motor cortex in voluntary movements a distributed system for distributed functions

MOTOR CORTEX IN VOLUNTARY MOVEMENTS A DISTRIBUTED SYSTEM FOR DISTRIBUTED FUNCTIONS EDITED BY Alexa Riehle and Eilon Vaadia CRC PR E S S Boca Raton London New York Washington, D.C Library of Congress Cataloging-in-Publication Data Motor cortex in voluntary movements : a distributed system for distributed functions / edited by Alexa Riehle and Eilon Vaadia p cm Includes bibliographical references and index ISBN 0-8493-1287-6 (alk paper) Motor cortex Human locomotion I Riehle, Alexa II Vaadia, Eilon III Series QP383.15.M68 2005 612.8′252—dc22 2004057046 This book contains information obtained from authentic and highly regarded sources Reprinted material is quoted with permission, and sources are indicated A wide variety of references are listed Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher All rights reserved Authorization to photocopy items for internal or personal use, or the personal or internal use of specific clients, may be granted by CRC Press, provided that $1.50 per page photocopied is paid directly to Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923 USA The fee code for users of the Transactional Reporting Service is ISBN 0-8493-1287-6/05/$0.00+$1.50 The fee is subject to change without notice For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged The consent of CRC Press does not extend to copying for general distribution, for promotion, for creating new works, or for resale Specific permission must be obtained in writing from CRC Press for such copying Direct all inquiries to CRC Press, 2000 N.W Corporate Blvd., Boca Raton, Florida 33431 Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe Visit the CRC Press Web site at www.crcpress.com © 2005 by CRC Press No claim to original U.S Government works International Standard Book Number 0-8493-1287-6 Library of Congress Card Number 2004057046 Printed in the United States of America Printed on acid-free paper Copyright © 2005 CRC Press LLC Methods & New Frontiers in Neuroscience Our goal in creating the Methods & New Frontiers in Neuroscience series is to present the insights of experts on emerging experimental techniques and theoretical concepts that are or will be at the vanguard of the study of neuroscience Books in the series cover topics ranging from methods to investigate apoptosis to modern techniques for neural ensemble recordings in behaving animals The series also covers new and exciting multidisciplinary areas of brain research, such as computational neuroscience and neuroengineering, and describes breakthroughs in classical fields such as behavioral neuroscience We want these to be the books every neuroscientist will use in order to graduate students and postdoctoral fellows when they are looking for guidance to start a new line of research Each book is edited by an expert and consists of chapters written by the leaders in a particular field Books are richly illustrated and contain comprehensive bibliographies Chapters provide substantial background material relevant to the particular subject; hence, they are not only “methods” books They contain detailed tricks of the trade and information as to where these methods can be safely applied In addition, they include information about where to buy equipment and about Web sites that are helpful in solving both practical and theoretical problems We hope that as the volumes become available, the effort put in by us, by the publisher, by the book editors, and by the individual authors will contribute to the further development of brain research The extent to which we achieve this goal will be determiend by the utility of these books Sidney A Simon, Ph.D Miguel A.L Nicolelis, M.D., Ph.D Series Editors Copyright © 2005 CRC Press LLC Preface Voluntary movement is undoubtedly the overt basis of human behavior Without movement we cannot walk, nourish ourselves, communicate, or interact with the environment This is one of the reasons why the motor cortex was one of the first cortical areas to be explored experimentally Historically, the generation of motor commands was thought to proceed in a rigidly serial and hierarchical fashion The traditional metaphor of the piano presents the premotor cortex “playing” the upper motoneuron keys of the primary motor cortex (M1), which in turn activate with strict point-to-point connectivity the lower motoneurons of the spinal cord Years of research have taught us that we may need to reexamine almost all aspects of this model Both the premotor and the primary motor cortex project directly to the spinal cord in highly complex overlapping patterns, contradicting the simple hierarchical view of motor control The task of generating and controlling movements appears to be subdivided into a number of subtasks that are accomplished through parallel distributed processing in multiple motor areas Multiple motor areas may increase the behavioral flexibility by responding in a context-related way to any constraint within the environment Furthermore, although more and more knowledge is accumulating, there is still an ongoing debate about what is represented in the motor cortex: dynamic parameters (such as specific muscle activation), kinematic parameters of the movement (for example, its direction and speed), or even more abstract parameters such as the context of the movement Given the great scope of the subject considered here, this book focuses on some new perspectives developed from contemporary monkey and human studies Moreover, many topics receive very limited treatment Section I, which includes the first two chapters, uses functional neuroanatomy and imaging studies to describe motor cortical function The objective of Chapter is to describe the major components of the structural framework employed by the cerebral cortex to generate and control skeletomotor function Dum and Strick focus on motor areas in the frontal lobe that are the source of corticospinal projections to the ventral horn of the spinal cord in primates These cortical areas include the primary motor cortex (M1) and the six premotor areas that project directly to it The results presented lead to an