardjian CD, Tan Y-F, Chung J-Y, Heskin R, Peterson MJ, Broussard DM (2005) The site of a memory shifts with consolidation. J Neurosci 25:7979–7985. Katoh A, Kitazawa H, Itohara S, Nagao S (1998) Dynamic characteristics and adaptability of mouse vestibulo-ocular and optokinetic response eye movements and the role of the flocculo-olivary system revealed by chemical lesions. PNAS USA. 95:7705–7710. Katoh A, Kitazawa H, Itohara S, Nagao S (2000) Inhibition of nitric oxide synthesis and gene knockout of neuronal nitric oxide synthase impaired adaptation of mouse optokinetic response eye movements. Learn Mem 7:220–226. Kawaguchi SY, Hirano T (2002) Signaling cascade regulating long-term potentiation of GABA(A) receptor responsiveness in cerebellar Purkinje neurons. J Neurosci 22:3969–3976. Kawaguchi Y (1985) Two groups of secondary vestibular neurons mediating horizontal canal signals, probably to the ipsilateral medial rectus muscle, under inhibitory influences from the cerebellar flocculus in rabbits. Neurosci Res 2:434–446. Kawano K, Miles FA (1896) Short-latency ocular following responses of monkey. II. Dependence on a prior saccadic eye movement. J Neurophysiol 56:1355–1380. Kawano K, Shidara M, Watanabe Y, Yamane S (1994) Neural activity in cortical area MST of alert monkey during ocular following responses. J Neurophysiol 71:2305–2324. Kawato M (1999) Internal models for motor control and trajectory planning. Cur Opinion Neurobiol 9:718–727. Kawato M, Furukawa K, Suzuki R (1987) A hierarchical neural-network model for control and learning of voluntary movement. Biol Cybern 57:169–185. Kawato M, Kuroda T, Imamizu H, Nakano E, Miyauchi S, Yoshioka T (2003) Internal forward models in the cerebellum: fMRI study on grip force and load force coupling. Prog Brain Res 142:171–188. Kazantsev VB, Nekorkin VI, Makarenko VI, Llinás R (2004) Self-referential phase reset based on inferior olive oscillator dynamics. PNAS USA 101:18183–18188.

234

the cerebellum

Keating JG, Thach WT (1995) Nonclock behavior of inferior olive neurons: interspike interval of Purkinje cell complex spike discharge in the awake behaving monkey is random. J Neurophysiol 73:1329–1340. Keller EL, Heinen SJ (1991) Generation of smooth-pursuit eye movements: neuronal mechanisms and pathways. Neurosci Res 79-107. Kellett DO, Fukunaga I, Chen-Kubota E, Dean P, Yeo CH (2010) Memory consolidation in the cerebellar cortex. PLoS ONE 5:e11737. Kelly RM, Strick PL (2003) Cerebellar loops with motor cortex and prefrontal cortex of a nonhuman primate. J Neurosci 23:8432–8444. Kenyon GT (1997) A model of long-term memory storage in the cerebellar cortex: a possible role for plasticity at parallel fiber synapses onto stellate/basket interneurons. PNAS USA 94:14200–14205. Kenyon GT, Medina JF, MD Mauk (1998a) A mathematical model of the cerebellar-olivary system I: self-regulating equilibrium of climbing fiber activity. J Comput Neurosci 5:71–90. Kenyon GT, Medina JF, MD Mauk (1998b) A mathematical model of the cerebellar-olivary system II. Motor adaptation through systematic disruption of climbing fiber equilibrium. J Comput Neurosci. 5:71–90. Kevetter GA, Hoffman RD (1991) Excitatory amino acid immunoreactivity in vestibulo-ocular neurons in gerbils. Brain Res 554:348–351. Khodakhah K, Armstrong CM (1997) Induction of long-term depression and rebound potentiation by inositol trisphosphate in cerebellar Purkinje neurons. PNAS USA 94:14009-14014. Kielan-Jaworowska Z, Lancaster TE (2004) A new reconstruction of multituberculate endocranial casts and encephalization quotient of Kryptobaatar. Acta Palaeontologica Polonica 49:177–188. Kim JJ, Krupa DJ, Thompson RF (1998) Inhibitory cerebello-olivary projections and blocking effect in classical conditioning. Science 279:570–573. Kimura T, Sugimori M, Llinás RR (2005) Purkinje cell long-term depression is prevented by T-588, a neuroprotective compound that reduces cytosolic calcium release from intracellular stores. Proc Natl Acad Sci USA 102:17160–17165. Kimura T, Sugimori M, Llinás RR (2005) Purkinje cell long-term depression is prevented by T-588, a neuroprotective compound that reduces cytosolic calcium release from intracellular stores. PNAS USA 102:17160-17165. King JS, Cummings SL, Bishop GA (1992) Peptides in cerebellar circuit. Prog Neurobiol 39:423-442. Kishimoto Y, Kawahara S, Fujimichi R, Mori H, Mishina M, Kirino Y (2001) Impairment of eyeblink conditioning in GluRdelta2-mutant mice depends on the temporal overlap between conditioned and unconditioned stimuli. Eur J Neurosci 14:1515–1521. Kishimoto Y, Fujimichi R, Araishi K, Kawahara S, Kano M, AibA A, Kirino Y (2002) mGluR1 in cerebellar Purkinje cells is required for normal association of temporally contiguous stimuli in classical conditioning. Eur J Neurosci 16:2416–2424. Kitazawa S, Kimura T, Yin PB (1998) Cerebellar complex spikes encode both destinations and errors in arm movements. Nature 392:494–497. Kitazawa S, Wolpert DM (2005) Rhythmicity, randomness and synchrony in climbing fiber signals. Trends Neurosci 28:611–619.

References

235

Kitazawa S, Yin P-B (2002) Prism adaptation with delayed visual error signals in the monkey. Exp Brain Res 144:258–261. Kleim JA, Freeman JH, Bruneau R, Nolan BC, Cooper NR, Zook A (2002) Synapse formation is associated with memory storage in the cerebellum. PNAS USA 99:13228–13231. Kobayashi Y, Kawano K, Takemura A, Inoue Y, Kitama T, Gomi H, Kawato M (1998) Temporal firing patterns of Purkinje cells in cerebellar ventral paraflocculus during ocular following responses in monkeys. II. Complex spikes. J Neurophysiol 80:832–848. Koch G, Oliveri M, Torriero S, Salerno S, Gerfo EL, Caltagirone C (2006) Repetitive TMS of cerebellum interferes with millisecond time processing. Exp Brain Res 179:291–299. Koekkoek SK, Hulscher HC, Dortland BR, Hensbroek RA, Elgersma Y, Ruigrok TJ, De Zeeuw CI (2003) Cerebellar LTD and learning-dependent timing of conditioned eyelid responses. Science 301:1736–1739. Koekkoek SK, Yamaguchi K, Milojkovic BA, Dortland BR, Ruigrok TJ, Maex R, De Graaf, W, Smit AE, Vanderwerf F, Bakker CE, Willemsen R, Ikeda T, Kakizawa S, Onodera K, Nelson DL, Mientjes E, Joosten M, De Schutter E, Oostra BA, Ito M, De Zeeuw CI (2005) Deletion of FMR1 in Purkinje cells enhances parallel fiber LTD, enlarges spines, and attenuates cerebellar eyelid conditioning in fragile X syndrome. Neuron 47:339–352. Kojima I, Kitaoka M, Ogata E (1990) Insulin-like growth factor-I stimulates diacylglycerol production via multiple pathways in Balb/c 3T3 cells. J Biol Chem 265:16846–16850. Kojima Y, Yoshida K, Iwamoto Y (2007) Microstimulation of the midbrain tegmentum creates learning signals for saccade adaptation. J Neurosci 27:3759–3767. Kojima Y, Yoshida K, Iwamoto Y (2007) Microstimulation of the midbrain tegmentum creates learning signals for saccade adaptation. J Neurosci 27:3759-3767. Kondo S, Marty A (1998) Synaptic currents at individual connections among stellate cells in rat cerebellar slices. J Physiol (London) 509:221–232. Korte GE, Mugnaini E (1979) The cerebellar projection of the vestibular nerve in the cat. J Comp Neurol 184:265–277. Krauzlis RJ, Lisberger SG (1994) Simple spike responses of gaze velocity Purkinje cells in the floccular lobe of the monkey during the onset and offset of pursuit eye movements. J Neurophysiol 72:2045–2050. Kremer S, Chassagnon S, Hoffmann D, Benabid AL, Kahane P (2001) The cingulate hidden hand. J Neurol Neurosurg Psychiatry 70:264–265. Krienen FM, Buckner RL (2009) Segregated fronto-cerebellar circuits revealed by intrinsic functional connectivity. Cereb Cortex 19:2485–2497. Krutki P, Jankowska E, Edgley SA (2003) Are crossed actions of reticulospinal and vestibulospinal neurons on feline motoneurons mediated by the same or separate commissural neurons? J Neurosci 23:8041–8050. Kuroda S, Schweighofer N, Kawato M (2001) Exploration of signal transduction pathways in cerebellar long-term depression by kinetic simulation. J Neurosci 21:5693–5702. Lachamp P, Balland B, Tell F, Baude A, Strube C, Crest M, Kessler J-P (2005) Early expression of AMPA receptors and lack of NMDA receptors in developing rat climbing fibre synapses. J Physiol (London) 564:751–763. Lachamp PM, Liu Y, Liu SJ (2009) Glutamatergic modulation of cerebellar interneuron activity is mediated by an enhancement of GABA release and requires protein kinase a/rim1 signaling. J Neurosci 29:381–392.

236

the cerebellum

Laine J, Axelrad H (2002) Extending the cerebellar Lugaro cell class. Neuroscience 115:363–374. Landsend AS, Amiry-Moghaddam M, Matsubara A, Bergersen L, Usami S, Wenthold RJ, Ottersen OP (1997) Differential localization of δ glutamate receptors in the rat cerebellum: coexpression with ampa receptors in parallel fiber–spine synapses and absence from climbing fiber–spine synapses. J Neurosci 17:834-842. Larsell O (1970) Comparative anatomy and histology of the cerebellum. Minneapolis: University of Minnesota Press. Launey T (2007) A computational approach to the study of AMPA receptor declustering at Purkinje cell synapses. Archives Italiennes de Biologie 145:99–310. Launey T, Endo S, Sakai R, Harano J, Ito M (2004) Protein phosphatase 2A inhibition induces cerebellar long-term depression and declustering of synaptic AMPA receptor. PNAS USA 101:676–681. Le TD, Shirai Y, Okamoto T, Tatsukawa T, Nagao S, Shimizu T, Ito M (2010) Lipid signaling in cytosolic phospholipase A2α cyclooxygenase-2 cascade mediates cerebellar long-term depression and motor learning. PNAS USA 107:3198-3203. Leclerc N, Schwarting GA, Herrup K, Hawkes R, Yamamoto M (1992) Compartmentation in mammalian cerebellum: zebrin II and P-path antibodies define three classes of sagittally organized bands of Purkinje cells. PNAS USA 89:5006–5010. Lee M, Martin C, Graham A, Court J, Jaros E, Perry R, Iversen P, Bauman M, Perry E (2002) Nicotinic receptor abnormalities in the cerebellar cortex in autism. Brain 125:1483–1495. Leggio MG, Silveri MC, Petrosini L, Molinari M (2000) Phonological grouping is specifically affected in cerebellar patients; a verbal fluency study. J Neurol Neurosurg Psychiatr 69:102–106. Leiner HC (2010) Solving the mystery of the human cerebellum. Neuropsychol Rev 3:229–235. Leiner HC, Leiner AL, Dow RS (1986) Does the cerebellum contribute to mental skills? Behav Neurosci 100:443–454. Leiner HC, Leiner AL, Dow RS (1987) Cerebro-cerebeUar learning loops in apes and humans. Ital J Neurol Sci 8:425-436. Leiner HC, Leiner AL, Dow RS (1993) Cognitive and language functions of the human cerebellum. Trends Neurosci 16:444–447. Lemon RN (2008) Descending pathways in motor control. Annu Rev Neurosci 31:195–218. Lévénès C, Daniel H, Soubrié P, Crépel F (1998) Cannabinoids decrease excitatory synaptic transmission and impair long-term depression in rat cerebellar Purkinje cells. J Physiol (London) 510:867–879. Levine JM, Card JP (1987) Light and electron microscopic localization of a cell surface antigen (NG2) in the rat cerebellum: association with smooth protoplasmic astrocytes. J Neurosci 7:2711-2720. Lev-Ram V, Jiang T, Wood J, Lawrence DS, Tsien RY (1997a) Synergies and coincidence requirements between NO, cGMP and Ca2+ in the induction of cerebellar long-term depression. Neuron 18:1025–1038. Lev-Ram V, Mehta SB, Kleinfeld D, Tsien RY (2003) Reversing cerebellar long-term depression. PNAS USA 100:15989–15993. Lev-Ram V, Nebyelul Z, Ellisman MH, Huang PL, Tsien RY (1997b) Absence of cerebellar longterm depression in mice lacking neuronal nitric oxide synthase. Learn Mem 3:169–177. Lev-Ram V, Wong ST, Storm DR, Tsien RY (2002) A new form of cerebellar long-term potentiation is postsynaptic and depends on nitric oxide but not cAMP. PNAS USA 99:8389–8393.

References

237

Levy JB, Canoll PD, Silvennoinen O, Barnea G, Morse B, Honegger AM, Huang JT, Cannizzaro LA, Park SH, Druck T, Huebner K, Sap J, Ehrlich M, Musacchiot JM, Schlessinger J (1993) The cloning of a receptor-type protein tyrosine phosphatase expressed in the central nervous system. J Biol Chem 268:10573–10581. Leyton M (1987) Nested structures of control: an intuitive view. Computer Vision, Graphics, and Image Processing 37:20–53. Lim R, Callister RJ, Brichta AM (2009) An increase in glycinergic quantal amplitude and frequency during early vestibular compensation in mouse. J Neurophysiol 103:16–24. Lin S-C, Huck JHJ, Roberts JDB, Macklin WB, Somogyi P, Bergles DE (2005) Climbing fiber innervation of NG2-expressing glia in the mammalian cerebellum. Neuron 46:773–785. Lincoln JS, McCormick DA, Thompson RF (1982) Ipsilateral cerebellar lesions prevent learning of the classically conditioned nictitating membrane/eyelid response. Brain Res 242:190–193. Linden DJ (1994) Input-specific induction of cerebellar long-term depression does not require presynaptic alteration. Learn Mem 1:121–128. Linden DJ (1996) A protein synthesis-dependent late phase of cerebellar long-term depression. Neuron 17:3483–3490. Linden DJ (2001) The expression of cerebellar LTD in culture is not associated with changes in AMPA-receptor kinetics, agonist affinity, or unitary conductance. PNAS USA 98:14066–14071. Linden DJ, Connor JA (1991) Long-term depression of glutamate currents in cultured cerebellar Purkinje neurons does not require nitric oxide signaling. Eur J Neurosci 4:10–15. Linden DJ, Dawson TM, Dawson VL (1995) An evaluation of the nitric oxide/cGMP/cGMPdependent protein kinase cascade in the induction of cerebellar long-term depression in culture. J Neurosci 15:5098–5105. Linden DJ, Dickinson MH, Smeyne M, Connor JA (1991) A long-term depression of AMPA currents in cultured cerebellar Purkinje neurons. Neuron 7:81–89. Lippman J, Dunaevsky A (2005) Dendritic spine morphogenesis and plasticity. J Neurobiol 64:47–57. Lisberger SG (2009) Internal models of eye movement in the floccular complex of the monkey cerebellum. Neuroscience 162:763–776. Lisberger SG, Fuchs AF (1978) Role of primate flocculus during rapid behavioral modification of vestibulo-ocular reflex. II. Mossy fiber firing patterns during horizontal head rotation and eye movement. J Neurophysiol 41:764–777. Lisberger SG, Morris EJ, Tychsen L (1987) Visual motion processing and sensory-motor integration for smooth pursuit eye movements. Annu Rev Neurosci 10:97–129. Lisberger SG, Sejnowski TJ (1992) Motor learning in a recurrent network model based on the vestibulo-ocular reflex. Nature 360:159–161. Lisberger SG, Sejnowski TJ (1993) Cerebellar flocculus hypothesis—reply. Nature 363:25. Lisberger SG (1994) Neural basis for motor learning in the vestibuloocular reflex of primates. III. Computational and behavioral analysis of the sites of learning. J Neurophysiol 72:974–998. Llinás R, Baker R, Sotelo C (1974) Electronic coupling between neurons in cat inferior olive. J Neurophysiol 37:560–571. Llinás R, Sugimori M (1980a) Electrophysiological properties of in vitro Purkinje cell somata in mammalian cerebellar slices. J Physiol (London) 305:171–195.

