computer science/neuroscience D T o o rn w h a e g r e d , Of related interest M B il r lá a n i New Directions in Statistical Signal Processing , n From Systems to Brains H - i C n Guido Dornhege is a Postdoctoral Researcher in the edited by Simon Haykin, José C. Príncipe, Terrence J. Sejnowski, and John McWhirter t o Toward e Intelligent Data Analysis Group at the Fraunhofer Insti- Signal processing and neural computation have separately and signiicantly inluenced many disciplines, but the cross- rb m Brain-Computer tute for Computer Architecture and Software Technol- e fertilization of the two ields has begun only recently. Research now shows that each has much to teach the other, as we p r ogy in Berlin. José del R. Millán is a Senior Researcher see highly sophisticated kinds of signal processing and elaborate hierarchical levels of neural computation performed side ge u Interfacing at the IDIAP Research Institute in Martigny, Switzer- r t by side in the brain. In New Directions in Statistical Signal Processing, leading researchers from both signal processing and , e M land, and Adjunct Professor at the Swiss Federal Insti- neural computation present new work that aims to promote interaction between the two disciplines. r edited by Guido Dornhege, José del R. Millán, c tute of Technology in Lausanne. Thilo Hinterberger is F I Thilo Hinterberger, Dennis J. McFarland, and a n with the Institute of Medical Psychology at the Univer- r Klaus-Robert Müller Dynamical Systems in Neuroscience l t sity of Tübingen and is a Senior Researcher at the Uni- a e The Geometry of Excitability and Bursting n r foreword by Terrence J. Sejnowski versity of Northampton. Dennis J. McFarland is a Re- d Eugene M. Izhikevich , fa search Scientist with the Laboratory of Nervous System a In order to model neuronal behavior or to interpret the results of modeling studies, neuroscientists must call upon methods n c Disorders, Wadsworth Center, New York State Depart- d i of nonlinear dynamics. This book offers an introduction to nonlinear dynamical systems theory for researchers and graduate n ment of Health. Klaus-Robert Müller is Head of the M g Intelligent Data Analysis Group at the Fraunhofer In- students in neuroscience. It also provides an overview of neuroscience for mathematicians who want to learn the basic facts ü l of electrophysiology. l stitute and Professor in the Department of Computer e r Toward , Science at the Technical University of Berlin. e Bayesian Brain d i Probabilistic Approaches to Neural Coding to r Interest in developing an effective communication in- edited by Kenji Doya, Shin Ishii, Alexandre Pouget, and Rajesh P. N. Rao s Brain-Computer terface connecting the human brain and a computer has A Bayesian approach can contribute to an understanding of the brain on multiple levels, by giving normative predictions grown rapidly over the past decade. A brain-computer about how an ideal sensory system should combine prior knowledge and observation, by providing mechanistic interpre- interface (BCI) would allow humans to operate comput- tation of the dynamic functioning of the brain circuit, and by suggesting optimal ways of deciphering experimental data. Interfacing ers, wheelchairs, prostheses, and other devices, using Bayesian Brain brings together contributions from both experimental and theoretical neuroscientists that examine the brain brain signals only. BCI research may someday provide mechanisms of perception, decision making, and motor control according to the concepts of Bayesian estimation. a communication channel for patients with severe physi- cal disabilities but intact cognitive functions, a working tool in computational neuroscience that contributes