emerging view that motor commands can arise from multiple motor areas and that each of these motor areas makes a specialized contribution to the planning, execution, or control of voluntary movement The purpose of Chapter is to provide an overview of the contribution of functional magnetic resonance imaging (fMRI) to some of the prevailing topics in the study of motor control and the function of the primary motor cortex Kleinschmidt and Toni claim that in several points the findings of functional neuroimaging seem to be in apparent disagreement with those obtained with other methods, which cannot always be attributed to insufficient sensitivity of this noninvasive technique In part, it may Copyright © 2005 CRC Press LLC reflect the indirect and spatio-temporally imprecise nature of the fMRI signal, but these studies remain informative by virtue of the fact that usually the whole brain is covered Not only does fMRI reveal plausible brain regions for the control of localized effects, but the distribution of response foci and the correlation of effects observed at many different sites can assist in the guidance of detailed studies at the mesoscopic or microscopic spatio-temporal level A prudently modest view might conclude that fMRI is at present primarily a tool of exploratory rather than explanatory value Section II provides a large overview of studies about neural representations in the motor cortex Chapter focuses on the neuromuscular evolution of individuated finger movements Schieber, Reilly, and Lang demonstrate that rather than acting as a somatotopic array of upper motor neurons, each controlling a single muscle that moves a single finger, neurons in the primary motor cortex (M1) act as a spatially distributed network of very diverse elements, many of which have outputs that diverge to facilitate multiple muscles acting on different fingers This biological control of a complex peripheral apparatus initially may appear unnecessarily complicated compared to the independent control of digits in a robotic hand, but can be understood as the result of concurrent evolution of the peripheral neuromuscular apparatus and its descending control from the motor cortex Chapter deals with simultaneous movements of the two arms, as a simple example of complex movements, and may serve to test whether and how the brain generates unique representations of complex movements from their constituent elements Vaadia and Cardoso de Oliveira present evidence that bimanual representations indeed exist, both at the level of single neurons and at the level of neuronal populations (in local field potentials) They further show that population firing rates and dynamic interactions between the hemispheres contain information about the bimanual movement to be executed In Chapter 5, Ashe discusses studies with respect to the debate as to whether the motor cortex codes the spatial aspects (kinematics) of motor output, such as direction, velocity, and position, or primarily controls, muscles, and forces (dynamics) Although the weight of evidence is in favor of M1 controlling spatial output, the effect of limb biomechanics and forces on motor cortex activity is beyond dispute The author proposes that the motor cortex indeed codes for the most behaviorally relevant spatial variables and that both spatial variables and limb biomechanics are reflected in motor cortex activity Chapter starts with the important issue of how theoretical concepts guide experimental design and data analysis Scott describes two conceptual frameworks for interpreting neural activity during reaching: sensorimotor transformations and internal models He claims that sensorimotor transformation have been used extensively over the past 20 years to guide neurophysiological experiments on reaching, whereas internal models have only recently had an impact on experimental design Furthermore, the chapter demonstrates how the notion of internal models can be used to explore the neural basis of movement by describing a new experimental tool that can sense and perturb multiple-joint planar movements Chapter deals with the function of oscillatory potentials in the motor cortex MacKay notes that from their earliest recognition, oscillatory EEG signals in the sensorimotor cortex have been associated with stasis: a lack of movement, static postures, and possibly physiological tremor It is now established that Copyright © 2005 CRC Press LLC 10-, 20-, and 40-Hz motor cortical oscillations are associated with constant, sustained muscle contractions, again a static condition Sigma band oscillations of about 14 Hz may be indicative of maintained active suppression of a motor response The dynamic phase at the onset of an intended movement is preceded by a marked decrease in oscillatory power, but not all frequencies are suppressed Fast gamma oscillations coincide with movement onset Moreover, there is increasing evidence that oscillatory potentials of even low frequencies (4–12 Hz) may be linked to dynamic episodes of movement Most surprisingly, the 8-Hz cortical oscillation — the neurogenic component of physiological tremor — is emerging as a major factor in shaping the pulsatile dynamic microstructure of movement, and possibly in coordinating diverse actions performed together In Chapter 8, Riehle discusses the main aspects of preparatory processes in the motor cortex Preparation for action is thought to be based on central processes, which are responsible for maximizing the efficiency of motor performance A strong argument in favor of such an efficiency hypothesis of preparatory processes is the fact that providing prior information about movement parameters or removing time uncertainty about when to move significantly shortens reaction time The types of changes in the neuronal activity of the motor cortex, and their selectivity during preparation, are portrayed and compared with other cortical areas that are involved in motor behavior Furthermore, linking motor cortical activity directly to behavioral performance showed that the trial-by-trial correlation between single neuron firing rates and reaction time revealed strong task-related cortical dynamics Finally, the cooperative interplay