238

the cerebellum

Llinás R, Sugimori M (1980b) Electrophysiological properties of in vitro Purkinje cell dendrites in mammalian cerebellar slices. J Physiol (London) 305:197–213. Llinás R, Sugimori M, Lin JW, Cherksey B (1989) Blocking and isolation of a calcium channel from neurons in mammals and cephalopods utilizing a toxin fraction (FTX) from funnel-web spider poison. PNAS USA 86:1689–1693. Llinás R, Yarom Y (1986) Oscillatory properties of guinea-pig inferior olivary neurones and their pharmacological modulation: an in vitro study J Physiol (London) 376:163-182. Loeb GE (1984) The control and responses of mammalian muscle spindles during normally executed motor tasks. In: Terjung RL, editor. Exercise and sports sciences reviews, Vol. 12,. Lexington (MA): D.C. Heath and Company, pp. 157–204. Loeb GE, Levine WS, He J (1990) Understanding Sensorimotor Feedback through Optimal Control. Cold Spring Harb Symp Quant Biol 55:791-803. Lonart G, Schoch S, Kaeser PS, Larkin CJ, Südhof TC, Linden DJ (2003) Phosphorylation of RIM1α by PKA triggers presynaptic long-term potentiation at cerebellar parallel fiber synapses. Cell 115:149–160. López-Muñoz F, Boya J, Alamo C (2006) Neuron theory, the cornerstone of neuroscience, on the centenary of the Nobel Prize award to Santiago Ramón y Cajal. Brain Res Bull 70:391–405. Loram ID, Lakie M (2002). Direct measurement of human ankle stiffness during quiet standing: the intrinsic mechanical stiffness is insufficient for stability. J Physiol (London) 545:1041–1053. Loram ID, Maganaris CN, Lakie M (2005) Human postural sway results from frequent, ballistic bias impulses by soleus and gastrocnemius. J Physiol (London) 564:295–311. Lou JS, Bloedel JR (1992) Responses of sagittally aligned Purkinje cells during perturbed locomotion: synchronous activation of climbing fiber inputs. J Neurophysiol 68:570–580. Lu H, Esquivel AV, Bower JM (2009) 3D electron microscopic reconstruction of segments of rat cerebellar Purkinje cell dendrites receiving ascending and parallel fiber granule cell synaptic inputs. J Comp Neurol 514:583–594. Luciani L (1891) cited from Dow and Moruzzi (1958) The physiology and pathology of the cerebellum. Lundberg A (1999) Descending control of forelimb movements in the cat. Brain Res Bull 50:323–324. Maass W, Natschläger T, Markram H (2002) Real-time computing without stable states: a new framework for neural computation based on perturbations. Neural Computation 14:2531–2560. Maejima T, Hashimoto K, Yoshida T, Aiba A, Kano M (2001) Presynaptic inhibition caused by retrograde signal from metabotropic glutamate to cannabinoid receptors. Neuron 31:463–475. Maekawa K, Simpson JI (1973) Climbing fiber responses evoked in vestibulocerebellum of rabbit from visual pathway. J Neurophysiol 36:649–666. Maex R, De Schutter E (2005) Oscillations in the cerebellar cortex: a prediction of their frequency bands. Prog Brain Res 148:181–188. Maffei A, Prestori F, Shibuki K, Rossi P, Taglietti V, D’Angelo E (2003) NO enhances presynaptic currents during cerebellar mossy fiber-granule cell LTP. J Neurophysiol 90:2478–2483. Mailleux P, Mitchell F, Vanderhaeghen JJ, Milligan G, Erneux C (1992) Immunohistochemical distribution of neurons containing the G-proteins alpha/G11 alpha in the adult rat brain. Neuroscience 51:311–316.

References

239

Malomo AO, Idowu OE, Osuagwu FC (2006) Lessons from history: human anatomy, from the origin to the Renaissance. Int J Morphol 24:99–104. Manni E, Petrosini L (2004) A century of cerebellar somatotopy: a debated representation Nat Rev Neurosci 5:241–249. Mann-Metzer P, Yarom Y (1999) Electrotonic coupling interacts with intrinsic properties to generate synchronized activity in cerebellar networks of inhibitory interneurons. J Neurosci 19:3298–3306. Mano N, Ito Y, Shibutani H (1991) Saccade-related Purkinje cells in the cerebellar hemispheres of the monkey. Exp Brain Res 84:465–470. Manto M-U, Pandolfo M (2001) The cerebellum and its disorders. Cambridge University Press Cambridge UK. Maravita A, Iriki A (2004) Tools for the body (schema). Trends Cog Sci 8:79–86. Marder E, Hooper SL, Siwicki KK (1986) Modulatory action and distribution of the neuropeptide proctolin in the crustacean stomatogastric nervous system. J Comp Neurol 243:454–467. Mariani J, Changeux JP (1981) Ontogenesis of olivocerebellar relationships. I. Studies by intracellular recordings of the multiple innervation of Purkinje cells by climbing fibers in the developing rat cerebellum. J Neurosci 1:696–702. Markert G, Büttner U, Straube A, Boyle R (1988) Neuronal activity in the flocculus of the alert monkey during sinusoidal optokinetic stimulation. Exp Brain Res 70:134–144. Marr D (1969) A theory of cerebellar cortex. J Physiol (London) 202:437–470. Martin TA, Keating JG, Goodkin HP, Bastian AJ, Thach WT (1996a) Throwing while looking through prisms. I. Focal olivocerebellar lesions impair adaptation. Brain 119:1183–1198. Martin TA, Keating JG, Goodkin HP, Bastian AJ, Thach WT (1996b) Throwing while looking through prisms. II. Specificity and storage of multiple gaze-throw calibrations. Brain 119:1199–1211. Masuda N, Amari S (2005) Modeling memory transfer and savings in cerebellar motor learning. Proceedings of NIPS’2005. Mathy A, Ho SSN, Davie JT, Duguid IC, Clark BA, Häusser M (2009) Encoding of Oscillations by Axonal Bursts in Inferior Olive Neurons. Neuron 62:388–399. Matsuda S, Launey T, Mikawa S, Hirai H (2000) Disruption of AMPA receptor GluR2 clusters following long-term depression induction in cerebellar Purkinje neurons. EMBO J 19:2765–2774. Matsuda K, Miura E, Miyazaki T, Kakegawa W, Emi K, Narumi S, Fukazawa Y, Ito-Ishida A, Kondo T, Shigemoto R, Watanabe M, Yuzaki M (2010) Cbln1 is a ligand for an orphan glutamate receptor δ2, a bidirectional synapse organizer. Science 328:363-368. Matsui K, Jahr CE (2003) Ectopic Release of Synaptic Vesicles. Neuron 40:1173-1183. Matsui K, Jahr CE (2004) Differential control of synaptic and ectopic vesicular release of glutamate. J Neurosci 24:8932–8939. Matsui K, Jahr CE, Rubio ME (2005) High-Concentration Rapid Transients of Glutamate Mediate Neural-Glial Communication via Ectopic Release. J Neurosci 25:7538-7547. Matsushita M, Yaginuma H, Tanami T (1992) Somatotopic termination of the spino-olivary fibers in the cat, studied with the wheat germ agglutinin-horseradish peroxidase technique. Exp Brain Res 89:397–407.

240

the cerebellum

Matthews PBC (2010) Muscle spindles: their messages and their fusimotor supply. [Internet] DOI:10.1002/cphy.cp010206, Wiley Published Online: [cited 1 Jun 2010]. Mazzarello P (1999) A unifying concept: the history of cell theory. Nature Cell Biol 1:E13–E15. McCarthy J, Minsky ML, Rochester N, Shannon CE (1955) A proposal for the Dartmouth Summer Research Project on artificial intelligence. [Internet] Available from: http://www-formal.stanford. edu/jmc/history/dartmouth/dartmouth.html: [cited 1 Jun 2010]. McCrea DA, Rybak IA (2008) Organization of mammalian locomotor rhythm and pattern generation. Brain Res Rev 57:134–146. McElligott JG, Beeton P, Polk J (1998) Effect of cerebellar inactivation by lidocaine microdialysis on the vestibuloocular reflex in goldfish. J Neurophysiol 79:1286–1294. McElvain LE, Bagnall MW, Sakatos A, du Lac S (2010) Bidirectional plasticity gated by hyperpolarization controls the gain of postsynaptic firing responses at central vestibular nerve synapses. Neuron 68:763–775. McKay BE, Engbers JD, Mehaffey WH, Gordon GR, Molineux ML, Bains JS, Turner RW (2007) Climbing fiber discharge regulates cerebellar functions by controlling the intrinsic characteristics of Purkinje cell output. J Neurophysiol 97:2590–2604. McLaughlin SC (1967) Parametric adjustment in saccadic eye movements. Percept Psychophys 2:359 –362. Medina JF, Lisberger SG (2008) Links from complex spikes to local plasticity and motor learning in the cerebellum of awake-behaving monkeys. Nature Neurosci 11:1185–1192. Medina JF, Mauk MD (1999) Simulations of cerebellar motor learning: computational analysis of plasticity at the mossy fiber to deep nucleus synapse. J Neurosci 19:7140-7151. Medina JF, Nores WL, Mauk MD (2002) Inhibition of climbing fibres is a signal for the extinction of conditioned eyelid responses. Nature 416:330–333. Medina JF, Nores WL, Ohyama T, Mauk MD (2000) Mechanisms of cerebellar learning suggested by eyelid conditioning. Cur Opinion Neurobiol 10:717–724. Meek J, Yang JY, Han VZ, Bell CC (2008) Morphological analysis of the mormyrid cerebellum using immunohistochemistry, with emphasis on the unusual neuronal organization of the valvula. J Comp Neurol 510: 296-421. Melzack R (1990) Phantom limbs and the concept of a neuromatrix. Trends Neurosci 13:88–92. Merton PA (1953) Speculations on the servo-control of movement. In Wlostenholme GEW, editor. The spinal cord. London: Churchill, pp. 247–255. Miall RC (2003) Connecting mirror neurons and forward models. NeuroReport 14:2135–2137. Miall RC, Christensen LOD, Cain O, Stanley J (2007) Disruption of state estimation in the human lateral cerebellum. PLoS Biol 5:e316. Miall RC, Weir DJ, Wolpert DM, Stein JF (1993) Is the cerebellum a Smith predictor? J Motor Behav 25:203–216. Mihailoff GA (1993) Cerebellar nuclear projections from the basilar pontine nuclei and nucleus reticularis tegmenti pontis as demonstrated with PHA-L tracing in the rat. J Comp Neurol 330:130–146. Miles FA, Kawano K (1986) Short-latency ocular following responses of monkey. III. Plasticity. J Neurophysiol 56:1381–1396.

References

241

Miles FA, Kawano K, Optican LM (1986) Short-latency ocular following responses of monkey. I. Dependence on temporospatial properties of visual input. J Neurophysiol 56:1321–1354. Miles FA, Lisberger SG (1981) Plasticity in the vestibuloocular reflex. A new hypothesis. Annu Rev Neurosci 4:273–299. Miles GB, Cerminara NL, Marple-Horvat DE (2006) Purkinje cells in the lateral cerebellum of the cat encode visual events and target motion during visually guided reaching. J Physiol (London) 571:619–637. Miller EK, Cohen JD (2001) An integrative theory of prefrontal cortex function. Annu Rev Neurosci 24:167–202. Miller S, Oscarsson O (1970) Termination and functional organization of spino-olivocerebellar paths. In: Fields WS, Willis WD, editors. The cerebellum in health and disease. St. Louis, MO: Green, pp. 172–200. Miller SG, Kennedy MB (1985) Distinct forebrain and cerebellar isozymes of type II Ca2+/ calmodulin-dependent protein kinase associate differently with the postsynaptic density fraction. J Biol Chem 260:9039–9046. Mink JW (1996) The basal ganglia: focused selection and inhibition of competing motor programs. Prog Neurobiol 50:381–425. Minsky M, Papert S (1988) Perceptrons: an introduction to computational geometry, 3rd ed. Cambridge, MA: MIT Press. Miyashita Y (1984) Eye velocity responsiveness and its proprioceptive component in the floccular Purkinje cells of the alert pigmented rabbit. Exp Brain Res 55:81–90. Miyashita Y (2004) Cognitive memory: cellular and network machineries and their top-down control. Science 306:435–440. Miyata M, Finch EA, Khiroug L, Hashimoto K, Hayasaka S, Oda S, Inouye M, Takagishi Y, Augustine GJ, Kno M (2000) Local calcium release in dendritic spines required for long-term synaptic depression. Neuron 28:233–244. Miyata M, Hashimoto K, Offermanns S, Simon MI, Kano M (1998) Deficient cerebellar LTD in mice lacking the subunit of heterotrimeric G protein Gq (Abstract). Annu Meet Soc Neurosci 28th, Los Angeles CA., p. 620.1. Miyata M, Kim HT, Hashimoto K, Lee TK, Cho SY, Jiang H, Wu Y, Jun K, Wu D, Kano M, Shin HS (2001) Deficient long-term synaptic depression in the rostral cerebellum correlated with impaired motor learning in phospholipase C beta4 mutant mice. Eur J Neurosci 13:1945–1954. Miyata M, Okada D, Hashimoto K, Kano M, Ito M (1999) Corticotropin-releasing factor plays a permissive role in cerebellar long-term depression. Neuron 22:763–775. Mizukoshi A, Kitama T, Omata T, Ueno T, Kawato M, Sato Y (2000) Motor dynamics encoding in the rostral zone of the cat cerebellar flocculus during vertical optokinetic eye movements. Exp Brain Res 132:260–268. Molinari M, Leggio MG, Solida A, Ciorra R, Misciagna S, Silveri MC, Petrosini L (1997) Cerebellum and procedural learning: evidence from focal cerebellar lesions. Brain 120:1753–1762. Monzée J, Drew T, Smith AM (2004) Effects of muscimol inactivation of the cerebellar nuclei on precision grip. J Neurophysiol 91:1240–1249. Monzée J, Smith AM (2004) Responses of cerebellar interpositus neurons to predictable perturbations applied to an object held in a precision grip. J Neurophysiol 91:1230–1239.

242

the cerebellum

Morgan JI, Curran T (1989) Stimulus-transcription coupling in neurons: role of cellular immediateearly genes. Trends Neurosci 12:459–462. Mori S, Matsui T, Kuze B, Asanome M, Nakajima K, Matsuyama K (1999) Stimulation of a restricted region in the midline cerebellar white matter evokes coordinated quadrupedal locomotion in the decerebrate cat. J Neurophysiol 82:290–300. Morishita W, Sastry BR (1996) Postsynaptic mechanisms underlying long-term depression of GABAergic transmission in neurons of the deep cerebellar nuclei. J Neurophysiol 76:59–68. Morton SM, Bastian AJ (2006) Cerebellar contributions to locomotor adaptations during splitbelt treadmill walking. J Neurosci 26:9107–9116. Moruzzi G (1940) Paleocerebellar inhibition of vasomotor and respiratory carotid sinus reflexes. J Neurophysiol 3:20–32. Moschovakis AK, Scudder CA, Highstein SM (1996) The microscopic anatomy and physiology of the mammalian saccadic system. Prog Neurobiol 50:133–254. Mugnaini E (1983) The length of cerebellar parallel fibers in chicken and rhesus monkey. J Comp Neurol 220:7–15. Müller F, Dichgans J (2002) Impairments of precision grip in two patients with acute unilateral cerebellar lesions: a simple parametric test for clinical use. Neuropsychologia 32:265–269. Murashima M, Hirano T (1999) Entire course and distinct phases of day-lasting depression of miniature EPSC amplitudes in cultured Purkinje neurons. J Neurosci 19:7326–7333. Nagahama Y, Fukuyama H, Yamauchi H, Matsuzaki S, Konishi J, Shibasaki H, Kimura J (1996) Cerebral activation during performance of a card sorting test. Brain 119:1667–1675. Nagao S (1988) Behavior of floccular Purkinje cells correlated with adaptation of horizontal optokinetic eye movement response in pigmented rabbits. Exp Brain Res 73:489–497. Nagao S (1989) Role of cerebellar flocculus in adaptive interaction between optokinetic eye movement response and vestibulo-ocular reflex in pigmented rabbits. Exp Brain Res 77:541–551. Nagao S, Ito M (1991) Subdural application of hemoglobin to the cerebellum blocks vestibuloocular reflex adaptation. NeuroReport 2:193–196. Nagao S, Ito M, Karachot L (1984) Sites in the rabbit flocculus specifically related to eye blinking and neck muscle contraction. Neurosci Res 1:149–152. Nagao S, Ito M, Karachot L (1985) Eye field in the cerebellar flocculus of pigmented rabbits determined with local electrical stimulation. Neurosci Res 3:39-51. Nagao S, Kitazawa H (2000) Subdural applications of NO scavenger or NO blocker to the cerebellum depress the adaptation of monkey post-saccadic smooth pursuit eye movements. NeuroReport 11:131–133. Nagao S, Kitazawa H (2003) Effects of reversible shutdown of the monkey flocculus on the retention of adaptation of the horizontal vestibulo-ocular reflex. Neuroscience 118:563–570. Nakadate K, Shutoh F, Nagao S, Shigemoto R (2004) Reduction of synapse density after long-term adaptation of horizontal optokinetic response in the mouse flocculus. Neurosci Res, 50:S129. Nakamagoe K, Iwamoto Y, Yoshida K (2000) Evidence for brainstem structures participating in oculomotor integration. Science 288:857–859. Nakazawa K, Karachot L, Nakabeppu Y, Yamamori T (1993) The conjunctive stimuli that cause long-term desensitization also predominantly induce c-Fos and Jun-B in cerebellar Purkinje cells. NeuroReport 4:1275–1278.

References

243

Napier JR, Napier PH (1967) A handbook of living primates. New York: Academic Press. Napper RM, Harvey RJ (1988a) Quantitative study of the Purkinje cell dendritic spines in the rat cerebellum. J Comp Neurol 274:158–167. Napper RM, Harvey RJ (1988b) Number of parallel fiber synapses on an individual Purkinje cell in the cerebellum of the rat. J Comp Neurol 274:168–177. Narasimhan K, Pessah IN, Linden DJ (1998) Inositol-1,4,5-trisphosphate receptor-mediated ca mobilization is not required for cerebellar long-term depression in reduced preparations. J Neurophysiol 80:2963-2974. Nashner LM (1976) Adapting reflexes controlling the human posture. Exp Brain Res 26:59–72. Nelson AB, Gittis AH, du Lac S (2005) Decreases in CaMKII activity trigger persistent potentiation of intrinsic excitability in spontaneously firing vestibular nucleus neurons. Neuron 46:623–631. Nelson AB, Krispel CM, Sekirnjak C, du Lac S (2003) Long-lasting increases in intrinsic excitability triggered by inhibition. Neuron 40:609–620. Netzeband JG, Parsons KL, Sweeney DD, Gruol DL (1997) Metabotropic glutamate receptor agonists alter neuronal excitability and Ca2+ levels via the phospholipase C transduction pathway in cultured Purkinje neurons. J Neurophysiol 78:63–75. Nichols R, Ross KT (2009) The implications of force feedback for the L model. Adv Exp Med Biol 629:663–679. Nicolson RI, Daum I, Schugens MM, Fawcett AJ, Schulz A (2002) Eyeblink conditioning indicates cerebellar abnormality in dyslexia. Exp Brain Res 143:42–50. Nieto-Bona MP, Garcia-Segura LM, Torres-Aleman I (1997) Transynaptic modulation by insulinlike growth factor I of dendritic spines in Purkinje cells. Int J Devl Neurosci 15:749–754. Nieuwenhuys R, Nicholson C (1967) The cerebellum of mormyrids. Nature 215:764-765. Nisimaru N (2004) Cardiovascular models of the cerebellum. J Physiol Sci (Jpn J Physiol) 54:431–448. Nisimaru N, Arata A, Ito M (2010) Hypothalamic orexinergic neurons and flocculus Purkinje cells regulate cardiovascular defense reactions. Neurosci Res. 66 supple.1, P1-k11. Nisimaru N, Mittal C, Sooksawate T, Shirai Y, Sakurai T, Yamamoto M, Ito M (2006) Hypothalamic defense reactions involve Purkinje cells in the flocculus folium p via orexin and GABA. Neurosci Res 55, supple.1, S61:OS2P-8-05. Nisimaru N, Mittal C, Sooksawate T, Shirai Y, Sakurai T, Yamamoto M, Ito M (2006) Hypothalamic defense reactions involve Purkinje cells in the flocculus folium p via orexin and GABA. Neurosci Res 55:S615. Nisimaru N, Okahara K, Yanai S (1998) Cerebellar control of the cardiovascular responses during postural changes in conscious rabbits. Neurosci Res 32:267–271. Nitta T, Akao T, Kurkin S, Fukushima K (2008) Involvement of the cerebellar dorsal vermis in vergence eye movements in monkeys. Cereb Cortex 18:1042-1057. Noble D (2006) The music of life. Biology beyond the genome. Oxford University Press. Oxford UK, New York USA Noda H, Fujikado T (1987) Involvement of Purkinje cells in evoking saccadic eye movements by microstimulation of the posterior cerebellar vermis of monkeys. J Neurophysiol 57:1247–1261. Noda H, Warabi T (1987) Responses of Purkinje cells and mossy fibres in the flocculus of the monkey during sinusoidal movements of a visual pattern. J Physiol (London) 387:611–628.