among neurons, expressed by precise synchronization of their action potentials, is illustrated and compared with changes in the firing rate of the same neurons New concepts including the notion of coordinated ensemble activity and their functional implication during movement preparation are discussed In the last chapter of Section II, Chapter 9, Jeannerod poses the question of the role of the motor cortex in motor cognition The classical view of the primary motor cortex holds that it is an area devoted to transferring motor execution messages that have been elaborated upstream in the cerebral cortex More recently, however, experimental data have pointed to the fact that the relation of motor cortex activity to the production of movements is not as simple as was thought on the basis of early stimulation experiments This revision of motor cortical function originated from two main lines of research, dealing first with the plasticity of the somatotopic organization of the primary motor cortex, and second with its involvement in cognitive functions such as motor imagery Section III is mainly concerned with motor learning Chapter 10 explores various conditions of mapping between sensory input and motor output Brasted and Wise claim that studies on the role of the motor cortex in voluntary movement usually focus on standard sensorimotor mapping, in which movements are directed toward sensory cues Sensorimotor behavior can, however, show much greater flexibility Some variants rely on an algorithmic transform between the location of the cue and that of the target The well-known “antisaccade” task and its analogues in reaching serve as special cases of such transformational mapping, one form of nonstandard mapping Other forms of nonstandard mapping differ strongly: they are arbitrary In arbitrary sensorimotor mapping, the cue’s location has no systematic spatial relationship with the response The authors explore several types of arbitrary mapping, Copyright © 2005 CRC Press LLC with emphasis on the neural basis of learning In Chapter 11, Shadmehr, Donchin, Hwang, Hemminger, and Rao deal with internal models that transform the desired movement into a motor command When one moves the hand from one point to another, the brain guides the arm by relying on neural structures that estimate the physical dynamics of the task Internal models are learned with practice and are a fundamental part of voluntary motor control What internal models compute, and which neural structures perform that computation? The authors approach these questions by considering a task where the physical dynamics of reaching movements are altered by force fields that act on the hand Many studies suggest that internal models are sensorimotor transformations that map a desired sensory state of the arm into an estimate of forces; i.e., a model of the inverse dynamics of the task If this computation is represented as a population code via a flexible combination of basis functions, then one can infer activity fields of the bases from the patterns of generalization Shadmehr and colleagues provide a mathematical technique that facilitates this inference by analyzing trial-by-trial changes in performance Results suggest that internal models are computed with bases that are directionally tuned to limb motion in intrinsic coordinates of joints and muscles, and this tuning is modulated multiplicatively as a function of static position of the limb That is, limb position acts as a gain field on directional tuning Some of these properties are consistent with activity fields of neurons in the motor cortex and the cerebellum The authors suggest that activity fields of these cells are reflected in human behavior in the way that we learn and generalize patterns of dynamics in reaching movements In the last chapter of Section III, Chapter 12, Padoa-Schioppa, Bizzi, and Mussa-Ivaldi address the question of the cortical control of motor learning In robotic systems, engineers coordinate the action of multiple motors by writing computer codes that specify how the motors must be activated for achieving the desired robot motion and for compensating unexpected disturbance Humans and animals follow another path Something akin to programming is achieved in nature by the biological mechanisms of synaptic plasticity — that is, by the variation in efficacy of neural transmission brought about by past history of pre- and post-synaptic signals However, robots and animals differ in another important way Robots have a fixed mechanical structure and dimensions In contrast, the mechanics of muscles, bones, and ligaments change in time Because of these changes, the central nervous system must continuously adapt motor commands to the mechanics of the body Adaptation is a form of motor learning Here, a view of motor learning is presented that starts from the analysis of the computational problems associated with the execution of the simplest gestures The authors discuss the theoretical idea of internal models and present some evidence and theoretical considerations suggesting that internal models of limb dynamics may be obtained by the combination of simple modules or “motor primitives.” Their findings suggest that the motor cortical areas include neurons that process well-acquired movements as well as neurons that change their behavior during and after being exposed to a new task The last section, Section IV, is devoted to the reconstruction of movements using brain activity For decades, science fiction authors anticipated the view that computers can be made to communicate directly with the brain Now, a rapidly expanding science community is making this a reality In Chapter 13, Carmena and Nicolelis Copyright © 2005 CRC Press LLC present and discuss the recent research in the field of brain–machine interfaces (BMI) conducted mainly on nonhuman primates In fact, this research field has supported the contention that we are at the brink of a technological revolution, where artificial devices may be “integrated” in the multiple sensory, motor, and cognitive representations that exist in the primate brain These studies have demonstrated that animals can learn to utilize their brain activity to control the displacements of computer cursors, the movements of simple and elaborate robot arms, and, more recently, the reaching and grasping movements of a robot arm In addition to the current research performed in rodents and primates, there are also preliminary studies using human subjects The ultimate goal of this emerging field of BMI is to allow human subjects to interact effortlessly with a variety of actuators and sensory devices through the expression of their voluntary brain activity, either for augmenting or restoring sensory, motor, and cognitive function In the last chapter, Chapter 14, Pfurtscheller, Neuper, and Birbaumer deal with BMIs, which transform signals originating from the human brain into commands that can control devices or applications BCIs provide a new nonmuscular communication channel, which can be used to assist patients who have highly compromised motor functions, as is the case with patients suffering from neurological diseases such as amyotrophic lateral sclerosis (ALS) or brainstem stroke The immediate goal of current research in this field is to provide these users with an opportunity to communicate with their environment Presentday BCI systems use different electrophysiological signals such as slow cortical potentials, evoked potentials, and oscillatory activity recorded from scalp or subdural electrodes, and cortical neuronal activity recorded from implanted electrodes Due to advances in methods of signal processing, it is possible that specific features automatically extracted from the electroencephalogram (EEG) and electrocorticogram (ECoG) can be used to operate computer-controlled devices The interaction between the BCI system and the user, in terms of adaptation and learning, is a challenging aspect of any BCI development and application It is the increased understanding of neuronal mechanisms of motor functions, as reflected in this book, that led to the success of BCI Yet, the success in tapping and interpreting neuronal activity and interfacing it with a machine that eventually executes the subject’s intention is amazing, considering the limited understanding we have of the system as a whole Perhaps ironically, the proof of our understanding of motor cortical activity will stem from how effectively we, as external observers of the brain, can tap into it and make use of it Alexa Riehle Eilon Vaadia Copyright © 2005 CRC Press LLC Dedication to Hanns-Günther Riehle Copyright © 2005 CRC Press LLC Editors Alexa Riehle received a B.Sc degree in biology (main topic: deciphering microcircuitries in the frog retina) from the Free University, Berlin, Germany, in 1976, and a Ph.D degree in neurophysiology (main topic: neuronal mechanisms of temporal aspects of color vision in the honey bee) from the Biology Department of the Free University in 1980 From 1980 to 1984, she was a postdoctoral fellow at the National Center for Scientific Research (CNRS) in Marseille, France (main topic: neuronal mechanisms of elementary motion detectors in the fly visual system) In 1984, she moved to the Cognitive Neuroscience Department at the CNRS and has been mainly interested since then in the study of cortical information processing and neural coding in cortical ensembles during movement preparation and execution in nonhuman primates Eilon Vaadia graduated from the Hebrew University of Jerusalem (HUJI) in 1980 and joined the Department of Physiology at Hadassah Medical School after postdoctoral studies in the Department of Biomedical Engineering at Johns Hopkins University Medical School in Baltimore, Maryland Vaadia studies cortical mechanisms of sensorimotor functions by combining experimental work (recordings of multiple unit activity in the cortex of behaving animals) with a computational approach He is currently the director of the Department of Physiology and the head of the Ph.D program at the Interdisciplinary Center for Neural Computation (ICNC) at HUJI, and a director of a European advanced course in computational neuroscience Copyright © 2005 CRC Press LLC 40 Annett, J., On knowing how to things: a theory of motor imagery, Brain Res Cogn Brain Res., 3, 65, 1996 41 Grafton, S.T et al., Premotor cortex activation during observation and naming of familiar tools, NeuroImage, 6, 231, 1997 42 Buccino, G et al., Action observation activates premotor and parietal areas in a somatotopic manner: an fMRI study, Eur J Neurosci., 13, 400, 2001 43 Klass, S.G and Bickford, R.G., Observations on the rolandic arceau rhythm, Electroenceph Clin Neurophysiol., 9, 570, 1957 44 Chatrian, G.E., Petersen, M.C., and Lazarte, J.A., The blocking of the rolandic wicket rhythm and some central changes related to movement, Electroenceph Clin Neurophysiol., 11, 497, 1959 45 Beisteiner, R et al., Mental representations of movements Brain potentials associated with imagination of hand movements, Electroenceph Clin Neurophysiol., 96, 183, 1995 46 Cunningham, R., Iansek, R., and Stokes, M.J., Movement-related potentials associated with movement preparation and motor imagery, Exp Brain Res., 111, 429, 1996 47 Schnitzler, A et al., Involvement of primary motor cortex in motor imagery: a neuromagnetic study, NeuroImage, 6, 201, 1997 48 Cassim, F et al., Does post-movement beta synchronization reflect an idling motor cortex? Neuroreport, 12, 3859, 2001 49 Hinterberger, T et al., A brain–computer interface (BCI) for the locked-in: comparison of different EEG classifications for the thought translation device (TTD), Clin Neurophysiol., 114, 3, 416, 2003 50 Stamm, J.S., Whipple, S., and Born, J., Effects of spontaneous cortical slow potentials on semantic information processing, Int J Psychophysiol., 5, 11, 1987 51 Stamm, J.S and Rosen, S.C., The locus and crucial time of implication of prefrontal cortex in the delayed response task, in Psychophysiology of the Frontal Lobes, Pribram, K and Luria, A., Eds., Academic Press, New York, 1973, 139 52 Fetz, E., Operant conditioning of cortical unit activity, Science, 163, 955, 1969 53 Hinterberger, T et al., Brain areas activated in fMRI during self-regulation of slow cortical potentials (SCP), Exp Brain Res., 152, 113, 2003 54 Birbaumer, N et al., Slow brain potentials, imagery