244

the cerebellum

Norman DA, Shallice T (1986) Attention to action: willed and automatic control of behaviour (Revised reprint of Norman & Shallice, 1980). In: Davidson RJ, Schwartz GE, Shapiro D, editors. Consciousness and self-regulation: advances in research and theory. Plenum Press, pp. 1–18. New York. Nunzi M-G, Shigemoto R, Mugnaini E (2002) Differential expression of calretinin and metabotropic glutamate receptor mGluR1α defines subsets of unipolar brush cells in mouse cerebellum. J Comp Neurol 451:189–199. Nusser Z., Mulvihill E, Streit P, Somogyi P (1994) Subsynaptic segregation of metabotropic and ionotropic glutamate receptors as revealed immunogold localization. Neuroscience 61:421–427. Obata K, Ito M, Ochi R, Sato N (1967) Pharmacological properties of the postsynaptic inhibition by Purkinje cell axons and the action of gamma-aminobutyric acid on Deiters neurons. Exp Brain Res 4:43–57. O’Donoghue DL, Bishop GA (1990) A quantitative analysis of the distribution of Purkinje cell axonal collaterals in different zones of the cat’s cerebellum: an intracellular HRP study. Exp Brain Res 80:63–71. O’Donoghue DL, King JS, Bishop GA (1989) Physiological and anatomical studies of the interactions between Purkinje cells and basket cells in the cat’s cerebellar cortex: evidence for a unitary relationship. J Neurosci 9:2141–2150. Oelschläger HHA (2000) Morphological and functional adaptation of the toothed whale head to aquatic life. Hist Biol 14:33–39. Oelschläger HHA (2008) The dolphin brain—A challenge for synthetic neurobiology. Brain Res Bull 75:450–459. Oelschläger HHA, Oelschläger JS (2009) Brain, In: Encyclopedia of Marine Mammals. 2nd edition, eds. Perrin WF, Wursig B, Thewissen JGM, pp 134-149. Amesterdam: Elsevier. Offermanns S, Hashimoto K, Watanabe M, Sun W, Kurihara H, Thompson RF, Inoue Y, Kano M, Simon MI (1997) Impaired motor coordination and persistent multiple climbing fiber innervation of cerebellar Purkinje cells in mice lacking Gaq. PNAS USA 94:14089–14094. Ohishi H, Ogawa-Meguro R, Shigemoto R, Kaneko T, Nakanishi S, Mizuno N (1994) Immunohistochemical localization of metabotropic glutamate receptors, mGluR2 and mGluR3, in rat cerebellar cortex. Neuron 13:55–60. Ohki M, Kitazawa H, Hiramatsu T, Kaga K, Kitamura T, Yamada J, Nagao S (2009) Role of primate cerebellar hemisphere in voluntary eye movement control revealed by lesion effects. J Neurophysiol 101:934–947. Erratum in: J Neurophysiol (2009) 101:2733. Ohtsuka K, Noda H (1992) Burst discharges of mossy fibers in the oculomotor vermis of macaque monkeys during saccadic eye movements. Neurosci Res 15:102–114. Ohyama T, Nores WL, Medina JF, Riusech FA, Mauk MD (2006) Learning-Induced Plasticity in Deep Cerebellar Nucleus. J Neurosci 26:12656–12663. Ojakangas CL, Ebner TJ (1992) Purkinje cell complex and simple spike changes during a voluntary arm movement learning task in the monkey. J Neurophysiol 68:2222–2236. Okamoto T, Endo S, Shirao T, Nagao S (2011) Role of cerebellar cortical protein synthesis in transfer of memory trace of cerebellum-dependent motor learning, J Neurosci 31:8958–8966. Omata T, Kitama T, Mizukoshi A, Ueno T, Kawato M, Sato, Y (2000) Purkinje cell activity in the middle zone of the cerebellar flocculus during optokinetic and vestibular eye movement in cats. J Physiol Sci (Jpn J Physiol) 50:357–370.

References

245

Onat F, Cavdar S (2003) Cerebellar connections: hypothalamus. Cerebellum 2:263-269. Orlovsky GN (1972a) Activity of vestibulospinal neurons during locomotion. Brain Res 46:85-98. Orlovsky GN (1972b) Activity of rubrospinal neurons during locomotion. Brain Res 46:99-112. Osanai R, Nagao S, Kitamura T, Kawabata I, Yamada J (1999) Differences in mossy and climbing afferent sources between flocculus and ventral and dorsal paraflocculus in the rat. Exp Brain Res 124:248–264. Oscarsson O (1979) Functional units of the cerebellum—sagittal zones and microzones. Trends Neurosci 2:144–145. Oscarsson O, Uddenberg N (1965) Properties of afferent connections to the rostral spinocerebellar tract in the cat. Acta Physiol Scand 64:143–153. Ottersen OP, Storm-Mathisen J, Somogyi P (1988) Colocalization of glycine-like and GABA-like immunoreactivities in Golgi cell terminals in the rat cerebellum: a postembedding light and electron microscopic study. Brain Res 450:342–353. Oztop E, Kawato M, Arbib M (2006) Mirror neurons and imitation: A computationally guided review. Neural Net 19:254-271. Palay SL, Chan-Palay V (1974) The Cerebellar Cortex. New York: Springer-Verlag. Palkovits M, Magyar P, Szentágothai J (1971) Quantitative histological analysis of the cerebellar cortex in the cat. II. Cell numbers and densities in the granular layer. Brain Res 32:15–30. Palkovits M, Mezey E, Hamori J, Szentágothai J (1977) Quantitative histological analysis of the cerebellar nuclei in the cat. I. Numerical data on cells and on synapses. Exp Brain Res 28:189–209. Parenti R, Zappalá A, Serapide MF, Pantó MR, Cicirata F (2002) Projections of the basilar pontine nuclei and nucleus reticularis tegmenti pontis to the cerebellar nuclei of the rat. J Comp Neurol 452:115–127. Parsons LM, Bower JM, Gao JH, Xiong J, Li J, Fox PT (1997) Lateral cerebellar hemispheres actively support sensory acquisition and discrimination rather than motor control. Learn Mem 4:49–62. Pastor AM, de la Cruz RR, Baker R (1994) Cerebellar role in adaptation of the goldfish vestibuloocular reflex. J Neurophysiol 72:1383–1394. Pause T (1996) Location and function of the human frontal eye-field: a selective review. Neuropsychologia 34:475–483. Pearson KG, Collins DF (1993) Reversal of the influence of group lb afferents from plantaris on activity in medial gastrocnemius muscle during locomotor activity. J Neurophysiol 70:1009–1017. Pekhletski R, Gerlai R, Overstreet LS, Huang XP, Agopyan N, Slater NT, Abramow-Newerly W, Roder JC, Hampson DR (1996) Impaired cerebellar synaptic plasticity and motor performance in mice lacking the mGluR4 subtype of metabotropic glutamate receptor. J Neurosci 16:6364–6373. Pellerin JP, Lamarre Y (1997) Local field potential oscillations in primate cerebellar cortex during voluntary movement. J Neurophysiol 78:3502–3507. Pellionisz A, Szentágothai J (1973) Dynamic single unit stimulation of a realistic cerebellar network model. Brain Res. 49:83–99. Penfield W, Perot P (1963) The brain’s record of auditory and visual experience; a final summary and discussion. Brain 86:595–696.

246

the cerebellum

Penfield W, Rasmussen T (1950) The cerebral cortex of man. New York: The Macmillan Company. Peterson BW, Pitts NG, Fukushima K (1979) Reticulospinal connections with limb and axial motoneurons. Exp Brain Res 1–20. Petit L, Clark VP, Ingeholm J, Haxby JV (1997) Dissociation of saccade-related and pursuit-related activation in human frontal eye fields as revealed by fMRI. J Neurophysiol 77:3386–3390. Petralia RS, Zhao HM, Wang YX, Wenthold RJ (1998) Variations in the tangential distribution of postsynaptic glutamate receptors in Purkinje cell parallel and climbing fiber synapses during development. Neuropharmacology 37:1321–1334. Pettersson LG, Lundberg A, Alstermark B, Isa T, Tantisira B (1997) Effect of spinal cord lesions on forelimb target-reaching and on visually guided switching of target-reaching in the cat. Neurosci Res 29:241–256. Philipona D, Coenen OJ-MD (2004) Model of granular layer encoding of the cerebellum. Neurocomputing 58:575–580. Piaget J (1951) The psychology of intelligence. London: Routledge and Kegan Paul. Pichitpornchai C, Rawson JA, Rees S (1994) Morphology of parallel fibres in the cerebellar cortex of the rat: an experimental light and electron microscopic study with biocytin. J Comp Neurol 342:206–220. Pierrot-Deseilligny E, Burke DC (2005) The circuitry of the human spinal cord: its role in motor control and movement disorders. Cambridge University Press. Cambridge UK. Pijpers A, Voogd J, Ruigrok TJH (2005) Topography of olivo-cortico-nuclear modules in the intermediate cerebellum of the Rat. J Comp Neurol 492:193–213. Piochon C, Levenes C, Ohtsuki G, Hansel CP (2010) Purkinje cell NMDA receptors assume a key role in synaptic gain control in the mature cerebellum. J Neurosci 30:15330–15335. Piomelli D, Wang JKT, Shira TS, Nairn AC, Czernik AJ, Greengard P (1989) Inhibition of Ca2+/calmodulin-dependent protein kinase II by arachidonic acid and its metabolites. PNAS USA 86:8550–8554. Ploghaus A, Tracey I, Clare S, Gati JS, Rawlins JNP, Matthews PM (2000) Learning about pain: the neural substrate of the prediction error for aversive events. PNAS USA 97:9281–9286. Prochazka A, Clarac F, Loeb GE, Rothwell JC, Wolpaw JR (2000) What do reflex and voluntary mean? Modern views on an ancient debate. Exp Brain Res 130:417–432. Prochazka A, Ellaway PH (2011) Sensory systems in the control of movement. In: Handbook of Exercise in Health and Disease. Baldwin K, Edgerton VR, Wagner P eds New York: Americal Physiological Society. In press. Prochazka A, Gillard D, Bennett DJ (1997) Implications of positive feedback in the control of movement. J Neurophysiol 77:3237–3251. Prochazka A, Hulliger M, Zangger P, Appenteng K (1985) ‘Fusimotor set’: new evidence for α-independent control of γ-motoneurones during movement in the awake cat. Brain Res 339:136-140. Pugh JR, Raman IM (2006) Potentiation of mossy fiber EPSCs in the cerebellar nuclei by NMDA receptor activation followed by postinhibitory rebound current. Neuron 51:113–123. Quallo MM, Pricec CJ, Uenod K, Asamizuyad T, Chengd K, Lemon RN, Iriki A (2009) Gray and white matter changes associated with tool-use learning in macaque monkeys. PNAS USA 106:18379-18384.

References

247

Quinn KJ, Helminski JO, Didier AJ, Baker JF, Peterson BW (1996) Changes in sensitivity of vestibular nucleus neurons induced by cross-axis adaptation of the vestibulo-ocular reflex in the cat. Brain Res 718:176–180. Rachmuth G, Poon CS (2008) Transistor analogs of emergent iono-neuronal dynamics. HFSP J 2:156–166. Racine RJ, Wilson DA, Gingell R, Sunderland D (1986) Long-term potentiation in the interpositus and vestibular nuclei in the rat. Exp Brain Res 63:158–162. Ralston DD (1994) Corticorubral synaptic organization in Macaca fascicularis: a study utilizing degeneration, anterograde transport of WGA-HRP, and combined immuno-GABA-gold technique and computer-assisted reconstruction. J Comp Neurol 350:657–673. Ramachandran VS, Hirsten W (1998) The perception of phantom limbs. Brain 121:1603–1630. Ramat S, Leigh RJ, Zee DS, Optican LM (2007) What clinical disorders tell us about the neural control of saccadic eye movements. Brain 130:10–35. Ramnani N (2006) The primate cortico-cerebellar system: anatomy and function. Nat Rev Neurosci 7:511–522. Rancz EA, Ishikawa T, Duguid I, Chadderton P, Mahon S, Häusser M (2007) High-fidelity transmission of sensory information by single cerebellar mossy fibre boutons. Nature 450:1245–1248. Raphan TH, Matsuo V, Cohen B (1979) Velocity storage in the vestibulo-ocular reflex arc (VOR). Exp Brain Res 35:229–248. Rathelot J-A, Strick PL (2009) Subdivisions of primary motor cortex based on cortico-motoneuronal cells. PNAS USA 106:918–923. Raymond J, Nieoullon A, Demêmes D, A. Sans (1984) Evidence for glutamate as a neurotransmitter in the cat vestibular nerve: radioautographic and biochemical studies. Exp Brain Res 56:523–531. Raymond JL, Lisberger SG, Mauk MD (1996) The cerebellum: a neuronal learning machine? Science 272:1126–1131. Reichenberger I, Streit P, Ottersen OP, Dieringer N (1993) GABA- and glycine-like immunoreactivities in the cerebellum of the frog. Neurosci Lett 154:89–92. Renzi, M and Farrant, M, Cull-Candy, SG (2007) Climbing-fibre activation of NMDA receptors in Purkinje cells of adult mice. J Physiol (London) 585:91-101. Reynolds T, Hartell NA (2000) An evaluation of the synapse specificity of long-term depression induced in rat cerebellar slices. J Physiol (London) 527:563–577. Reynolds T, Hartell NA (2001). Roles for nitric oxide and arachidonic acid in the induction of heterosynaptic cerebellar LTD. NeuroReport 12:133–136. Riklan M, Cullinan T, Shulman M, Cooper IS (1976) A psychometric study of chronic cerebellar stimulation in man. Biol Psychiatry 11:543–574. Rizzolatti G, Craighero L (2004) The mirror-neuron system. Annu Rev Neurosci 27:169–192. Robertson EM (2007) The serial reaction time task: implicit motor skill learning? J Neurosci 27:10073-10075. Robinson DA (1972) Eye movements evoked by collicular stimulation in the alert monkey. Vision Res 12:1795–1808.

248

the cerebellum

Robinson DA (1976) Adaptive gain control of vestibulo-ocular reflex by the cerebellum. J Neurophysiol 39:954–969. Robinson FR, Houk JC, Gibson AR (1987) Limb specific connections of the cat magnocellular red nucleus. J Comp Neurol 257:553–577. Erratum in: J Comp Neurol (1987) 259:622. Robinson FR, Straube A, Fuchs AF (1993) Role of the caudal fastigial nucleus in saccade generation. II. Effects of muscimol inactivation. J Neurophysiol 70:1741–1758. Roitman AV, Pasalar S, Johnson MTV, Ebner TJ (2005) Position, direction of movement, and speed tuning of cerebellar Purkinje cells during circular manual tracking in monkey. J Neurosci 25:9244–9257. Ron S, Robinson DA (1973) Eye movements evoked by cerebellar stimulation in the alert monkey. J Neurophysiol 36:1004–1022. Rosenblatt F (1962) Principles of neurodynamics: perceptron and the theory of brain mechanisms. Washington, DC: Spartan Books. Rossi DJ, Alford S, Mugnaini E, Slater NT (1995) Properties of transmission at a giant glutamatergic synapse in cerebellum: the mossy fiber-unipolar brush cell synapse. J Neurophysiol 74:24–42. Rossi DJ, Hamann M (1998) Spillover-mediated transmission at inhibitory synapses promoted by high affinity alpha6 subunit GABA(A) receptors and glomerular geometry. Neuron 20:783–795. Erratum in: Neuron (1998) 21:527. Roustan P, Abitbol M, Menini C, Ribeaudeau F, Gerard M, Vekemans M, Mallet J, Dufier J (1995) The rat phospholipase C beta gene is expressed at high abundance in cerebellar Purkinje cells. NeuroReport 2:1837–1841. Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagating errors. Nature 323:533–536.. Sadakata T, Kakegawa W, Mizoguchi A, Washida M, Katoh-Semba R, Shutoh F, Okamoto T, Nakashima H, Kimura K, Tanaka M, Sekine Y, Itohara S, Yuzaki M, Nagao S, Furuichi T (2007) Impaired cerebellar development and function in mice lacking CAPS2, a protein involved in neurotrophin release. J Neurosci 27:2472–2482. Sadakata T, Washida M, Iwayama Y, Shoji S, Sato Y, Ohkura T, Katoh-Semba R, Nakajima M, Sekine Y, Tanaka M, Nakamura K, Iwata Y, Tsuchiya KJ, Mori N, Detera-Wadleigh SD, Ichikawa H, Itohara S, Yoshikawa T, Furuichi T (2007) Autistic-like phenotypes in Cadps2–knockout mice and aberrant CADPS2 splicing in autistic patients. J Clin Invest 117:931–943. Sahin M, Hockfield S (1990) Molecular identification of Lugaro cells in the cat cerebellar cortex. J Comp Neurol 301:575–584. Saint-Cyr JA, Courville J (2004) Sources of descending afferents to the inferior olive from the upper brain stem in the cat as revealed by the retrograde transport of horseradish peroxidase. J Comp Neurol 198:567–581. Sakagami H, Kondo H (1993) Cloning and screening of a gene encoding the b polypeptide of Ca2+/calmodulin-dependent protein kinase IV and its expression confined to the mature cerebellar granule cells. Mol Brain Res 19:215–218. Sakurai M (1987) Synaptic modification of parallel fibre-Purkinje cell transmission in in vitro guinea-pig cerebellar slices. J Physiol (London) 394:463–480. Salin PA, Malenka RC, Nicoll RA (1996) Cyclic AMP mediates a presynaptic form of LTP at cerebellar parallel fiber synapses. Neuron 16:797–803. Sanes JR, Lichtman JW (1999) Can molecules explain long-term potentiation? Nat Neurosci 2:597–604.

References

249

Sasaki K (1985) Cerebro-cerebellar interactions and organization of a fast and stable hand movement: cerebellar participation in voluntary movement and motor learning. In: Bloedel JR, Dichgans J, Precht W, editors. Cerebellar functions. Berlin: Springer, pp. 70–85. Sasaki K, Gemba H (1982) Development and change of cortical field potentials during learning processes of visually initiated hand movements in the monkey. Exp Brain Res 48:429–437. Sasaki K, Gemba H, Hashimoto S (1981) Influences of cerebellar hemispherectomy upon cortical potentials preceding visually initiated hand movements in the monkey. Brain Res 205:425–430. Sato T, Yokoyama R, Fukushima J, Fukushima K (1999) Latency of cross-axis vestibulo-ocular reflex induced by pursuit training in monkeys. Neurosci Res 33:65–70.16. Sato Y, Kawasaki T (1984) Functional localization in the three floccular zones related to eye movement control in the cat. Brain Res 290:25–31. Sato Y, Kawasaki T, Ikarashi K (1983) Afferent projections from the brainstem to the three floccular zones in cats. II. Mossy fiber projections. Brain Res 272:137–148. Scelfo B, Sacchetti B, Strata P (2008) Learning-related long-term potentiation of inhibitory synapses in the cerebellar cortex. PNAS USA 105:769–774. Scelfo B, Strata P (2005) Correlation between multiple climbing fibre regression and parallel fibre response development in the postnatal mouse cerebellum. Eur J Neurosci 21:971–978. Schall JD (2002) The neural selection and control of saccades by the frontal eye field. Phil Trans R Soc London B 357:1073–1082. Schall JD, Morel A, King DJ, Bullie J (1995) Topography of visual cortex connections with frontal eye field in macaque: convergence and segregation of processing streams. J Neurosci 15:4464–4487. Scheibel ME, Scheibel AB (1955) The inferior olive. A Golgi study. J Comp Neurol 102:77–131. Schlag J, Schlag-Rey M (1987) Evidence for a supplementary eye field. J Neurophysiol 57:179–200. Schlag-Rey M, Amador N, Sanchez H, Schlag J (1997) Antisaccade performance predicted by neuronal activity in the supplementary eye field. Nature 390:398–491. Schlösser R, Hutchinson M, Joseffer S, Rusinek H, Saarimaki A, Stevenson J, Dewey SL, Brodie JD (1998) Functional magnetic resonance imaging of human brain activity in a verbal fluency task. J Neurol Neurosurg Psychiatr 64:492–498. Schmahmann JD (1991) An emerging concept: the cerebellar contribution to higher function. Arch Neurol 48:1178–1187. Schmahmann JD, Doyon J, McDonald D, Holmes C, Lavoie K, Hurwitz AS, Kabani N, Toga A, Evans A, Petrides M (1999) Three-dimensional MRI atlas of the human cerebellum in proportional stereotaxic space. NeuroImage 10:233–260. Schmajuk NA, Lam YW, Gray JA (1996) Latent inhibition: a neural network approach. J Exp Psychol 22:321–349. Schonewille M, Luo C, Ruigrok TJ, Voogd J, Schmolesky MT, Rutteman M, Hoebeek FE, De Jeu MT, De Zeeuw CI (2006) Zonal organization of the mouse flocculus: physiology, input, and output. J Comp Neurol 497:670–682. Schonewille M, Gao Z, Boele H-J, Veloz MFV, Amerika WE, Simek AAM, De Jeu MT, Steinberg JP, Takamiya K, Hoebeek FE, Linden DJ, Huganir RL, De Zeeuw CI (2011) Reevaluating the role of LTD in cerebellar motor learning. Neuron 70:43–50. Schreurs BG, Oh MM, Alkon DL (1996) Pairing-specific long-term depression of Purkinje cell excitatory postsynaptic potentials results from a classical conditioning procedure in the rabbit cerebellar slice. J Neurophysiol 75:1051–1060.

250

the cerebellum

Schreurs BG, Tomsic D, Gusev PA, Alkon DL (1997) Dendritic excitability microzones and occluded long-term depression after classical conditioning of the rabbit’s nictitating membrane response. J Neurophysiol 77:86–92. Schuman EM, Dynes JL, Steward O (2006) Synaptic Regulation of Translation of Dendritic mRNAs. J Neurosci 26:7143–7146. Schweighofer N, Arbib MA, Dominey PF (1996a) A model of the cerebellum in adaptive control of saccadic gain. I. The model and its biological substrate. Biol Cybern 75:19–28. Schweighofer N, Arbib MA, Dominey PF (1996b) A model of the cerebellum in adaptive control of saccadic gain. II. Simulation results. Biol Cybern 75:29–36. Schweighofer N, Doya K, Fukai H, Chiron JV, Furukawa T, Kawato M (2004a) Chaos may enhance information transmission in the inferior olive. PNAS USA 101:4655–4660. Schweighofer N, Doya K, Kuroda S (2004b) Cerebellar aminergic neuromodulation: towards a functional understanding. Brain Res Rev 44:103–116. Schweighofer N, Spoelstra J, Arbib MA, Kawato M (1998) Role of the cerebellum in reaching movements in humans. II. A neural model of the intermediate cerebellum. Eur J Neurosci 10:95–105. Scudder CA, Fuchs AF (1992) Physiological and behavioral identification of vestibular nucleus neurons mediating the horizontal vestibuloocular reflex in trained rhesus monkeys. J Neurophysiol 68:244–264. Scudder CA, Kaneko CRS, AF Fuchs (2002) The brainstem burst generator for saccadic eye movements. A modern synthesis. Exp Brain Res 142:439–462. Sdrulla AD, Linden DJ (2007) Double dissociation between long-term depression and dendritic spine morphology in cerebellar Purkinje cells. Nat Neurosci 10:546–548. Sekerková G, Ilijic E, Mugnaini E (2004) Time of origin of unipolar brush cells in the rat cerebellum as observed by prenatal bromodeoxyuridine labeling. Neuroscience 127:845-858. Sekimjak C, Vissel B, Bollinger J, Faulstich M, du Lac S (2003) Purkinje cell synapses target physiologically unique brainstem neurons. J Neurosci 23:6392-6398. Selverston AI (1995) Modulation of circuits underlying rhythmic behaviors. J Comp Physiol 176:139–147. Shambes GM, Gibson JM, Welker W (1978) Fractured somatotopy in granule cell tactile areas of rat cerebellar hemispheres revealed by micromapping. Brain Behav Evol 15:116–140. Sheinberg DL, Logothetis NK (1997) The role of temporal areas in perceptual organization. PNAS USA 94:3408–3413. Shen K, Teruel MN, Subramanian K, Meyer T (1998) CaMKII functions as an F-actin targeting module that localizes CaMKIIa heterooligomers to dendritic spines. Neuron 21:593–606. Shen Y, Hansel C, Linden DJ (2002) Glutamate release during LTD at cerebellar climbing fiberPurkinje cell synapses. Nat Neurosci. 5:725-726. Sherrington CS (1906) The integrative action of the nervous system. 2nd edition, 1948. New Haven: Yale Univ. Press. Shibuki K, Gomi H, Chen L, Bao S, Kim JJ, Wakatsuki H, Fujisaki T, Fujimoto K, Katoh A, Ikeda T, Chen C, Thompson RF, Itohara S (1996) Deficient cerebellar long-term depression, impaired eyeblink conditioning, and normal motor coordination in GFAP mutant mice. Neuron 16:587–599. Shibuki K, Kimura S (1997) Dynamic properties of nitric oxide release from parallel fibres in rat cerebellar slices. J Physiol (London) 498:443–452.

References

251

Shibuki K, Okada D (1991) Endogenous nitric oxide release required for long-term synaptic depression in the cerebellum. Nature 349:326–328. Shidara M, Kawano K, Gomi H, Kawato M (1993) Inverse-dynamics model eye movement control by Purkinje cells in the cerebellum. Nature 365:50–52. Shigemoto R, Abe T, Nomura S, Nakanishi S, Hirano T (1994) Antibodies inactivating mGluR1 metabotropic glutamate receptor block long-term depression in cultured Purkinje cells. Neuron 12:1245–1255. Shigemoto R, Nakadate K, Shutoh F, Wang W, Fukazawa Y, Molnar E, Nagao S (2004) Reduction of AMPA receptor content in parallel fiber—Purkinje cell synapses by cerebellar motor learning. San Diego: 34th Annual meeting of Society for Neuroscience. Shik ML, Severin FV, Orlovsky GN (1966) Control of walking and running by means of electrical stimulation of the mid-brain. Biophysics 11:756–765. Shin JH, Linden DJ (2005) An NMDA receptor/nitric oxide cascade is involved in cerebellar LTD but it is not localized to the parallel fiber terminal. J Neurophysiol 94:4281–4289. Shinoda Y, Sugiuchi Y, Futami T, Izawa R (1992) Axon collaterals of mossy fibers from the pontine nucleus in the cerebellar dentate nucleus. J Neurophysiol 67:547–560. Shoshani J, Kupsky W, Marchant GH (2006) Elephant brain. Party I: cross morphology, function, comparative anatomy, and evolution. Brain Res Bull 70:124–157. Shutoh F, Katoh A, Kitazawa H, Aiba A, Itohara S, Nagao S (2002) Loss of adaptability of horizontal optokinetic response eye movements in mGluR1 knockout mice. Neurosci Res 42:141–145. Shutoh F, Ohki M, Kitazwa H, Itohara S, Nagao S (2006) Memory trace of motor learning shifts transsynaptically from cerebellar cortex to nuclei for consolidation. Neuroscience 139:767–777. Silva AJ, Kogan JH, Frankland PW, Kida S (1998) CREB and memory. Annu Rev Neurosci 21:127–148. Simat M, Parpan F, Fritschy JM (2007) Heterogeneity of glycinergic and GABAergic interneurons in the granule cell layer of mouse cerebellum. J Comp Neurol 500:71–83. Simpson JI (1984) The accessory optic system. Annu Rev Neurosci 7:13–41. Simpson JI, Belton T, Suh M, Winkelman B (2002) Complex spike activity in the flocculus signals more than the eye can see. Ann NY Acad Sci 978:232–236. Sims RE, Hartell NA (2005) Differences in transmission properties and susceptibility to LTD reveals functional specialization of ascending axon and parallel fiber synapses to Purkinje cells. J Neurosci 25:3246–3257. Sims RE, Hartell NA (2006) Differential susceptibility to synaptic plasticity reveals a functional specialization of ascending axon and parallel fiber synapses to cerebellar Purkinje cells. J Neurosci 26:5153–5159. Smith EE, Jonides J (1997) Working memory: a view from neuroimaging. Cognit Psychol 33:5–42. Smith KU, Putz V, Molitor K (1969) Eye movement-retina delayed feedback. Science 166:1542–1544. Smith-Hicks C, Xiao B, Deng R, Ji Y, Zhao, Shepherd JD, Posern G, Kuhl D, Huganir RL, Ginty DD, Worley PF, Linden DL (2010) SRF binding to SRE 6.9 in the Arc promoter is essential for LTD in cultured Purkinje cells. Nat Neurosci 13:1082–1089. Sola C, Tusell JM, Serratosa J (1999) Comparative study of the distribution of calmodulin kinase II and calcineurin in the mammalian brain. J Neurosci Res 1:651–662. Solms M, Turnbull O (2002) The brain and the inner world. OTHER Press New York USA.

252

the cerebellum

Solomon D, Cohen B (1994) Stimulation of the nodulus and uvula discharges velocity storage in the vestibulo-ocular reflex. Exp Brain Res 102:57–68. Sotelo C (2003) Viewing the brain through the master hand of Ramon y Cajal. Nat Rev Neurosci 4:71–77. Sotelo C, Gotow T, Wassef M (2004) Localization of glutamic-acid- decarboxylase-immunoreactive axon terminals in the inferior olive of the rat, with special emphasis on anatomical relations between GABAergic synapses and dendrodendritic gap junctions. J Comp Neurol 252:32–50. Sotelo C, Llinás R, Baker R (1974) Structural study of inferior olivary nucleus of the cat: morphological correlates of electronic coupling. J Neurophysiol 37:541–559. Spacek J (1985) Three-dimensional analysis of dendrtic spines III glial sheath. Anat Embryol 171:245-252. Sparks D, Nelson J (1987) Sensory and motor maps in the superior colliculus. Trends Neurosci 10:312–317. Spencer RF, Wenthold RJ, Baker R (1989) Evidence for glycine as an inhibitory neurotransmitter of vestibular, reticular, and prepositus hypoglossi neurons that project to the cat abducens nucleus. J Neurosci 9:2718–2736. Sperling AJ, Lu ZL, Manis FR, Seidenberg MS (2005) Deficits in perceptual noise exclusion in developmental dyslexia. Nature Neurosci 8:862–863. Squire LR (2009) Memory and brain systems:1969–2009. J Neurosci 29:12711–12716. Stamenov MI (2005) Body schema, body image, and mirror neurons. In: De Preester H, Knockaert V, editors. Body image and body schema. John Benjamins Publishing Co., pp. 21–43. Philadelphia USA. pp.21–43. Stein J, Walsh V (1997) To see but not to read: the magnocellular theory of dyslexia. Trends Neurosci 20:147–152. Sternweiss PC, Smrcka AV (1993) G protein in signal transduction: the regulation of phospholipase. In: Proceedings of Ciba Found Symp 176:96–106. Steuber V, Mittmann W, Hoebeek FE, Silver RA, De Zeeuw CI, Häusser M, De Schutter E (2007) Cerebellar LTD and pattern recognition by Purkinje cells. Neuron 54:121–136. Stone LS, Lisberger SG (1990a) Visual responses of Purkinje cells in the cerebellar flocculus during smooth-pursuit eye movements in monkeys. I. Complex spikes. J Neurophysiol 63:1241–1261. Stone LS, Lisberger SG (1990b) Visual responses of Purkinje cells in the cerebellar flocculus during smooth-pursuit eye movements in monkeys. II. Simple spikes. J Neurophysiol 63:1262–1275. Stoodley CJ, Fawcett AJ, Nicolson RI, Stein JF (2006) Balancing and pointing tasks in dyslexic and control adults. Dyslexia 12:276–288. Strassman A, Highstein SM, McCrea RA (1986a) Anatomy and physiology of saccadic burst neurons in the alert squirrel monkey. I. Excitatory burst neurons. J Comp Neurol 249:337–357. Strassman A, Highstein SM, McCrea RA (1986b) Anatomy and physiology of saccadic burst neurons in the alert squirrel monkey. II. Inhibitory burst neurons. J Comp Neurol 249:358–380. Strata P (2002) Dendritic spines in Purkinje cells. Cerebellum 1:230–232. Straube A, Scheuerer W, Eggert T (1997) Unilateral cerebellar lesions affect initiation of ipsilateral smooth pursuit eye movements in humans. Ann Neurol 42:891–898. Strick PL, Dum RP, Fiez JA (2009) Cerebellum and nonmotor function. Annu Rev Neurosci 32:413–434.

References

253

Stuart DG (2007) Reflections on integrative and comparative movement neuroscience. Integr Comp Biol 47:482–504. Stuart DG, Hultborn H (2008) Thomas Graham Brown (1882–1965), Anders Lundberg (1920–), and the neural control of stepping. Brain Res Rev 59:74–95. Stuart DG, McDonagh JC (1998) Reflections on a Bernsteinian approach to systems neuroscience: the controlled locomotion of high-decerebrate cats. In: Latash ML, editor. Progress in motor control: Bernstein’s tradition in movement studies. 1:21–49. Stuart DG, Pierce PA (2006) The academic lineage of Sir John Carew Eccles (1903-1997). Prog Neurobiol 78:136-155. Stuss DT, Knight RT (2002) Principles of frontal lobe function. New York: Oxford University Press. Sugihara I, Wu H, Shinoda Y (1996) Morphology of axon collaterals of single climbing fibers in the deep cerebellar nuclei of the rat. Neurosci Lett 2:1733–1736. Sugihara I, Wu H-S, Shinoda Y (1999) Morphology of single olivocerebellar axons labeled with biotinylated dextran amine in the rat. J Comp Neurol 414:131–148. Sugihara I, Shinoda Y (2004) Molecular, topographic, and functional organization of the cerebellar cortex: a study with combined aldolase C and olivocerebellar labeling. J Neurosci 24:8771-8785. Sugiyama T, Hirono M, Suzuki K, Nakamura Y, Aiba A, Nakamura K, Nakao K, Katsuki M, Yoshioka T (1999) Localization of phospholipase Cbeta isozymes in the cerebellum. Biochem Biophys Res Commun 19: 473-478. Sugiyama T, Hirono M, Suzuki K, Nakamura Y, Aiba A, Nakamura K, Nakao K, Katsuki M, Yoshioka T (1999) Localization of phospholipase Cbeta isozymes in the cerebellum. Biochem Biophys Res Commun 19:473–478. Sugrue MM, Brugge JS, Marshak DR, Greengard P, Gustafson EL (1990) Immunocytochemical localization of the neuron-specific form of the c-src gene product, pp60c-src(+), in rat brain. J Neurosci 10:2513–2527. Suh M, Leung H-C, Kettner RE (2000) Cerebellar flocculus and ventral paraflocculus Purkinje cell activity during predictive and visually driven pursuit in monkey. J Neurophysiol 84:1835–1850. Suzuki DA, Betelak KF, Yee RD (2009) Gaze pursuit responses in nucleus reticularis tegmenti pontis of head-unrestrained macaques. J Neurophysiol 101:460–473. Swensen AM, Marder E (1986) Modulators with convergent cellular actions elicit distinct circuit outputs. J Neurosci 21:4050–4058. Szapiro G, Barbour B (2007) Multiple climbing fibers signal to molecular layer interneurons exclusively via glutamate spillover. Nat Neurosci 10:735–742. Szentágothai J (1963) New data on the functional anatomy of synapses. Magy Tud Akad Biol Orv Tud Osztal Kozl 6:217–227. Szentágothai J, Palkovits K (1959) Uber den Ursprung der Kletterfasern des Kleinhirn. Z Anat Entwickl-Gesch 1231:130–141. Szpunar KK, Watson JM, McDermott KB (2007) Neural substrates of envisioning the future. PNAS USA 2:642–647. Tabata HK, Yamamoto K, Kawato M (2002) Computational study on monkey VOR adaptation and smooth pursuit based on the parallel control-pathway theory. J Neurophysiol 87:2176–2189. Taillanter M, Lannou J (1988) Responses of nucleus reticularis tegmenti pontis neurons to vestibular stimulation in the rat. Exp Brain Res 69:2417–423.