and hemispheric differences, J Neurosci., 39, 101, 1988 55 Kotchoubey, B et al., Is there a mind? Electrophysiology of unconscious patients, News Physiol Sci., 17, 38, 2002 56 Wolpaw, J.R., McFarland, D.J., and Vaughan, T.M., Brain–computer interface research at the Wadsworth Center, IEEE Trans Rehab Eng., 8, 222, 2002 57 Sutter, E.E., The visual evoked response as a communicatioan channel, Proc IEEE/NSF Symp Biosensors, 95, 1984 58 Donchin, E., Spencer, K.V and Wijesinghe, R., The mental prosthesis: assessing the speed of a P300-based brain–computer interface, IEEE Trans Rehab Eng., 8, 174, 2000 59 Middendorf, M et al., Brain–computer interfaces based on the steady-state visualevoked response, IEEE Trans Rehab Eng., 2, 211, 2000 60 Wolpaw, J.R et al., An EEG-based brain–computer interface for cursor control, Electroenceph Clin Neurophysiol., 78, 252, 1991 61 McFarland, D.J., Lefkowicz, A.T., and Wolpaw, J.R., Design and operation of an EEG-based brain–computer interface (BCI) with digital signal processing technology, Behav Res Methods Instrum Comput., 29, 337, 1997 62 Miner, L.A., McFarland, D.J., and Wolpaw, J.R., Answering questions with an EEGbased brain–computer interface (BCI), Arch Phys Med Rehab., 79, 1029, 1998 Copyright © 2005 CRC Press LLC 63 Wolpaw, J.R et al., EEG-based communication: improved accuracy by response verification, IEEE Trans Rehab Eng., 6, 326, 1998 64 McFarland, D.J et al., EEG-based brain–computer interface (BCI) communication: effects of target number and trial length on information transfer rate, Soc Neurosci Abst., 26, 1228, 2000 65 McFarland, D.J et al., Spatial filter selection for EEG-based communication, Electroenceph Clin Neurophysiol., 103, 386, 1997 66 Ramoser, H., Wolpaw, J.R., and Pfurtscheller, G., EEG-based communication: evaluation of alternative signal prediction methods, Biomed Technik, 42, 226, 1997 67 Schalk, G et al., EEG-based communication and control: presence of an error potential, Clin Neurophysiol., 111, 2138, 2000 68 Vaughan, T.M et al., EEG-based brain–computer interface: development of a speller, Soc Neurosci Abstr., 27, 167, 2001 69 Kalcher, J et al., Graz Brain–Computer Interface II: towards communication between humans and computers based on online classification of three different EEG patterns, Med Biol Eng Comput., 34, 382, 1996 70 Pregenzer, M and Pfurtscheller, G., Frequency component selection for an EEGbased Brain–Computer Interface (BCI), IEEE Trans Rehab Eng., 7, 413, 1999 71 Pfurtscheller, G et al., EEG-based discrimination between imagination of right and left hand movement, Electroenceph Clin Neurophysiol., 103, 642, 1997 72 Schlögl, A., Neuper C., and Pfurtscheller, G., Estimating the mutual information of an EEG-based Brain–Computer Interface, Biomed Technik, 47, 3, 2002 73 Pfurtscheller, G et al., Separability of EEG signals recorded during right and left motor imagery using adaptive autoregressive parameters, IEEE Trans Rehab Eng., 6, 316, 1998 74 Neuper, C., Schlögl, A., and Pfurtscheller, G., Enhancement of left-right sensorimotor EEG differences during feedback-regulated motor imagery, J Clin Neurophysiol., 16, 373, 1999 75 Obermaier, B et al., Hidden Markov Models for online classification of single trial EEG data, Pattern Recog Lett., 22, 1299, 2001 76 Obermaier, B., Müller, G.R., and Pfurtscheller, G., ‘Virtual Keyboard’ controlled by spontaneous EEG activity, IEEE Trans Neural Sys Rehab Eng., 11, 422, 2003 77 Guger, C., Ramoser, H and Pfurtscheller, G., Real-time EEG analysis for a brain– computer interface (BCI) with subject-specific spatial patterns, IEEE Trans Rehab Eng., 8, 447, 2000 78 Müller-Gerking, J., Pfurtscheller, G., and Flyvbjerg, H., Designing optimal spatial filters for single-trial EEG classification in a movement task, Clin Neurophysiol., 110, 787, 1999 79 Birbaumer, N et al., Biofeedback of event-related slow potentials of the brain, Int J Psychol., 16, 389, 1981 80 Elbert, T et al., Biofeedback of slow cortical potentials Part I, Electroenceph Clin Neurophysiol., 48, 293, 1980 81 Kotchoubey, B et al., Modification of slow cortical potentials in patients with refractory epilepsy: a controlled outcome study, Epilepsia, 42, 406, 2001 82 Birbaumer, N et al., A spelling device for the paralysed, Nature, 398, 297, 1999 83 Müller, K.R et al., An introduction to kernel-based learning algorithms, IEEE Trans Neural Networks, 12, 181, 2001 84 Perelmouter, J et al., Language support program for thought-translation-devices, Automedica, 18, 67, 1999 Copyright © 2005 CRC Press LLC 85 Cheng, M., Gao, X., and Gao, S., Design and implementation of a brain–computer interface with high transfer rates, IEEE Trans Biomed Eng., 49, 1181, 2003 86 Kennedy, P.R and Bakay, R.A.E., Restoration of neural output from a paralyzed patient by a direct brain connection, NeuroReport, 9, 1707, 1998 87 Salmoni, A.W., Schmidt, R.A., and Walter, C.B., Knowledge of results and motor learning: a review and critical reappraisal, Psycholog Bulletin, 95, 355, 1984 88 McFarland, D.J., McCane, L.M., and Wolpaw, J.R., EEG-based communication and control: short-term role of feedback, IEEE Trans Rehab Eng., 6, 7, 1998 89 Kuhlman, W.N., Functional topography of the human mu rhythm, Electroenceph Clin Neurophysiol 44, 83, 1978 90 Westphal, K.P et al., EEG-blocking before and during voluntary movements: differences between the eyes closed and the eyes open condition, Arch It Biol., 131, 25, 1993 91 Koshino, Y and Niedermeyer, E., Enhancement of rolandic mu-rhythm by pattern vision, Electroenceph Clin Neurophysiol., 38, 535, 1975 92 Cochin, S et al., Perception of motion and EEG activity in human adults, Electroenceph Clin Neurophysiol., 107, 287, 1998 93 Rizzolatti, G., Fogassi, L., and Gallese, V., Neurophysiological mechanisms underlying the understanding and imitation of action, Perspectives, 2, 661, 2001 94 Pfurtscheller, G et al., Visually guided motor imagery activates sensorimotor areas in humans, Neurosci Lett., 269, 153, 1999 95 Kübler, A et al., Brain–computer communication: unlocking the locked-in, Psychol Bulletin, 127, 358, 2001 96 Neuper, C et al., Clinical application of an EEG-based brain–computer interface: a case study in a patient