254

the cerebellum

Takagi M, Tamargo R, Zee DS (2003) Effects of lesions of the cerebellar oculomotor vermis on eye movements in primate: binocular control. Prog Brain Res 142:19–33. Takagi M, Zee DS, Tamargo RJ (2000) Effects of lesions of the oculomotor cerebellar vermis on eye movements in primate: smooth pursuit. J Neurophysiol 83:2047–2062. Takahashi KA, Linden DJ (2000) Cannabinoid receptor modulation of synapses received by cerebellar Purkinje cells. J Neurophysiol 83:1167–1180. Takayasu Y, Iino M, Furuya N, Ozawa S (2003) Muscarine-induced increase in frequency of spontaneous EPSCs in Purkinje cells in the vestibulo-cerebellum of the rat. J Neurosci 23:6200–6208. Takemura A, Inoue Y, Gomi H, Kawato M, Kawano K (2001) Change in neuronal firing patterns in the process of motor command generation for the ocular following response. J Neurophysiol 86:1750–1763. Takeuchi T, Ohtsuki G, Yoshida T, Fukaya M, Wainai T, Yamashita M, Yamazaki Y, Mori H, Sakimura K, Kawamoto S, Watanabe M, Hirano T, Mishina M (2008) Enhancement of both longterm depression induction and optokinetic response adaptation in mice lacking delphilin. PLoS ONE 3:e2297. Takeuchi T, Ohtsuki G, Yoshida T, Fukaya M, Wainai T, Yamashita M, Yamazaki Y, Mori H, Sakimura K, Kawamoto S, Watanabe M, Hirano T, Mishina M (2008) Enhancement of both longterm depression induction and optokinetic response adaptation in mice lacking delphilin. PLoS ONE 3:e2297. Tanaka J, Nakagawa S, Kushiya E, Yamasaki M, Fukaya M, Iwanaga T, Simon MI, Sakimura K, Kano M, Watanabe M (2000) Gq protein α subunits Gαq and Gα11 are localized at postsynaptic extrajunctional membrane of cerebellar Purkinje cells and hippocampal pyramidal cells. Eur J Neurosci. 12:781–792. Tanaka K (1996) Inferotemporal cortex and object vision. Annu Rev Neurosci 19:109–139. Tanaka K, Augustine GJ (2008) A positive feedback signal transduction loop determines timing of cerebellar long-term depression. Neuron 59:608–620. Tanimura A, Yamazaki M, Hashimotodani Y, Uchigashima M, Kawata, S, Abe M, Kita Y, Hashimoto K, Shimizu T, Watanabe M, Sakimura K, Kano M (2010) The endocannabinoid 2-arachidonoylglycerol produced by diacylglycerol lipase α mediates retrograde suppression of synaptic transmission. Neuron 65:320–327. Tatsukawa T, Chimura T, Miyakawa H, Yamaguchi K (2006) Involvement of basal protein kinase c and extracellular signal-regulated kinase 1/2 activities in constitutive internalization of AMPA receptors in cerebellar Purkinje cells. J Neurosci 26:4820–4825. Tatsukawa T, Chimura T, Miyakawa H, Yamaguchi K (2006) Involvement of basal protein kinase c and extracellular signal-regulated kinase 1/2 activities in constitutive internalization of AMPA receptors in cerebellar Purkinje cells. J Neurosci 26:4820-4825. Tehovnik EJ, Sommer MA, Chou IH, Slocum WM, Schiller PH (2000) Eye fields in the frontal lobes of primates. Brain Res Rev 32:413–448. Tempia F. Miniaci MC, Anchisi D, Strata P (1998) Postsynaptic current mediated by metabotropic glutamate receptors in cerebellar Purkinje cells. J Neurophysiol 80:520-528. Thach WT (1967) Somatosensory receptive fields of single units in cat cerebellar cortex. J Neurophysiol 30:675–696. Thier P, Dicke PW, Haas R, Barash S (2000) Encoding of movement time by populations of cerebellar Purkinje cells. Nature 405:72–76.

References

255

Thier P, Ilg UJ (2005) The neural basis of smooth-pursuit eye movements. Cur Opinion Neurobiol 15:645–652. Thompson RF (1988) The neural basis of basic associative learning of discrete behavioral responses. Trends Neurosci 11:152–155. Tian JR, Lynch JC (1996) Functionally defined smooth and saccadic eye movement subregions in the frontal eye field of cebus monkeys. J Neurophysiol 76:2740–2753. Tian L, Wen YQ, Li HZ, Zuo CC, Wang JJ (2000) Histamine excites rat cerebellar Purkinje cells via H2 receptors in vitro. Neurosci Res 36:61–66. Tighilet B, Lacour M (2001) Gamma amino butyric acid (GABA) immunoreactivity in the vestibular nuclei of normal and unilateral vestibular neurectomized cats. Eur J Neurosci 13:2255–2267. Tigyi G, Matute C, Miledi R (1990) Monoclonal antibodies to cerebellar pinceau terminals obtained after immunization with brain mRNA-injected Xenopus oocytes. PNAS USA 87:528–532. Titley HK, Heskin-Sweezie R, Chung J-Y, Kassardjian CD, Razik F, Broussard DM (2007) Rapid consolidation of motor memory in the vestibuloocular reflex. J Neurophysiol 98:3809–3812. Torriero S, Oliveri M, Koch G, Caltagirone C, Petrosin L (2004) Interference of left and right cerebellar rTMS with procedural learning. J Cog Neurosci 16:1605–1611. Toyama K, Tsukahara N, Kosaka K, Matsunami K (1970) Synaptic excitation of red nucleus neurons by fibres from interpositus nucleus. Exp Brain Res 11:187–198. Tracey DJ, Asanuma C, Jones EG, Porter R (1980) Thalamic relay to motor cortex: afferent pathways from brain stem, cerebellum and spinal cord in monkeys. J Neurophysiol 44:532. Traynelis SF, Silver RA, Cull-Candy SG (1993) Estimated conductance of glutamate receptor channels activated during EPSCs at the cerebellar mossy fiber-granule cell synapse. Neuron 11:279–289. Tsukahara N, Kosaka K (1968) The mode of cerebral excitation of red nucleus neurons. Exp Brain Res 5:102–117. Tsunoda K, Yamane Y, Nishizaka M, Tanifuji M (2001) Complex objects are represented in the macaque inferotemporal cortex by the combination of feature columns. Nat Neurosci 4:832–838. Uchino Y, Sasaki M, Sato H, Imagawa M, Suwa H, Isu N (1996) Utriculoocular reflex arc of the cat. J Neurophysiol 76:1896–1903. Udo M, Kamei H, Matsukawa K, Tanaka K (1982) Interlimb coordination in cat locomotion investigated with perturbation. Exp Brain Res 46:438–447. Udo M, Matsukawa K, Kamei H, Minoda K, Oda Y (1981) Simple and complex spike activities of Purkinje cells during locomotion in the cerebellar vermal zones of decerebrate cats. Exp Brain Res 41:292–300. Uemura T, Lee SJ, Yasumura M, Takeuchi T, Yoshida T, Ra M, Taguchi R, Sakimura K, Mishina M. (2010) Trans-synaptic interaction of GluRdelta2 and Neurexin through Cbln1 mediates synapse formation in the cerebellum. Cell 141:1068-1079. Ulloa A, Bullock D (2003) A neural network simulating human reach–grasp coordination by continuous updating of vector positioning commands. Neural Netw 16:1141–1160. Ulloa A, Bullock D, Rhodes BJ (2003) Adaptive force generation for precision-grip lifting by a spectral timing model of the cerebellum. Neural Netw 16:521–528. Ulloa A, Bullock D, Rhodes BJ (2010) A model of cerebellar adaptation of grip force during lifting. Generation for precision-grip lifting by a spectral timing model of the cerebellum. Neural Netw 16:521–528.

256

the cerebellum

Umemura H, Hayashi T, Inoue T, Nakanishi S, Mikoshiba K, Yamamoto T (1999) Involvement of protein tyrosine phosphatases in activation of the trimeric protein Gq/11. Oncogene 18:7399–7402. Uusisaari M, Obata K, Knöpfel T (2007) Morphological and electrophysiological properties of GABAergic and non-GABAergic cells in the deep cerebellar nuclei. J Neurophysiol 97:901–911. van Alphen AM, De Zeeuw CI (2002) Cerebellar LTD facilitates but is not essential for long-term adaptation of the vestibulo-ocular reflex. Eur J Neurosci 16:486–490. van der Steen J, Simpson JI, Tan J (1994) Functional and anatomic organization of three-dimensional eye movements in rabbit cerebellar flocculus. J Neurophysiol 72:31–46. van Hemmen LJ, Sejnowski TJ (2005) 23 problems in systems neuroscience (Computational Neuroscience Series). Oxford University Press. Oxford UK. van Vreeswijk C, Sompolinsky H (1996) Chaos in neuronal networks with balanced excitatory and inhibitory activity. Science 274:1724–1726. van Woerden GM, Hoebeek FE, Gao Z, Nagaraja RY, Hoogenraad CC, Kushner SA, Hansel C, De Zeeuw CI, Elgersma Y (2009) CaMKII controls the direction of plasticity at parallel fiber–Purkinje cell synapses. Nature Neurosci 12:823–825. Vandervert LR, Schlimp PH, Liu H (2007) How working memory and the cerebellum collaborate to produce creativity and innovation. Creativity Res J 19:1–18. Vervaeke K, Lörincz A, Gleeson P, Farinella M, Nusser Z, Silver RA (2010) Rapid desynchronization of an electrically coupled interneuron network with sparse excitatory synaptic input. Neuron 67:435–451. Vidal P-P, De Waele C, Vibert N, Mühlethaler M (1998) Vestibular compensation revisited. Otolaryngol 119:34–42. Vigot R, Batini C, Kado RT, Yamamori T (2002) Synaptic long-term depression (LTD) in vivo recorded on the rat cerebellar cortex. Archives Italiennes de Biologie 140(1):1–12. von Holst, E (1954) Relations between the central nervous system and peripheral organs. Br J Anim Behav 2:89–94. Voogd J (1964) The cerebellum of the cat [Ph.D. dissertation]. Assen, Germany: Van Gorcum. Voogd, J. (2004) Cerebellum and precerebellar nuclei. In Paxinos G and Mai JK (Eds.) The Human Nervous System. 2nd edition Amsterdam, Elsevier, pp. 321/392. Voogd J (2010) Cerebellar zones: a personal history. [Internet] Cerebellum Oct 22, 2010. Voogd J, Glickstein M (1998) The anatomy of the cerebellum. Trends Neurosci 21:370–375. Vos BP, Maex R, Volny-Luraghi A, De Schutter E (2000) Precise spike timing of tactile-evoked cerebellar Golgi cell responses: a reflection of combined mossy fiber and parallel fiber activation? In: De Zeeuw CI, Ruigrok TJH, Gerrits NM, editors. Prog Brain Res 124:95–105. Wadiche JI, Jahr CE (2005) Patterned expression of Purkinje cell glutamate transporters controls synaptic plasticity. Nat Neurosci 8:1329–1334. Walaas SI, Lai Y, Gorelick FS, De Camilli P, Moretti M, Greengard P (1988) Cell-specific localization of the a-subunit of calcium/calmodulin-dependent protein kinase II in Purkinje cells in rodent cerebellum. Mo Brain Res 4:233–242. Walter JT, Khodakhah K (2006) The linear computational algorithm of cerebellar Purkinje cells. J Neurosci 26:12861–12872. Walter JT, Khodakhah K (2009). The advantages of linear information processing for cerebellar computation. PNAS USA 106:4471–4476.

References

257

Wan X, Nakatani H, Ueno K, Asamizuya T, Cheng K, Tanaka K (2011) The neural basis of intuitive best next-move generation in board game experts. Science 331:341-346. Wang JJ, Kim JH, Ebner, TJ (1987) Climbing fiber afferent modulation during a visually guided, multi-joint arm movement in the monkey. Brain Res 410:323–329. Wang SS, Denk W, Hausser M (2000a) Coincidence detection in single dendritic spines mediated by calcium release. Nat Neurosci 3:1266–1273. Wang SS, Khiroug L, Augustine GJ (2000b) Quantification of spread of cerebellar long-term depression with chemical two-photon uncaging of glutamate. PNAS USA 97:8635–8640. Wang Y, Pillai S, Wolpaw JR, Chen XY (2006) Motor learning changes GABAergic terminals on spinal motoneurons in normal rats. Eur J Neurosci 23:141–150. Wang Y, Pillai S, Wolpaw JR, Chen XY (2009) H-reflex down-conditioning greatly increases the number of identifiable GABAergic interneurons in rat ventral horn. Neurosci Lett 452:124–129. Watanabe D, Hashimoto K, Suzuki N, Kano M, Shigemoto R, Hirano T, Toyama K, Kaneko S, Yokoi M, Moriyaoshi K, Suzuki M, Kobayashi K, Nagatsu T, Kreitman RJ, Pastan I, Nakanishi S (1998) Ablation of cerebellar Golgi cells disrupts synaptic integration involving GABA inhibition and NMDA receptor activation in motor coordination. Cell 95:17–27. Watanabe D, Nakanishi S (2003) mGluR2 Postsynaptically Senses Granule Cell Inputs at Golgi Cell Synapses. Neuron 39:821-829. Watanabe M, Nakamura M, Sato K, Kano M, Simon MI, Inoue Y (1998) Patterns of expression for the mRNA corresponding to the four isoforms of phospholipase Cb in mouse brain. Eur J Neurosci 10:2016–2025. Wegner DM (2002) The illusion of conscious will. Cambridge, MA: MIT Press. Welsh JP, Yamaguchi H, Zeng XH, Kojo M, Nakada Y, Takagi A, Sugimori M, Llinás RR (2005) Normal motor learning during pharmacological prevention of Purkinje cell long-term depression. PNAS USA 102:17166–17171. Westenbroek RE, Sakurai T, Elliott EM, Hell JW, Starr TV, Snutch TP, Catterall WA (1995) Immunochemical identification and subcellular distribution of the alpha 1A subunits of brain calcium channels. J Neurosci 15:6403–6418. Wharton SM, Payne JN (1985) Axonal branching in parasagittal zones of the rat olivocerebellar projection: a retrograde fluorescent double-labelling study. Exp Brain Res 58:183–1189. Wiener N (1948) Cybernetics or control and communication in the animal and the machine. Paris: Hermann & Cie Editeurs, Paris; Cambridge, MA: The Technology Press; New York: John Wiley & Sons Inc. Wiesendanger R, Wiesendanger M, Ruegg DG (1979) An anatomical investigation of the corticopontine projection in the primate (Macaca fascicularis and Sai miri seiureus). II. The projection from frontal and parietal association areas. Neuroscience 4: 747–765. Williams PT, Hilmas CJ (2010) Cholinergic effects on ocular flutter in guinea pigs following nerve agent exposure: a review [Internet]. JMedCBR 8:24 November. Available from: http://www.jmedcbr. org/issue_0801/PWilliams/PWilliams_11_10.html cited 21 Jan. 2011 Wilson VJ, Yoshida M (1969) Comparison of effects of stimulation of Deiters’ nucleus and medial longitudinal fasciculus on neck, forelimb, and hindlimb motoneurons. J Neurophysiol 32:743–758. Winkelman BH, Frens MA (2006) Motor coding in floccular climbing fibers. J Neurophysiol 95:2342–2351.

258

the cerebellum

Wise SP, Boussaoud D, Johnson PB, Caminiti R (1997) Premotor and parietal cortex: corticocortical connectivity and combinatorial computations. Annu Rev Neurosci 20:25–42. Wolpaw JR, Chen XY (2006) The cerebellum in maintenance of a motor skill: a hierarchy of brain and spinal cord plasticity underlies H-reflex conditioning. Learn Mem 13:208–215. Wolpert DM, Doya K, Kawato M (2003) A unifying computational framework for motor control and social interaction. Philos Trans R Soc London B 358:593–602. Wolpert DM, Kawato M (1989) Multiple paired forward and inverse models for motor control. Neural Netw 11:1317–1329. Xia J, Chung HI, Wihler C, Huganir RL, Linden DJ (2000) Cerebellar long-term depression requires PKC-mediated interactions between GluR2/3 and PDZ domain-containing proteins. Neuron 28:499–510. Xiong G, Nagao S, Kitazawa H (2010) Mossy and climbing fiber collateral inputs in monkey cerebellar paraflocculus lobulus petrosus and hemispheric lobule VII and their relevance to oculomotor functions. Neurosci Lett 468:282–386. [Epub 2009 Nov 10] Xu W, Edgley SA (2008) Climbing fibre-dependent changes in Golgi cell responses to peripheral stimulation. J Physiol (London) 586:4951–4959. Yamada J, Noda H (1987) Afferent and efferent connections of the oculomotor cerebellar vermis in the macaque monkey. J Comp Neurol 265:224–241. Yamada K, Fukaya M, Shimizu H, Sakimura K, Watanabe M (2001) NMDA receptor subunits GluRepsilon1, GluRepsilon3 and GluRzeta1 are enriched at the mossy fibre-granule cell synapse in the adult mouse cerebellum. Eur J Neurosci 13:2025–2036. Yamaguchi K (2009) An integrative kinetic model of AMPA-receptors constitutive trafficking and LTD/LTP in cerebellar Purkinje cell. Poster for 36th International Congress of Physiological Sciences (IUPS2009) Kyoto. J Physiol Sci 59 Suppl 1:183. Yamamori T, Mikawa S, Kado R (1995) Jun-B expression in Purkinje cells by conjunctive stimulation of climbing fibre and AMPA. NeuroReport 6:793–796. Yamamoto K, Kawato M, Kotosaka S, Kitazawa S (2007) Encoding of movement dynamics by Purkinje cell simple spike activity during fast arm movements under resistive and assistive force fields. J Neurophysiol 97:1588–1599. Yamamoto K, Kobayashi Y, Takemura A, Kawano K, Kawato M (1997) A mathematical model that reproduces vertical ocular following responses from visual stimuli by reproducing the simple spike firing frequency of Purkinje cells in the cerebellum. Neurosci Res 29:161–169. Yamamoto K, Kobayashi Y, Takemura A, Kawano K, Kawato M (2002) Computational studies on acquisition and adaptation of ocular following responses based on cerebellar synaptic plasticity. J Neurophysiol 87:1554–1571. Yamasaki T, Kawaji1 K, Ono K, Bito H, Hirano T, Osumi N, Kengaku M (2001) Pax6 regulates granule cell polarization during parallel fiber formation in the developing cerebellum. Dev 128:3133–3144. Yamazaki T, Nagao S (2008) A computational model of the cerebellum for motor-memory transfer. Society for Neuroscience 38th Annual Meeting. Washington. Abstract Viewer/Itinerary Planner. 471.6. Yamazaki T, Tanaka S (2005) Neural modeling of an internal clock, Neural Computation 17:1032–1058.

References

259

Yamazaki T, Tanaka S (2007) A spiking network model for passage-of-time representation in the cerebellum. Eur J Neurosci 26:2279–2292. Yanagihara D, Kondo I (1996) Nitric oxide plays a key role in adaptive control of locomotion in cat. PNAS USA 93:13292–13297. Yanagihara D, Udo M (1994) Climbing fiber responses in cerebellar vermal Purkinje cells during perturbed locomotion in decerebrate cats. Neurosci Res 19:245–24. Yano R, Nakazawa K, Kado RT, Karachot L, Ito M, Mikawa S, Komine Y, Yamamori T (1996) Cerebellar long-term plasticity and gene expression. In: Ito M, Miyashita Y, editors. Integrative and molecular approach to brain function. Amsterdam: Elsevier. pp. 35–44. Yeo CH, Hardiman MJ, Glickstein M (1985a) Classical conditioning of the nictitating membrane response of the rabbit. I. Lesions of the cerebellar nuclei. Exp Brain Res 60:87–98. Yeo CH, Hardiman MJ, Glickstein M (1985b) Classical conditioning of the nictitating membrane response of the rabbit. II. Lesions of the cerebellar cortex. Exp Brain Res 60:99–113. Yeo CH, Hardiman MJ, Glickstein M (1985c) Classical conditioning of the nictitating membrane response of the rabbit. IIl. Connections of cerebellar lobule HVI. Exp Brain Res 60:114–126. Yttri E, Smith A, Reid E, Thach, WT (2006) Inactivation of parvocellular red nucleus impairs prism adaptation to and learning of contralateral gaze-reach shift. (Abstr.) 440.15/K8 Society for Neuroscience Annual Meeting, Atlanta. Yuzaki M, Furuichi T, Mikoshiba K, Kagawa Y (1994) A stimulus paradigm inducing long-term desensitization of AMPA receptors evokes a specific increase in BDNF mRNA in cerebellar slices. Learn Mem 1:230–242. Zagon IS, McLaughlin PI, Smith S (1977) Neural populations in the human cerebellum: estimations from isolated cell nuclei. Brain Res 127:279–282. Zee DS, Tusa RJ, Herdman SJ, Butler PH, Gucer G (1987) Effects of occipital lobectomy upon eye movements in primate. J Neurophysiol 58:883–907. Zhang J, Li B, Yu L, He Y-C, Li H-Z, Zhu J-N, Wang J-J (2011) A role for orexin in central vestibular motor control. Neuron 69:793–804. Zhang W, Linden DJ (2006) Long-term depression at the mossy fiber–deep cerebellar nucleus synapse. J Neurosci 26:6935–6944. Zhang W, Shin JH, Linden DJ (2004) Persistent changes in the intrinsic excitability of rat deep cerebellar nuclear neurons induced by EPSP or IPSP bursts. J Physiol (London) 561:703–719. Zhao Y, Baker H, Walaas SI, Sudol M (1991) Localization of p62 protein in mammalian neural tissues. Oncogene 6:1725–1733. Zhu J-N, Yung W-H, Chow BK-C, Chan Y-S, Wang J-J (2006) The cerebellar-hypothalamic circuits: Potential pathways underlying cerebellar involvement in somatic-visceral integration. Brain Res Rev 52:93-106.

This page intentionally left blank

Index A action controllers, premotor cortex, 181-182 action recognition, mirror neurons, 189 action schema, 183-184 adaptive control, 10, 38-40 cognitive functions, 193-194 activity in the cerebellum, 198-201 cerebellar internal model for thoughts, 196 cerebral cortical model for thoughts, 195-196 explicit/implicit thoughts, 197-198 mental disorders associated with cerebellar dysfunction, 201-203 neural systems, 193-195 eye-blink conditioning, 132-136 locomotion, 130-132

models ocular reflexes, 139-145 prototypes, 146-149 saccadic eye movement, 146-147 somatic reflexes, 147-148 nociceptive withdrawal reflex, 127-128 OKR (Optokinetic EyeMovement Response), 114 saccades, 119-120 sympathetic reflexes, 136-138 VOR (Vestibuloocular Reflex) climbing fiber input, 110-111 eye movement-related signals, 111-112 flocculus, 107-109 memory sites, 112 vestibular mossy fiber input, 110 adaptive filter model cerebellum, 95 Fujita, 34-35

261

262

index

afferent fibers, 44-45 afferent muscles, 121 Albus, James, 33 alpha-amino-3-hydroxy5-methyl-4-isoxazolone propionate. See AMPA alpha-motoneurons, 121 AMPA receptors (alpha-amino-3-hydroxy5-methyl-4-isoxazolone propionate), 45 conjunctive LTD, 74-75 mediation of mossy fibergranule cell synapses, 45 mediation of parallel fiber input to Golgi cells, 54 archicerebellum, 23 arm movements, voluntary motor control, 151-154 autism, 201 autonomic reflexes. See cardiovascular function

B B-zone Purkinje cells, locomotion, 131 basal ganglia, 2, 210-211 basket cells, 51-53, 83 BDNF (brain-derived neurotrophic factor), 80 beaded fibers cells of origin, 66 input/output pathways of cerebellar cortex, 50

Bergmann glial cells, 57-58 beta-motoneurons, 121 blocking studies, eye-blink conditioning, 133 body schema, 183-184 brain divisions, 2 brain-derived neurotrophic factor (BDNF), 80 brainstem, 2, 68

C c-Fos/Jun-B, 79 C-type microcomplex, adaptive control, 146-149 C3 zone, 104 Ca2+ surge (Purkinje cell dendrites), 71-74 Ca2+/calmodulin-dependent protein kinase (CaMKII), 78 calretinin-expressed unipolar brush cells, 47 CaMKII (Ca2+/calmodulindependent protein kinase), 78 canal specific pathways, VOR (Vestibuloocular Reflex), 105 cannabinoid-receptormediated presynaptic LTD, 82 cardiovascular function somatosympathetic reflex, 137 vestibulosympathetic reflex, 136-137 caudate neurons, disinhibition of basal ganglia components, 210

index

cells Bergmann glial, 57-58 globular (cerebellar cortex granular layer), 56-57 Golgi, 53-55 granule, 45-46 Lugaro, 55-56 NG2+, 58-59 origin beaded fibers, 66 mossy fibers, 60-63 Purkinje B-zone, 131 Ca2+ surge in dendrites, 71-74 convergence of climbing and parallel fibers, 32 inhibitory neurons, 30 input/output pathways of cerebellar cortex, 47-49 models for neuronal circuits, 92-94 synapses with parallel fibers, 70 synaptic plasticity, 81-83 unipolar brush, 46-47 central nervous system (CNS) components, 2 neural networks, 8-9 neuronal circuits decomposition and reconstruction, 1-7 neurons and synapses, 7-8 systems control mechanisms, 9 cognitive functions, 19-20

263

feedback control systems, 10 intelligence and conscious awareness, 20-21 model-based control systems, 10-11 reflexes, 12-18 voluntary movements, 12-13, 18-19 central pattern generator (CPG) mechanism, 15 cerebellar control ocular reflexes, 105, 117 OFR (Ocular Following Response), 116 OKR (Optokinetic EyeMovement Response), 113-115 saccades, 117-120 VOR (Vestibuloocular Reflex), 105-113 voluntary eye movement, 159 DMFC (dorsomedial frontal cortex), 165-166 frontal eye field, 159-165 voluntary motor control hand grip, 154-156 instruction signals, 157-158 load compensation, 150 multijoint arm movements, 151-154 operant conditioning, 156-157 reaction time, 151

264

index

cerebellar cortex divisions, 2, 23-24 inhibitory neurons, 51 basket/stellate cells, 5153 Bergmann glial cells, 57-58 Golgi cells, 53-55 Lugaro cells, 55-56 small globular cells in granular layer, 56-57 input and output pathways, 44 beaded fibers, 50 climbing fibers, 49-50 granule cells, 45-46 mossy fibers, 44-45 Purkinje cells, 47-49 unipolar brush cells, 46-47 cerebellar cortical synapses, synaptic plasticity, 84 cerebellar internal models, 11, 196 cerebellar model articulation controller (CMAC), 99 cerebellar mutism, 201 cerebellar nuclear neurons, 67-68 cerebellar roles, motor actions, 185-187 cerebellar valvula, 25 cerebellar/vestibular nuclear neurons, synaptic plasticity, 84-85 cerebellothalamocortical circuit, motor actions, 186

The Cerebellum and Adaptive Control, 21 The Cerebellum and Neural Control, 21 The Cerebellum as a Neuronal Machine, 30 cerebral cortex divisions, 2 models for cognitive function, 11, 195-196 cerebrocerebellar communication loop, 23, 41-42, 168 cerebrum divisions, 2 circuit models, adaptive control of ocular reflexes, 142-144 climbing fiber signals forward/inverse models, 174-175 VOR adaptation, 110-111 climbing fiber-Purkinje cell synapses, homosynaptic LTD, 83 climbing fibers conjunctive stimulation, 72 input/output pathways of cerebellar cortex, 49-50 clock model, granule cellGolgi cell loop, 95 CMAC (cerebellar model articulation controller), 99 CNS (central nervous system) components, 2 neural networks, 8-9

index

neuronal circuits decomposition and reconstruction, 1-7 neurons and synapses, 7-8 systems control mechanisms, 9 cognitive functions, 19-20 feedback control systems, 10 intelligence and conscious awareness, 20-21 model-based control systems, 10-11 reflexes, 12-18 voluntary movements, 12-13, 18-19 codons (Marr), 90 cognitive functions, 19-20, 193-195 activity in the cerebellum, 198-201 cerebellar internal model for thoughts, 196 cerebellum as a neuronal machine, 43 cerebral cortical model for thoughts, 195-196 explicit/implicit thoughts, 197-198 mental disorders associated with cerebellar dysfunction, 201-203 components of the CNS, 2 compounded reflexes, 14-15

265

conditioned reflex pathway (eye-blink reflex), 132 conjunctive LTD (long-term depression), 69 motor learning, 206-207 properties, 69-71 signal transduction pathways, 71 AMPA receptors, 74-75 brain-derived neurotrophic factor (BDNF), 80 c-Fos/Jun-B, 79 Ca2+ surge in Purkinje cell dendrites, 71-74 Ca2+/calmodulindependent protein kinase (CaMKII), 78 glial fibrillary acidic protein (GFAP), 79 glutamate receptor δ2 (GluR δ2), 76-77 insulin-like growth factor-1 (IGF-1), 80 Nitric oxide (NO) synthase, 75-76 PKCα and lipid-signaling cascade, 74 protein phosphatases, 78 protein tyrosine kinases, 77 synaptic plasticity, 205-206 conscious awareness, 20-21 context dependencies, functional stretch reflex, 126 contralateral inferior olive complex, 63-65

266

index

contralateral lateralization, frontocerebellar circuits, 194 control systems, 9 adaptive cognitive functions, 193-203 eye-blink conditioning, 132-136 locomotion, 130-132 models, 139-149 nociceptive withdrawal reflex, 127-128 OKR (Optokinetic Eye-Movement Response), 114 saccades, 119-120 sympathetic reflexes, 136-138 VOR (Vestibuloocular Reflex), 107-112 cognitive functions, 19-20, 193-194 activity in the cerebellum, 198-201 cerebellar internal model for thoughts, 196 cerebral cortical model for thoughts, 195-196 explicit/implicit thoughts, 197-198 mental disorders associated with cerebellar dysfunction, 201-203 combination with microcomplexes, 176

feedback control systems, 10 intelligence and conscious awareness, 20-21 model-based control systems, 10-11 reflexes, 12-18 voluntary movements, 12-13, 18-19 controlled objects (G) (control systems), 9 controllers (g) (control systems), 9 convergence, Purkinje cells, climbing and parallel fibers, 32 cortico-nuclear microcomplex, 102 CPG (central pattern generator) mechanism, 15 Craik’s mental models, 195 CREB (Cyclic AMP response element-binding protein), 88-89 crus I and crus II, eye movements, 159 cuneocerebellar tract, 62 curly cells (inferior olive complex), 64 Cyclic AMP response element-binding protein (CREB), 88-89

index

D DAG (1, 2-diacylglycerol), 71 DAO (dorsal accessory olive) subdivision, 63 declarative memory, 20 decomposition, neuronal circuits, 1-7 defective central coherence, 202 Deiters neurons, 126 Deiters, Otto, 30 Delta/Notch-like EGF-related receptor (DNER), 57 Descartes, René, clock metaphor for complex mechanisms, 1 descending control, stretch reflex, 124-126 development, 211-212 developmental dyslexia, 202 digital dexterity, neuronal mechanisms, 154-156 disinhibitory pathways, basal ganglia components, 210 divisions brain, 2 cerebellar cortex, 2, 23-24 cerebral cortex, 2 cerebrum, 2 lobules I-X, 23 DMFC (dorsomedial frontal cortex), 165-166 DNER (Delta/Notch-like EGF-related receptor), 57 dorsal accessory olive (DAO) subdivision, 63

267

dorsal spinocerebellar tract (DSCT), 60 dorsomedial frontal cortex (DMFC), 165-166 Dow, Robert, 43 DSCT (dorsal spinocerebellar tract), 60 dual memory mechanism model, 102 dysmetria, 26

E EAAT4, 57 Eccles, John, excitatory and inhibitory neurons, 29 efference copy, 142 EPSCs (excitatory postsynaptic currents), 29 EPSPs (excitatory postsynaptic potentials), 29 equilibrium point hypothesis, myotatic reflexes, 124 Erasistratus, 22 evolution cerebellar circuits, 208-209 lobular structures, 23 excitatory neurons, 29 excitatory postsynaptic currents (EPSCs), 29 excitatory postsynaptic potentials (EPSPs), 29 excitatory projection neurons, 67 excitatory synapses, cerebellar/vestibular nuclear neurons, 85

268

index

explicit/implicit thoughts, 197-198 eye movements saccades, 117-120 voluntary control, 159 DMFC (dorsomedial frontal cortex), 165-166 frontal eye field, 159-165 VOR adaptation, 111-112 eye-blink conditioning, 132-136 eyeballs, inverse model, 144145

F fear conditioning, Purkinje cells, 83 feedback control systems, 10 feedforward control systems, 10, 39 flocculonodular lobe, 23 flocculus, connection to VOR, 107-109 Flourens, Jean Pierre, 26 forward models, voluntary motor control, 167-170, 177-178 frontal eye field, voluntary eye movement, 159 saccades, 163-164 smooth pursuit, 160-162 vergence, 164-165 frontocerebellar circuits, contralateral lateralization, 194 Fujita, adaptive filter model of the cerebellum, 34-35

functional stretch reflex, 126 fusimotor set hypothesis, muscle contractions, 123 future research, 205 conjunctive LTD, 205-207 development, 211-212 evolution in cerebellar circuits, 208-209 internal models, 208 IO function, 207 motor actions, 209-210 nuclear memory, 208

G G (controlled objects) (control systems), 9 g (controllers) (control system), 9 G-protein (Gq/11), 71 GABA-releasing inhibitory neurons, Purkinje cells, 30 GABAergic neurons, 67 gamma-motoneurons, stretch reflex, 121 GFAP (glial fibrillary acidic protein), conjunctive LTD, 79 giant neurons, 30 gigantocerebellum, 25 glial cells, Bergmann, 57-58 glial cells, NG2+, 58-59 glial fibrillary acidic protein (GFAP), conjunctive LTD, 79 globular cells (cerebellar cortex granular layer), 56-57 GluR δ2 (glutamate receptor δ2), conjunctive LTD, 76-77

index

Glutamate, transmission of climbing fibers, 49 glutamate receptors, conjunctive LTD, 70, 76-77 glycinergic neurons, 67 Golgi cells, 53-55, 84 Gq/11 (G-protein), 71 Granit, Ragner, 28 granule cell-Golgi cell loops, 94-97 granule cells, 45-46 grasp-lift-hold tasks, neuronal mechanisms, 155

H half-fused control systems, 13-14 hallucination, 202 hand grip, voluntary motor control, 154-156 Hebb, Donald, 8 history cerebellar involvement in motor skills, 25-27 cerebellum as neuronal machine adaptive control, 38-40 cognitive functions, 43 discoveries in the 1960s, 29-30 internal models, 41-42 LTD (long-term depression), 35-37 Marr-Albus model, 33-35 morphological map of the cerebellum, 22-25 Purkinje cells, 27-28