with severe motor impairment, Clin Neurophysiol., 114, 399, 2003 97 Müller, G.R., Neuper, C., and Pfurtscheller, G., Implementation of a telemonitoring system for the control of an EEG-based brain computer interface, IEEE Trans Neural Systems Rehab Eng., 11, 54, 2003 98 Levine, S.P et al., Identification of electrocorticogram patterns as the basis for a direct brain interface, Clin Neurophysiol., 16, 439, 1999 99 Jensen, A and LaCour-Harbo, A., Ripples in Mathematics: The Discrete Wavelet Transform, Springer, Berlin, 2001 100 Goldberg, D.E., Genetic Algorithms in Search, Optimization, and Machine Learning, Addison-Wesley, Reading, MA, 1989 101 Graimann, B., Movement-related patterns in ECoG and EEG: visualization and detection, T.U Graz, Austria, Ph.D thesis, 2002 102 Krausz, G et al., Critical decision speed and information transfer in the Graz brain– computer interface, Appl Psychophysiol Biofeedback, 28, 233, 2003 103 Makeig, S et al., Dynamic brain sources of visual evoked responses, Science, 295, 690, 2002 104 Moller, E et al., Instantaneous multivariate EEG coherence analysis by means of adaptive high-dimensional autoregressive models, J Neurosci Meth., 105, 143, 2001 105 Meeker, D et al., Cognitive control signals for prosthetic systems, Soc Neurosci Abstr., 27, 63, 2001 106 Andersen, R.A and Buneo, C.A., Intentional maps in posterior parietal cortex, Annu Rev Neurosci., 25, 189, 2002 107 Kennedy, P.R et al., Direct control of a computer from the human central nervous system, IEEE Trans Rehab Eng., 8, 198, 2000 Copyright © 2005 CRC Press LLC A Medial Ce ntr al Sulc us , A nterio r Ban k Elbow Elbow Wrist Rostral Fundus Wrist Shoulder Shoulder Elbow Digits Digits + Wrist Digits Digits + Wrist Wrist Wrist Wrist Shoulder Shoulder Area B 10 Area 3a Area Central Sulcus mm C Hindlimb Trunk T ArS Fund us 10 CS mm EDC BIS 15 Face Distal Distal + Proximal Arm Hand + arm Bimodal/defensive Hand to Mouth Proximal COLOR FIGURE 1.2 Intracortical stimulation maps of M1 in macaque monkeys Note that in each map, hand movements form a central core (red) (A) Summary map of the movements evoked by intracortical stimulation (2–30: A) in an awake macaque monkey (Adapted with permission from Kwan, H C et al., J Neurophysiol., 41, 1120, 1978 Copyright 1978 by the American Physiological Society.) (B) Summary map of muscle representation in M1 derived from stimulus-triggered averages of rectified EMG activity (15: A at 15 Hz) in an awake monkey Sites that influenced only proximal muscles are indicated by light shading, those that influenced only distal muscles by dark shading, and those sites that influenced both proximal and distal muscles by intermediate shading Sites of significant stimulus-triggered averages of rectified EMG activity for the shorthead of biceps (BIS, blue) and extensor digitorum communis (EDC, red) are indicated with size-coded dots (3, 4, 5, S.D levels above pre-trigger level baseline activity) (Adapted with permission from Park, M C., BelhajSaif, A., Gordon, M., and Cheney, P D., J Neurosci., 21, 2784, 2001 Copyright 2001 by the Society for Neuroscience.) (C) Summary of hand and arm postures produced by long train (0.5 sec), high intensity (25–150: A) intracortical stimulation in M1, the PMd, and the PMv of an awake monkey Arm sites evoked postures involving the arm but without changes in the configuration of the hand Hand + arm indicates sites where stimulation evoked postures involving both the hand and arm Hand to mouth indicates sites that evoked grasp-like movements of the hand which was brought to the mouth Bimodal/defensive indicates sites where neurons received visual input and stimulation moved the arm into a defensive posture See text for further explanation (Adapted with permission from Graziano, M S., Taylor, C S., and Moore, T., Neuron, 34, 841–51, 2002 Copyright 2002 by Cell Press.) Copyright © 2005 CRC Press LLC e M1 Leg (Jo17) a b M1 Arm (Jo19) C M1 Face (Jo18) c d a b c -1 Ventral Distance (mm) B Dorsal A d -2 -4 e 13-16 10-12 8-9 4-7 -5 50 70 90 110 130 150 Section Number 11-13 10 7-9 3-6 100 120 140 160 180 Section Number 10-11 7-8 3-6 Caudal Dorsal -3 200 220 240 260 280 300 Section Number COLOR FIGURE 1.10 Somatotopic organization of dentate output channels to M1 Unfolded maps of the dentate illustrate the neurons labeled after HSV1 injections into the (A) leg, (B) arm, and (C) face representations of M1 These maps of the dentate were created by unfolding serial coronal sections through the nucleus Inset in part (A) illustrates a coronal section of the dentate where each segment in the unfolded map is identified The dashed vertical line indicates the rostro-caudal center of the nucleus (Adapted with permission from Dum, R P and Strick, P L., J Neurophysiol., 89, 634, 2003 Copyright 2003 by the American Physiological Society.) Copyright © 2005 CRC Press LLC COLOR FIGURE 3.3 Neuromuscular evolution A speculative scheme is illustrated through which a parent muscle (A) could become partially subdivided (B) and eventually divide into two daughter muscles (C), while still retaining some of the distributed descending neural control of the parent muscle A Parent Muscle A single tendon inserts broadly on multiple digits For simplicity only two digits are illustrated here The four motoneurons each innervate muscle fibers distributed widely in the muscle belly, so that the four motor unit territories overlap The motoneurons in turn are innervated by five descending neurons that each synapse widely within the motoneuron pool Here every descending neuron innervates every motoneuron B Partially Subdivided Muscle The tendon has become partially divided to act differentially on the two digits The motor unit territories also have become partially selective: the red and orange motoneurons innervate muscle fibers to the left, the green and blue motoneurons muscle fibers to the right, with a central region of overlap The red and orange motoneurons thus act more strongly on one digit and the green and blue motoneurons act more strongly on the other digit The descending inputs also have become more selective: the red and orange descending neurons no longer innervate the blue motoneuron; the green and blue motoneurons no longer innervate the red motoneuron; hence these descending neurons can act somewhat differentially on the digits The yellow descending neuron, however, still facilitates all four motoneurons C Daughter Muscles The tendon now has divided completely in two, as has the muscle belly The red and orange motoneurons exclusively innervate the left muscle; the green and blue motoneurons exclusively innervate the right The descending neurons also have become more, though not completely, selective: the red descending neuron now innervates only the red and orange motoneurons, and the blue descending neuron now innervates only the green and blue motoneurons These two descending neurons therefore selectively facilitate only the left or right daughter muscle, respectively The orange descending neuron facilitates the left muscle more than the right, the green descending neuron facilitates the right more than the left, and the yellow descending neuron still facilitates the left and right equally Copyright © 2005 CRC Press LLC B imanual par allel M ovement tr ajector ies Unimanual left a 35 sp/s b -750 T ime (ms) 750 cm COLOR FIGURE 4.6 Raster displays and PETHs illustrating the activity from a right MI cell in four conditions The activity of the cell during bimanual parallel movements is on the left (red) The activity of the cell during unimanual left movements is on the right (blue) The middle plots show the movement paths of the left hand for bimanual parallel (red) and unimanual left (blue) movements Row A only contains trials in which the movement path passed through a narrow band (thick green line) located between the origin and the target Row B only contains trials that did not pass through the band The green band was placed to maximize the difference between the trajectories in the lower display PETHs are centered on the beginning of movement, and the scale for all PETHs is the same The trajectories begin in the upper right and end in the lower left of the frame Note that the cell activity in bimanual trials (in red) remains similar regardless of the precise trajectories Copyright © 2005 CRC Press LLC COLOR FIGURE 4.13 Population vectors calculated for unimanual and bimanual movements to 315º For each movement, two PVs (colored arrows) of two neuronal subpopulations were calculated using an estimated best-fit PD The different colors represent the PVs of the different neuronal subpopulations A PVs constructed by dividing all cells into two subpopulations according to the hemisphere in which they reside B PVs constructed by dividing all cells into two subpopulations according to their arm preference (Reproduced with permission from Steinberg, O., Donchin, O., Gribova, A., Cardosa de Oliveira, S., Bergman, H., and Vaadia, E., Neuronal populations in primary motor cortex encode bimanual arm movements, Eur J Neurosci., 15, 1371, 2002.) Copyright © 2005 CRC Press LLC 0.6 0.2 0.6 0.4 0.2 0 0 -0.2 -0.2 -0.4 -0.4 -0.6 mEP -500 -500 T ime [ms] 500 0.4 B 500 A -0.6 mEP -500 500 Movement onset 1000 -500 500 1000 Movement onset COLOR FIGURE 4.15 Example of joint peri-event time correlograms (JPETCs) of a pair of recording sites from different hemispheres demonstrating different correlation patterns during different movements Each pixel in the JPETC represents the correlation coefficient between all the (single-trial) values of one local field potential (LFP) channel at the corresponding time bin of the x-axis and the values of a second LFP channel at the respective time bin of the y-axis Correlation is expressed as correlation coefficient (CC) and is shown in a color code, with the color scale given on the right side of the JPETC The main diagonal depicts the correlation at delay = A The correlation pattern during unimanual movements of the contralateral (left) arm to the front No correlation is apparent between the two electrodes in this condition B The correlation pattern for the same pair during a bimanual movement of the same amplitude Movement directions of the two arms differ by 90 degrees, with the left arm moving to the front (like in the unimanual condition shown in A) and the right arm moving to the right Note the strong correlation with side peaks that arises around movement onset, and lasts for about 100 msec Copyright © 2005 CRC Press LLC 0.8 0.6 0.2 0 -0.2 -0.4 -500 T ime [ms] 500 0.4 -0.6 mEP -500 500 Movement onset COLOR FIGURE 4.16 JPTEC depicting dynamics of correlation between two recording sites in different hemispheres around the time of movement onset The movements were bimanual non-symmetric with the left arm moving to the left, and the right arm to the front) Note that the high level of correlation before movement onset (CC ≈ 0.5) is significantly reduced near movement onset and throughout the movement duration COLOR FIGURE 5.2 An example of the population coding of movement direction The blue lines represent the vectorial contribution of individual cells in the population (N = 475) The actual movement direction is in yellow and the direction of the population vector is in red (From Georgopoulos, A.P., Kettner, R.E., and Schwartz, A.B., Primate motor cortex and free arm movements to visual targets in three-dimensional space II Coding of the direction of movement by a neuronal population, J Neurosci., 8, 2928, 1988, Figure 1, with permission.) Copyright © 2005 CRC Press LLC Early Force Late Baseline Early Washout Late Force Late Washout COLOR FIGURE 12.6 Psychophysics of motor learning Data are shown from a representative experimental session (A) Trajectories in real space The trajectories are roughly straight when the movements are not perturbed (baseline) When a counterclockwise (CCW) force field is turned on, trajectories are deviated at first (early force) After the perturbing force is turned off, the first movements show an aftereffect, inasmuch as they are deviated in the clockwise direction (early washout) Within a few trials, however, the monkey readapts to the unperturbed condition, and trajectories become straight again (late washout) (From Gandolfo, F., Li, C., Benda, B.J., Padoa-Schioppa, C., and Bizzi, E., Cortical correlates of learning in monkeys adapting to a new dynamical environment, Proc Natl Acad Sci USA, 97, 2259, 2000, with permission.) 