269

Holmes, Gordon, 26 homosynaptic LTD climbing fiber-Purkinje cell synapses, 83 parallel fiber-Purkinje cell synapses, 81 hybrid control reflexes and voluntary movements, 13-14 voluntary movements, 179

I Ia muscle afferent, 121 Ib muscle afferent, 121-122 IGF-1 (Insulin-like growth factor-1), conjunctive LTD, 80 imitation hypothesis, integration of reflexes, 17 imitation learning, mirror neurons, 189 implicit/explicit thoughts, 197198 in vivo Purkinje cells, 53 inferior olive-climbing fiber system, 97-99 information storage capacity, Purkinje cells, 92 inhibiting mechanism, basal ganglia, 210 inhibitory postsynaptic currents (IPSCs), 29 inhibitory postsynaptic potentials (IPSPs), 29 inhibitory synapses, 84

270

index

inositol-trisphosphate (IP3), 71 input pathways, cerebellar cortex, 44 beaded fibers, 50 climbing fibers, 49-50 granule cells, 45-46 mossy fibers, 44-45 Purkinje cells, 47-49 unipolar brush cells, 46-47 input-specificity, conjunctive LTD, 70 instruction signals forward/inverse models, 172-174 voluntary motor control, 157-158 instrumental conditioning, voluntary motor control, 156-157 insulin-like growth factor-1 (IGF-1), conjunctive LTD, 80 integration of reflexes, 13 compounded reflexes, 14 half-fused control systems, 13 hybrid control, 13 imitation hypothesis, 17 multi-input control systems, 13 mutual interactions, 14 nesting, 16 neuromodulation, 16 “The Integrative Action of the Nervous System,” 13 intelligence, 20-21

intention tremor, 26 internal forward model model-based control system, 11 voluntary motor control, 167-170 climbing fiber signals, 174-175 sensory/motor signals, 172-174 internal inverse model model-based control system, 11 voluntary motor control, 170-172 climbing fiber signals, 174-175 sensory/motor signals, 172-174 internal models cerebellum as a neuronal machine, 41-42 future research, 208 voluntary motor control, 167 internal forward model, 11, 167-175 internal inverse model, 11, 170-175 MOSAIC (modular selection and identification for control), 175 sensory cancellation, 178-180 transition from reflex control to voluntary, 176-179

index

intuitive thought, 21 inverse models eyeballs, 144-145 voluntary motor control, 170-172 IO function, 207 IO neurons, 97-99 IP3 (inositol-trisphosphate), 71 IPSCs (inhibitory postsynaptic currents), 29 IPSPs (inhibitory postsynaptic potentials), 29

J–K–L Jones, Geoffrey Melvill, 39 landmarks (longitudinal zones of cerebellum), 23-24 language acquisition, cerebellar mutism, 201 lateral reticular nucleus, mossy fibers, 62 lateral vestibular nucleus (LVN), 125 lateral vestibulospinal tract (LVST), 125 length-servo-assisted motion hypothesis, muscle contractions, 123 lesions studies, motor skills, 25-27 limb withdrawal reflex, 127-128 limbic system, 2 lipid signaling cascade, 74 liquid state machine model, 100-101

271

load compensation, voluntary motor control, 150 lobular structures, 23 lobules, 23 lobulus petrosus of the paraflocculus, 159 locomotion, 15, 128-132 long-lasting memory, synaptic plasticity, 88-89 long-latency stretch reflex, 126 long-term depression (LTD), 35-37 basket/stellate cells, 83 conjunctive, 69 motor learning, 206-207 properties, 69-71 signal transduction pathways, 71-80 synaptic plasticity, 205-206 excitatory synapses, 85 homosynaptic climbing fiber-Purkinje cell synapses, 83 parallel fiber-Purkinje cell synapses, 81 inhibitory synapses, 84 long-term potentiation (LTP), 69 basket/stellate cells, 83 mossy fiber-granule cell synapses, 84 longitudinal zones of cerebellum, 23-24

272

index

LTD (long-term depression), 35-37 basket/stellate cells, 83 conjunctive, 69 motor learning, 206-207 properties, 69-71 signal transduction pathways, 71-80 synaptic plasticity, 205-206 excitatory synapses, 85 homosynaptic climbing fiber-Purkinje cell synapses, 83 parallel fiber-Purkinje cell synapses, 81 inhibitory synapses, 84 LTP (long-term potentiation), 69 basket/stellate cells, 83 mossy fiber-granule cell synapses, 84 parallel fiber-Purkinje cell synapses, 81-82 Lugaro cells, 55-56 LVN (lateral vestibular nucleus), 125 LVST (lateral vestibulospinal tract), 125

M magnocellular nuclei (lateral reticular nucleus), 62 magnocellular red nucleus neurons, 30

major part of the lateral reticular nucleus (mLRN), 62 MAO (medial accessory olive) subdivision, 63 Marr, David, 33 Marr-Albus model, cerebellum as a neuronal machine, 33-35 mechanisms, neuronal circuitry, 4 medial accessory olive (MAO) subdivision, 63 medial superior temporal area (MST), 116 memory synaptic plasticity, 88-89 sites VOR adaptation, 112 VOR/OKR adaption, 144 mental activity, 193 activity in the cerebellum, 198-201 cerebellar internal model for thoughts, 196 cerebral cortical model for thoughts, 195-196 disorders associated with cerebellar dysfunction, 201-203 explicit/implicit thoughts, 197-198 neural systems, 193-195 mental disorders, association with cerebellar dysfunction, 201-203

index

mental models cognitive functions, 19 cortical circuits, 195 mesencephalic-locomotorcenter-evoked locomotion, 130 mGluR1-initiated signal transduction conjunctive LTD, 73 mGluR1a-expressed unipolar brush cells, 47 microcomplexes, 102-104, 176, 205 middle temporal area (MT), 116 mirror neurons, 187-189 mLRN (major part of the lateral reticular nucleus), 62 model-based control systems, 10-11 models adaptive control systems ocular reflexes, 139-145 prototypes, 146-149 saccadic eye movement, 146-147 somatic reflexes, 147-148 cognitive functions cerebellar internal model, 196 cerebral cortical model, 195-196 network models for neuronal circuits, 90 granule cell-Golgi cell loop, 94-97

273

inferior olive-climbing fiber system, 97-99 mossy fiber-granule cell relay, 90-92 multilayered network models, 99-101 Purkinje cell models, 92-94 nuclear circuits, 101-102 voluntary motor control, 167 internal forward model, 167-175 internal inverse model, 170-175 MOSAIC (modular selection and identification for control), 175 sensory cancellation, 178-180 transition from reflex control to voluntary, 176-179 modular selection and identification for control (MOSAIC), 175 morphological map, 22-25 MOSAIC (modular selection and identification for control), 175 mossy fiber-granule cell relay (neuronal circuit network model), 90-92 mossy fiber-granule cell synapses, 84

274

index

mossy fibers cells of origin, 60-63 input/output pathways of cerebellar cortex, 44-45 motoneurons, stretch reflex, 121 motor actions action schema, 183-184 cerebellar roles, 185-187 future research, 209-210 mirror neurons, 187-189 model-based control, 184 premotor cortex action controllers, 181-182 sensory cancellation, 189-190 tools, 190-192 voluntary movements, 18-19 motor control parallels to social interaction, 196 voluntary, 150 hand grip, 154-156 instruction signals, 157-158 internal models, 167-180 load compensation, 150 multijoint arm movements, 151-154 operant conditioning, 156-157 reaction time, 151 motor learning, conjunctive LTD, 206-207 motor schema, 183-184 motor signals, forward/inverse models, 172-174 motor skills, 25-27

MST (medial superior temporal area), 116 MT (middle temporal area), 116 multi-input reflex, 13-14 multicompartmental model, recognition capacity of Purkinje cells, 93 multijoint arm movements, voluntary motor control, 151-154 multilayered network models, neuronal circuits, 99-101 multiplicity, synaptic plasticity, 81 basket/stellate cells, 83 cerebellar cortical synapses, 84 cerebellar/vestibular nuclear neurons, 84-85 Purkinje cells, 81-83 muscle afferents, 121 muscle contractions, 123 mutism, 201 mutual interactions, integration of reflexes, 14 myotatic reflexes, 124

N N-ethylmaleimide-sensitive fusion (NSF), 82 N-methyl-D-aspartate (NMDA) receptors activation, 84 mediation of mossy fibergranule cell synapses, 45 mediation of parallel fiber input to Golgi cells, 54

index

negative feedback systems, 41 negative force feedback hypothesis, muscle contractions, 123 negative length feedback hypothesis, muscle contractions, 123 neocerebellum, 193 neocortex, 2 nesting reflexes, 16 network models neuronal circuits, 8-9, 90 granule cell-Golgi cell loop, 94-97 inferior olive-climbing fiber system, 97-99 mossy fiber-granule cell relay, 90-92 multilayered network models, 99-101 Purkinje cell models, 92-94 nuclear circuits, 101-102 neural circuits cognitive functions, 193-195 eye-blink reflex, 132-136 reflexes myotatic reflex, 124-126 nociceptive withdrawal reflex, 127 stretch reflex, 121-127 sympathetic reflexes, 136-138 somatosympathetic reflex, 137 vestibulosympathetic reflex, 136-137

275

neural integrator, VOR (Vestibuloocular Reflex), 107 neural networks, 8-9 neurobiotin, 64 neurocomputer (neuron assembly), 9 neuromodulation, 16 neuronal circuits, 1 concluding thoughts, 204-205 decomposition and reconstruction, 1-7 input and output pathways, 44 beaded fibers, 50 climbing fibers, 49-50 granule cells, 45-46 mossy fibers, 44-45 Purkinje cells, 47-49 unipolar brush cells, 46-47 network models, 90 granule cell-Golgi cell loop, 94-97 inferior olive-climbing fiber system, 97-99 mossy fiber-granule cell relay, 90-92 multilayered network models, 99-101 Purkinje cell models, 92-94 neurons and synapses, 7-8 pre- and postcerebellar cortex neurons, 60 cells of origin of beaded fibers, 66

276

index

cells of origin of mossy fibers, 60-63 cerebellar nuclear neurons, 67-68 contralateral inferior olive complex, 63-65 preolivary nuclei, 65-66 vestibular nuclear neurons, 68 revealing mechanisms through reconstruction, 4 structure-function relationships, 1-21 synaptic plasticity as memory element, 33-35 voluntary eye movement, 159 DMFC (dorsomedial frontal cortex), 165-166 frontal eye field, 159-165 voluntary motor control, 150 hand grip, 154-156 instruction signals, 157-158 load compensation, 150 multijoint arm movements, 151-154 operant conditioning, 156-157 reaction time, 151 VOR adaptation, 110-111 neuronal machine, history, 29 adaptive control, 38-40 cognitive functions, 43 discoveries in the 1960s, 29-30

internal models, 41-42 LTD (long-term depression), 35-37 Marr-Albus model, 33-35 neurons, 7-8 assembly, 8-9 brainstem, 68 Deiters, 126 DSCT (dorsal spinocerebellar tract), 60-62 excitatory, 29 granule cells, 45-46 inhibitory, 29 basket/stellate cells, 51-53 Bergmann glial cells, 57-58 Golgi cells, 53-55 Lugaro cells, 55-56 small globular cells in granular layer, 56-57 IO, 97-99 mirror, 187-189 orexinergic, 66 pre- and postcerebellar cortex, 60 cells of origin of beaded fibers, 66 cells of origin of mossy fibers, 60-63 cerebellar nuclear neurons, 67-68 contralateral inferior olive complex, 63-65

index

preolivary nuclei, 65-66 vestibular nuclear neurons, 68 synaptic plasticity, 81 basket/stellate cells, 83 cerebellar cortical synapses, 84 cerebellar/vestibular nuclear neurons, 84-85 conjunctive LTD (longterm depression), 69-80 persistency, 86-87 protein synthesis, 88-89 Purkinje cells, 81-83 VSCT (ventral spinocerebellar tract), 60-62 neuropeptides, 66 NG2+ cells, 58-59 Nitric oxide (NO) synthase, conjunctive LTD, 75-76 NMDA (N-methyl-Daspartate) receptors activation, 84 mediation of mossy fibergranule cell synapses, 45 mediation of parallel fiber input to Golgi cells, 54 NO (Nitric oxide) synthase, conjunctive LTD, 75-76 nociceptive withdrawal reflex, 127-128 non-declarative memory, 20 noxious stimuli, nociceptive withdrawal reflex, 127-128

277

NRTP (nucleus reticularis tegmenti pontis), 63 NSF (N-ethylmaleimidesensitive fusion), 82 nuclear circuits, 101-102 nuclear memory, future research, 208 nucleo-olivary inhibitory projection, 133 nucleus, 8 nucleus reticularis tegmenti pontis (NRTP), 63

O Ocular Following Response (OFR), 116, 140-141 ocular reflexes, 105 adaptive control system models, 139 circuit models, 142-143 inverse model of the eyeballs, 144-145 memory sites, 144 OFR, 140-141 OKR, 141 VOR, 139-140 cerebellar control, 117 OFR (Ocular Following Response), 116 OKR (Optokinetic EyeMovement Response), 113-115 saccades, 117-120

278

index

VOR (Vestibuloocular Reflex), 105-106 control system, 107-112 neural integrator, 107 velocity storage, 107 vestibular compensation, 112-113 OFR (Ocular Following Response), 116, 140-141 OKR (Optokinetic EyeMovement Response), 113-115, 141 oligodendrocyte precursor cells, 58 operant conditioning, voluntary motor control, 156-157 Optokinetic Eye-Movement Response (OKR), 113-115, 141 orexinergic neurons, 66 output pathways, cerebellar cortex, 44 beaded fibers, 50 climbing fibers, 49-50 granule cells, 45-46 mossy fibers, 44-45 Purkinje cells, 47-49 unipolar brush cells, 46-47

P P/Q channels, 71 paleocerebellum, 23 paraflocculus/flocculus, eye movements, 159

parallel-fiber Purkinje cell synapses cannabinoid-receptormediated presynaptic LTD, 82 conjunctive LTD, 70 homosynaptic LTD, 81 presynaptic LTP, 81-82 parietal association cortex, motor actions, 185 parvocellular nuclei (lateral reticular nucleus), 62 performance, cerebellar mutism, 201 peripheral stimuli, voluntary motor control, 13 persistency, synaptic plasticity, 86-87 pharmacological agents effects on eye-blink conditioning, 134 blocking conjunctive LTD, 109 parallel fiber stimulation, 70 Phillips, Charles, 28 phosphatidyl-inositol 4,5diphosphate (PIP2), 71 Piaget, Jean, 20 pinceau, 51 PIP2 (phosphatidyl-inositol 4,5-diphosphate), 71 PKA-mediated phosphorylation of RIM1a, presynaptic LTP, 82 PKCa, conjunctive LTD, 74

index

PO (principal olive) subdivision, 63 pontine nuclei, 62-63 postcerebellar cortex neurons, 60 cells of origin beaded fibers, 66 mossy fibers, 60-63 cerebellar nuclear neurons, 67-68 contralateral inferior olive complex, 63-65 preolivary nuclei, 65-66 vestibular nuclear neurons, 68 posterior vermis, saccade control, 118 posterolateral fissure, 23 postsynaptic density (PSD), 47 posture, cerebellar control, 127 PP1, 78 PP2, 78 precerebellar cortex neurons, 60 cells of origin beaded fibers, 66 mossy fibers, 60-63 cerebellar nuclear neurons, 67-68 contralateral inferior olive complex, 63-65 preolivary nuclei, 65-66 vestibular nuclear neurons, 68

279

prefrontal cortex, 193 premotor cortex, 18, 181-182 preolivary nuclei, 65-66 presupplementary eye field, 165 presynaptic LTD, 82 presynaptic LTP, 81-82 primary fissure, 23 primary motor cortex as controller for voluntary movement, 167 cerebrocerebellar communication loop, 168 forward-model-based control, 177-178 sites for instructional signals, 158 voluntary eye movement, 159 DMFC (dorsomedial frontal cortex), 165-166 frontal eye field, 159-165 voluntary motor control, 150 hand grip, 154-156 instruction signals, 157-158 load compensation, 150 multijoint arm movements, 151-154 operant conditioning, 156-157 reaction time, 151 primate cerebellum, 24-25 principal olive (PO) subdivision, 63 prism distortion, 202

280

index

progression, decomposition/ reconstruction cycles, 5 properties, conjunctive LTD, 69-71 protein phosphatases, conjunctive LTD, 78 protein synthesis, synaptic plasticity, 88-89 protein tyrosine kinases, conjunctive LTD, 77 prototypes, adaptive control of reflexes, 146-149 PSD (postsynaptic density), 47 Purkinje cells, 27-28 B-zone, 131 Ca2+ surge in dendrites, 71, 73-74 convergence of climbing and parallel fibers, 32 induced inhibitory postsynaptic potentials, 30 inhibitory neurons, 30 input/output pathways of cerebellar cortex, 47-49 models for neuronal circuits, 92-94 synapses with parallel fibers, 70 synaptic plasticity, 81-83 Purkinje, Jan E., 27

Q–R reaching experiments, voluntary motor control, 151-154

reaction time, voluntary motor control, 151 readout neurons, 100 rebound depolarization, cerebellar/vestibular nuclear neurons, 85 rebound potentiation, Purkinje cells, 83 recognition capacity, Purkinje cells, 93-94 reconstruction, neuronal circuits, 1-7 reflexes, 12 compounding, 15 hybrid control, 13-14 integration of, 13-18 compounded reflexes, 14 half-fused control systems, 13 hybrid control, 13 imitation hypothesis, 17 multi-input control systems, 13 mutual interactions, 14 nesting, 16 neuromodulation, 16 ocular, 105 adaptive control system models, 139-145 cerebellar control, 117 OFR (Ocular Following Response), 116 OKR (Optokinetic EyeMovement Response), 113-115

index

saccades, 117-120 VOR (Vestibuloocular Reflex), 105-113 prototypes of adaptive control, 146-149 somatic and autonomic adaptive control system models, 147-148 eye-blink conditioning, 132-136 locomotion, 128-132 nociceptive withdrawal reflex, 127-128 stretch reflex, 121-127 sympathetic, 136-138 transition from reflex control to voluntary, 176-179 research for the future conjunctive LTD, 205-207 development, 211-212 evolution in cerebellar circuits, 208-209 internal models, 208 IO function, 207 motor actions, 209-210 nuclear memory, 208 respiration, cerebellar involvement, 137-138 reversible neurotransmission blocking (RNB), 136 right-left transversal lobular structure, 23 RNB (reversible neurotransmission blocking), 136