90 Memory I 90 45 135 180 225 180 225 315 Memory II 315 90 45 180 315 270 45 135 225 315 270 90 180 270 225 315 135 45 225 180 270 90 45 135 270 135 90 45 135 180 225 315 270 COLOR FIGURE 12.7 The tuning curves are plotted in polar coordinates For each cell, the three plots represent the movement-related activity in the Baseline (left), in the Force epoch (center), and in the Washout (right) In each plot, the circle in dashed line represents the average activity during the center hold time window, when the monkey holds the manipulandum inside the center square and waits for instructions Examples of memory I and memory II cells, in terms of the modulation of the Pd All cells were recorded with a clockwise force field (From Li, C.S., Padoa-Schioppa, C., and Bizzi, E., Neuronal correlates of motor performance and motor learning in the primary motor cortex of monkeys adapting to an external force field, Neuron, 30, 593, 2001, with permission.) Copyright © 2005 CRC Press LLC B T ask T ask T ask C D Observed Predicted COLOR FIGURE 13.2 (A) Experimental setup and control loops, consisting of a data acquisition system, a computer running multiple linear models in real time, a robot arm equipped with a gripper, and a monkey visual display The pole was equipped with a gripping force transducer Robot position was translated into cursor position on the screen, and feedback of the gripping force was provided by changing the cursor size (B) Schematics of three behavioral tasks In task the monkey’s goal was to move the cursor to a visual target (green) that appeared at random locations on the screen In task the pole was stationary, and the monkey had to grasp a virtual object by developing a particular gripping force instructed by red circles displayed on the screen Task was a combination of tasks and The monkey had to move the cursor to the target and then develop a gripping force necessary to grasp a virtual object (C) Motor parameters (blue) and their prediction using linear models (red) From top to bottom, hand position (HPX, HPY) and velocity (HVx, HVy) during execution of task 1, and gripping force (GF) during execution of tasks and (D) Surface EMGs of arm muscles recorded in task for pole control (left) and brain control without arm movements (right) Top plots show X-coordinate of the cursor; plots below display EMGs of wrist flexors, wrist extensors, and biceps EMG modulations were absent in brain control (Extracted from Reference 10.) Copyright © 2005 CRC Press LLC A B C D E F G P O L E C O NTR O L r g r g r g r g r Virtual object g H B RAIN C O NTR O L r g r g r g r g r g r = reach g = grasp COLOR FIGURE 13.3 (A–F) Contribution of different cortical areas to model predictions of hand position, velocity (task 1), and force (task 2) For each area, neuronal dropping curves represent average prediction accuracy (R2) as a function of the number of neurons needed to attain it Contributions of each cortical area vary for different parameters Typically more than 30 randomly sampled neurons were required for an acceptable level of prediction (G–I) Comparison of the contribution of single units (blue) and multiple units (red) to predictions of HP, HV, and GF Single units and multiple units were taken from all cortical areas Single units’ contribution exceeded that of multiple units by ~20% (G, H) Representative robot trajectories and gripping force profiles in an advanced stage of training in task during both pole and brain control The bottom graphs show trajectories and the amount of the gripping force developed during grasping of each virtual object The dotted vertical lines in the panels indicate the end of reach Note that during both modes of BMI operation, the patterns of reaching and grasping movements (displacement followed by force increase) were preserved (Extracted from Reference 10.) Copyright © 2005 CRC Press LLC POSIT ION V EL OCIT Y FOR CE PM d M1 S1 SM A M1 ips COLOR FIGURE 13.4 Variability in contributions of individual neurons and cortical areas to the representation of multiple motor parameters (from top to bottom: hand position, hand velocity, and gripping force) Note the clear increase of accuracy in predictions for individual neurons and cortical areas during the day period During the same period, a high degree of variability in both neuronal and real contributions was observed The color bar indicates the sample size for each cortical area (Extracted from Reference 45.) Copyright © 2005 CRC Press LLC COLOR FIGURE 14.2 a) Simultaneous ERD/ERS in the mu (10−12 Hz) and gamma band (36−40 Hz), recorded from the left sensorimotor area and processed synchronous to the offset of voluntary right finger movements b) Superimposed ERD/ERS time courses from individual subjects (thin lines) and grand average curve (thick line) calculated for subject-specific beta frequency bands Data were recorded on electrode C3 and processed synchronous to the offset of wrist movements c) Simultaneous ERD/ERS in the mu (8−12 Hz) and gamma (70−80 Hz) band in ECoG recordings during voluntary finger movement d) ECoG electrode locations COLOR FIGURE 14.4 Left side: examples of single EEG trials recorded from electrode position C3 during right hand (upper panel) and foot (lower panel) motor imagery Right side: ERD/ERS time frequency maps and time curves of the frequency band 11−13 Hz recorded from electrode position C3 during right hand (upper panels) vs foot (lower panels) motor imagery Onset of cue presentation at second Copyright © 2005 CRC Press LLC ... cytoarchitectonic areas are delineated with dotted lines M1 and the premotor areas are shaded Abbreviations: AIP, LIP, MIP, VIP: anterior, lateral, medial, and ventral intraparietal areas; ArS: arcuate... transformational mapping, one form of nonstandard mapping Other forms of nonstandard mapping differ strongly: they are arbitrary In arbitrary sensorimotor mapping, the cue’s location has no systematic... demonstrate that rather than acting as a somatotopic array of upper motor neurons, each controlling a single muscle that moves a single finger, neurons in the primary motor cortex (M1) act as a spatially