281

Robinson, David, 39 Rolando, Luigi, 26 Rosenblatt, Frank, 8 rostral spinocerebellar tract, 62

S saccades adaptive control system models, 146-147 eye movement, 117-120 frontal eye field, 163-164 Salishan Lodge symposium, 32 schemas (Piaget), 20 schizophrenic patients, prism distortion, 202 selection process, basal ganglia, 210-211 self-monitoring, motor actions, 189-190 sensory cancellation, 178-180, 189-190 sensory signals, 172-174 serial reaction time, voluntary motor control, 151 serotonin, Lugaro cell sensitivity to, 56 serum response factor (SRF), 89 signal transduction pathways (conjunctive LTD), 71 AMPA receptors, 74-75 brain-derived neurotrophic factor (BDNF), 80 c-Fos/Jun-B, 79

282

index

Ca2+ surge in Purkinje cell dendrites, 71-74 Ca2+/calmodulin-dependent protein kinase (CaMKII), 78 glial fibrillary acidic protein (GFAP), 79 glutamate receptor d2 (GluR d2), 76-77 insulin-like growth factor-1 (IGF-1), 80 Nitric oxide (NO) synthase, 75-76 PKCα and lipid-signaling cascade, 74 protein phosphatases, 78 protein tyrosine kinases, 77 silent synapses, 86-87 simple perceptron model network, 8 simple reaction time, voluntary motor control, 151 Smith Predictor, 169 smooth pursuit, frontal eye field, 160-162 SNARE proteins, 82 social interaction, parallels to motor control, 196 somatic reflexes adaptive control system models, 147-148 eye-blink conditioning, 132-136 locomotion, 128-132 nociceptive withdrawal reflex, 127-128 stretch reflex, 121-127 somatosympathetic reflex, 137

sources, voluntary motor control instruction signals, 157-158 spike discharges, locomotion, 130 spinal neural control mechanisms, muscle contractions, 123 spinal reflex circuitry, 2-4 spindle II muscle afferent, stretch reflex, 121-123 spine configuration, synaptic plasticity, 86 spinocerebellum, 23 split-belt treadmill locomotion, 132 SRF (serum response factor), 89 stabilization-by-selection mechanism, basal ganglia, 210-211 stellate cells, 51-53, 83 storage capacity, Purkinje cells, 92 straight cells (inferior olive complex), 64 stretch reflex, 12, 121-127 structure-function relationships, neuronal circuits, 1-21 subtrigeminal nuclei (lateral reticular nucleus), 62 subtypes, synaptic plasticity, 81 basket/stellate cells, 83 cerebellar cortical synapses, 84

index

cerebellar/vestibular nuclear neurons, 84-85 persistency, 86-87 Purkinje cells, 81-83 supplementary eye field, 165 sympathetic reflexes, 136-138 synapses, 7-8 synaptic plasticity, 81 basket/stellate cells, 83 cerebellar cortical synapses, 84 cerebellar/vestibular nuclear neurons, 84-85 conjunctive LTD (long-term depression), 205-206 properties, 69-71 signal transduction pathways, 71-80 as memory element in neuronal circuits, 33-35 persistency, 86-87 protein synthesis, 88-89 Purkinje cells, 81-83 synthase, conjunctive LTD, 75-76 systems control mechanisms (CNS), 9 cognitive functions, 19-20 feedback control systems, 10 intelligence and conscious awareness, 20-21 model-based control systems, 10-11 reflexes, 12-18 voluntary movements, 12-13, 18-19 Szentágothai, Janos, 28

283

T task dependencies, functional stretch reflex, 126 temporoparietal cortex, 193 tendon tap, 122 thought processes activity in the cerebellum, 198-201 cerebellar internal model for thoughts, 196 cerebral cortical model for thoughts, 195-196 explicit/implicit thoughts, 197-198 mental disorders associated with cerebellar dysfunction, 201-203 neural systems, 193-195 throwing a ball, voluntary motor control, 153-154 tools, motor actions, 190-192 traditional views of cerebellum involvement in motor skills, 25-27 morphological map, 22-25 Purkinje cells, 27-28 transcriptional inhibitors, protein synthesis, 88 transduction pathways (conjunctive LTD), 71 AMPA receptors, 74-75 brain-derived neurotrophic factor (BDNF), 80 c-Fos/Jun-B, 79 Ca2+ surge in Purkinje cell dendrites, 71-74

284

index

Ca2+/calmodulin-dependent protein kinase (CaMKII), 78 glial fibrillary acidic protein (GFAP), 79 glutamate receptor d2 (GluR d2), 76-77 insulin-like growth factor-1 (IGF-1), 80 Nitric oxide (NO) synthase, 75-76 PKCα and lipid-signaling cascade, 74 protein phosphatases, 78 protein tyrosine kinases, 77 translational inhibitors, protein synthesis, 88 Tsukahara, Nakakira, 30

U–V unipolar brush cells, 46-47 V-type microcomplex, adaptive control prototype, 146-149 velocity storage, VOR (Vestibuloocular Reflex), 107 ventral spinocerebellar tract (VSCT), 60 vergence, frontal eye field, 164-165 vermal lobules VI/VII, eye movements, 159 vermis, 23 vestibular compensation, VOR (Vestibuloocular Reflex), 112-113

vestibular mossy fiber input, VOR adaptation, 110 vestibular nuclear neurons, 68 vestibular organs, 23 vestibulocerebellum, 23 Vestibuloocular Reflex (VOR), 36, 105-106 adaptation control system climbing fiber input, 110-111 eye movement-related signals, 111-112 flocculus, 107-109 memory sites, 112 vestibular mossy fiber input, 110 adaptive control, 139-140 feedforward control system, 39 neural integrator, 107 velocity storage, 107 vestibular compensation, 112-113 vestibulosympathetic reflex, 136-137 voltage-gated Ca2+ channels, 71 voluntary eye movement, 159 DMFC (dorsomedial frontal cortex), 165-166 frontal eye field, 159-165 voluntary motor control, 150 hand grip, 154-156 instruction signals, 157-158

index

internal models, 167 internal forward model, 167-175 internal inverse model, 170-175 MOSAIC (modular selection and identification for control), 175 sensory cancellation, 178-180 transition from reflex control to voluntary, 176-179 load compensation, 150 multijoint arm movements, 151-154 operant conditioning, 156-157 reaction time, 151 voluntary movements, 12 hybrid control, 13-14 motor actions, 18-19 VOR (Vestibuloocular Reflex), 36, 105-106 adaptation control system climbing fiber input, 110-111 eye movement-related signals, 111-112 flocculus, 107-109 memory sites, 112 vestibular mossy fiber input, 110

285

adaptive control, 139-140 feedforward control system, 39 neural integrator, 107 velocity storage, 107 vestibular compensation, 112-113 VSCT (ventral spinocerebellar tract), 60

W–Z whale cerebellum, 24-25 wiring diagrams, 31 Wisconsin card-sorting test, 199 XXVII Congress of the International Physiological Union, 36 Yoshida, Mitsuo, 30 zebrin, 24

press FINANCIAL TIMES

In an increasingly competitive world, it is quality of thinking that gives an edge—an idea that opens new doors, a technique that solves a problem, or an insight that simply helps make sense of it all. We work with leading authors in the various arenas of business and finance to bring cutting-edge thinking and best-learning practices to a global market. It is our goal to create world-class print publications and electronic products that give readers knowledge and understanding that can then be applied, whether studying or at work. To find out more about our business products, you can visit us at www.ftpress.com.

Color Plate I MRI image of the human cerebellum. Recent remarkable progress in brain imaging has made it possible to observe the human cerebellum in vivo. This horizontal section image was taken by the author on June 7, 2010. Abbreviations: b, basilar artery; h, cerebellar hemisphere; p, pons; v, 4th ventricle. For a three-dimensional atlas of the human cerebellum, see Schmahmann et al., 1999.

Color Plate II A schematized map of the divisions of the cerebellar cortex. On the outline of the anatomical divisions of the cerebellum (vermis, paravermis, and hemisphere) a lattice is formed by crossing transverse lobules and longitudinal zones to set coordinates for defining the position of each cortical area. Note that subzones A2, X, and Y are not shown. Note also that the zones and lobules are shown with the same width for simplicity despite that the fact that their widths vary considerably. The nomenclature used is that for the generalized mammalian cerebellum. Abbreviations: I-X, lobular divisions of the cerebellar vermis and paravermis; A-D2, longitudinal zones of the cerebellum; HI-HX, lobuli extending laterally from I-X. APF, ansoparamedian fissure; PFR, primary fissure; PLF, posterolateral fissure.

Color Plate III Longitudinal zonal organization in the Purkinje cell output and climbing fiber input. Upper half shows longitudinal zones and flocculonodular lobe after Color Plate II. Lower half, Target nuclei for Purkinje cell zones in the left half of the diagram. The right side shows the input connection from the inferior olive. Abbreviations: A-D2, zonal bands; AIP, anterior interpositus nucleus; c, caudal (MAO or DAO); DC, dorsal cap; Dc, caudal subnucleus of the dentate nucleus; Dr, rostral subnucleus of the dentate nucleus; dl, dorsal lamella; FAS, fastigial nucleus; FNL, flocculonodular lobe; LVN, lateral vestibular nucleus; MAO, medial accessory olive; PIP, posterior interpositus nucleus; PO, principal olive; r, rostral (MAO or DAO); VN, vestibular nuclei; vl, ventral lamella. (For further details, see Voogd, 2010.)

Color Plate IV A classic Ramón y Cajal drawing of cerebellar cells. Shown are Purkinje (A, B), stellate (D), Golgi (F), and granule (H) cells in the cerebellum. Also note the basket-cell axons (S) terminating freely around the Purkinje cell bodies. (Courtesy of the Instituto de Neurobiología “Ramón y Cajal,” Madrid, Spain.)

Color Plate V Sketch of neuronal circuits in the cerebellum. Long-term depression of Purkinje cells is induced by the conjunctive activation of parallel fibers and climbing fibers. (From Ito, 2009.)

Color Plate VI Unipolar brush cells in the adult mouse cerebellum. Shown are orange-stained cell bodies of unipolar brush cells, using the transcription factor, Tbr2/E, as the marker. Also shown is one unipolar brush cell of the calretinin subtype, which is green-stained in the outer edges of the cell body (at arrow) and in the dendritic brush (arrowhead) with an anticalretinin antibody. The yellow staining in the center of the cell body is due to staining with Tbr2+. Calibration bar, 13 micrometers. (From Englund et al., 2006.)

Color Plate VII Purkinje cell dendrites. Shown are the dendrites of two rat Purkinje cells injected with Alexa-594 from another rat. (A) A Purkinje cell on Z-stack image obtained from optical sections using a confocal laser scanning microscope. Scale, 20 micrometers. (B) Purkinje cell dendrites reconstructed three-dimensionally. Scale, 2 micrometers. (Courtesy of Tetsuya Tatsukawa.)

Color Plate VIII Dendritic spines of Purkinje cells. (A) A dendrite segment from an adult rat Purkinje cell, reconstructed from consecutive 90 nm serial sections. In this 3D rendition, the dendritic shaft (orange) at a branch point is studded with synaptic spines (in semitransparent green). Only a subset of the parallel fibers (gray) crossing the dendritic plane is shown here, with one displaying the typical “en passant” synaptic contact, with its postsynaptic density (PSD) shown in red (lower-left quadrant). In this reconstruction, only two climbing fiber synapses were found, with one visible here, as identified by its PSD (also red) in apposition with the Purkinje cell dendritic shaft. (B) A single spine reconstructed by electron tomography. The complex branched structure of the spine’s endoplasmic reticulum (purple) is visible through the semitransparent representation of the plasma membrane. In the dendritic shaft, note the continuity with endoplasmic reticulum. In A, average spine length is 1.36 micrometers. The spine in B is 1.4 micrometers long—i.e., from the base of its neck to its apex. (Courtesy of Thomas Launey.)

Color Plate IX Climbing fibers. After the injection of the tracer, BDA (biotinylated dextran amine), into the inferior olive of a mouse, labeled climbing fibers are visualized with immunofluorescent (red) methods in the cerebellar cortex. (A) Purkinje cells were also visualized by immunohistochemistry for a calcium binding protein, Calbindin (green). (B) Without labeling Calbindin. Scale bars: 100 micrometers. (Courtesy of Tsutomu Hashikawa.)

Color Plate X A Golgi cell and a Lugaro cell. (A) This Golgi cell in a cerebellar slice was filled with AlexaFluo through a patchclamp pipette (removed). It was reconstructed into a stack view by use of a confocal microscope. Note the cell’s broad extension of its axonal plexus, multiple short basal dendrites, and two apical dendrites, which climb into the molecular layer (not shown). (B) * marks a Lugaro cell observed under a microscope in a cerebellar slice derived from a gene-manipulated mouse that expressed GFP specifically in GABAergic neurons. Note also the green-labeled Purkinje and stellate cells, including their dendrites, in the molecular and Purkinje cell layers. Scale for B, 50 micrometers. (A, from D’Angelo, 2008; B, courtesy of Moritoshi Hirono.)

Color Plate XI Gap junction couplings among inferior olive neurons. (A) Dye injection into a “straight” (term based on shape of dendritic morphology) neuron resulted in indirect labeling of nine additional neurons, two of which (also straight) are shown in this panel. The darkly stained dendrite belongs to the cell that was labeled directly. Arrows denote examples of intersections of the dendrites of different cells, these being the possible locations of gap junctions. (B) Dye injected into a curly neuron (note its large dark cell body) resulted in indirect labeling of curly neurons. Indirectly labeled neurons had clearly stained cell bodies (e.g., the one at the arrow), but very weakly labeled dendrites. Scale bars, 20 micrometers. (From Devor and Yarom, 2002.)

Color Plate XII A Ca2+ surge induced by conjunctive stimulation in a Purkinje cell. (A) This cell was visualized using infrared differential interference contrast optics. A patch pipette containing a Ca2+ indicator was attached to the cell’s soma. Parallel fibers (PF) were stimulated using an extracellular electrode placed in the molecular layer. The dendritic region enclosed in the red-line rectangle was receiving synapses from the stimulated PFs as indicated by an increase in the fluorescent signals. Conjunctive stimulation was performed with a combination of 2 PF-stimulation at 30 ms intervals and somatic depolarization from– 70 to –20 mV for 150 ms, repeated at 1 Hz for 5 minutes. (B) Ca2+ images in the red-line rectangle in A shown before (a) and then 0.5, 2.5, and 4.5 (b, c, and d) minutes after the onset of conjunctive stimulation. (C) Changes in fluorescence intensity as a function of time, these being normalized relative to the average intensity recorded before conjunctive stimulation. Curves recorded for 10 seconds at times a–d in B are shown in different colors (a in black, b in red, c in green, and d in brown). Note that the curves in c and d nearly overlap each other. (Supporting data for Le et al., 2010.)

Color Plate XIII Declustering of GluR2/3 brought on by coapplication of AMPA and PP-2A. Shown are fluoromicrographs of two Purkinje cells’ dendrites immunolabeled with an antibody binding the extracellular domain of GluR2/3. (Upper panel) The control situation showing normal levels of GluR2/3 expression in cultured Purkinje cell dendrites. Note the small bright puncta (yellow to red), which indicate GluR2/3 clustering at granule cell-Purkinje cell synapses. (Lower panel) Equivalent dendrites studied in a sister culture. Treatment with cytostatin plus AMPA caused a reduction in the density, area, and fluorescence intensity of the GluR2/3 clusters. For both panels, the scale bar is 10 micrometers. (From Launey et al., 2004.)

Color Plate XIV GluR2 internalization of granule cell-Purkinje cell synapses. The shift of GluR2 from the dendritic membrane to its inside is visualized using an antibody against the N terminus of GluR, surface GluR2 (red), and internalized GluR2 (blue) under nonpermeant and permeant conditions, respectively. The LY panels show control (Ctrl.) and treated (TeTx) fluorescent images (in green) of Lucifer yellow infused into Purkinje cells without and with TeTx, respectively. Scale bar, 5 micrometers. (From Tatsukawa et al., 2006.)

Color Plate XV Disturbances of an arm-reaching movement by transcranial magnetic stimulation of the cerebellum. (A) The experimental task, with further details provided in the text. TMS was applied over the right lateral cerebellum. TMS trials are plotted in red lines, and control nonTMS trials are in blue lines. (B and C) Individual trial data of the finger trajectory on the frontal plane (Y–Z) viewed from behind and on the sagittal plane (X–Z) to the right of the subject, respectively. (D and E) Session averaged data (n = 30 trials) from one typical participant. Note deviations of the trajectory under TMS from that without TMS. (From Miall et al., 2007.)

Color Plate XVI Cerebral cortical areas for the mirror system. This lateral view of the monkey brain shows the motor areas of the frontal lobe and the main areas of the posterior parietal cortex. See Rizzolatti and Craighero (2004) for the nomenclature and definition of frontal motor areas (F1–F7) and posterior parietal areas (PE, PEc, PF, PFG, PFop, PG, PGop, and Opt),). Additional abbreviations: AI, inferior arcuate sulcus; AS, superior arcuate sulcus; C, central sulcus; FEF, frontal eye field; IO, inferior occipital sulcus; L, lateral fissure; Lu, lunate sulcus; P, principal sulcus; STS, superior temporal sulcus. (From Rizzolatti and Craighero, 2004.)

Join the

FT Press AffiliAte teAm

You love our books and you love to share them with your colleagues and friends...why not earn some money doing it?

If you have a website, blog or even a Facebook page, you can start earning money by putting an FT Press link on your page. If a visitor clicks on that link and purchases something from ftpress.com, you earn commissions* on all sales! Every sale you bring to our site will earn you a commission. All you have to do is post an ad and we’ll take care of the rest.

ApplY And get stArted It’s quick and easy to apply. To learn more go to: http://www.ftpress.com/affiliates/ *Valid for all books, eBooks and video sales at www.ftpress.com