to a better understanding of the brain, and a novel inde- Neural Information Processing series The MIT Press edited by pendent interface for human-machine communication Massachusetts Institute of Technology that offers new options for monitoring and control. This Cambridge, Massachusetts 02142 Guido Dornhege, José del R. Millán, volume presents a timely overview of the latest BCI http://mitpress.mit.edu research, with contributions from many of the impor- Thilo Hinterberger, Dennis J. McFarland, tant research groups in the ield. The book covers a broad range of topics, describing and Klaus-Robert Müller work on both noninvasive (that is, without the implanta- 978-0-262-04244-4 0-262-04244-4 tion of electrodes) and invasive approaches. Other chap- ters discuss relevant techniques from machine learning and signal processing, existing software for BCI, and pos- foreword by Terrence J. Sejnowski sible applications of BCI research in the real world. Toward Brain-Computer Interfacing NeuralInformationProcessingSeries MichaelI.JordanandThomasDietterich,editors Advances in Large Margin Classifiers, Alexander J. Smola, Peter L. Bartlett, Bernhard Scho¨lkopf,andDaleSchuurmans,eds.,2000 Advanced Mean Field Methods: Theory and Practice, Manfred Opper and David Saad, eds.,2001 Probabilistic Models of the Brain: Perception and Neural Function, Rajesh P. N. Rao, BrunoA.Olshausen,andMichaelS.Lewicki,eds.,2002 ExploratoryAnalysisandDataModelinginFunctionalNeuroimaging,FriedrichT.Som- merandAndrzejWichert,eds.,2003 AdvancesinMinimumDescriptionLength:TheoryandApplications,PeterD.Grunwald, InJaeMyung,andMarkA.Pitt,eds.,2005 Nearest-Neighbor Methods in Learning and Vision: Theory and Practice, Gregory Shakhnarovich,PiotrIndyk,andTrevorDarrell,eds.,2006 NewDirectionsinStatisticalSignalProcessing:FromSystemstoBrains,SimonHaykin, Jose´ C.Pr´ıncipe,TerrenceJ.Sejnowski,andJohnMcWhirter,eds.,2007 PredictingStructuredData,Go¨khanBakir,ThomasHofmann,BernardScho¨lkopf,Alexan- derJ.Smola,BenTaskar,S.V.N.Vishwanathan,eds.,2007 Toward Brain-Computer Interfacing, Guido Dornhege, Jose´ del R. Milla´n, Thilo Hinter- berger,DennisJ.McFarland,Klaus-RobertMu¨ller,eds.,2007 LargeScaleKernelMachines,Le´onBottou,OlivierChapelle,DenisDeCoste,JasonWest- man,eds.,2007 Toward Brain-Computer Interfacing editedby GuidoDornhege Jose´ delR.Milla´n ThiloHinterberger DennisJ.McFarland Klaus-RobertMu¨ller forewordby TerrenceJ.Sejnowski ABradfordBook TheMITPress Cambridge,Massachusetts London,England (cid:2)c 2007MassachusettsInstituteofTechnology Allrightsreserved.Nopartofthisbookmaybereproducedinanyformbyanyelectronicormechanicalmeans (includingphotocopying,recording,orinformationstorageandretrieval)withoutpermissioninwritingfromthe publisher. Forinformationaboutspecialquantitydiscounts,[email protected]. ThisbookwassetinLaTexbytheauthors. PrintedandboundintheUnitedStatesofAmerica. LibraryofCongressCataloging-in-PublicationData Towardbrain-computerinterfacing/editedbyGuidoDornhege...[etal.];forewordbyTerrenceJ.Sejnowski. p.; cm. –(Neuralinformationprocessingseries) ”ABradfordbook.” Includesbibliographicalreferencesandindex. ISBN978-0-262-04244-4(hardcover:alk.paper) 1.Brain-computerinterfaces.I.Dornhege,Guido.II.Series. [DNLM: 1. Brain Mapping. 2. User-Computer Interface. 3. Brain–physiology. 4. Psychomotor Performance. 5.Rehabilitation–instrumentation.WL335T7372007] QP360.7.T692007 612.8’2–dc22 2007000517 10987654321 Contents Foreword . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix TerrenceJ.Sejnowski Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi 1 AnIntroductiontoBrain-ComputerInterfacing . . . . . . . . . . . . . . 1 AndreaKu¨blerandKlaus-RobertMu¨ller I BCISystemsandApproaches 27 2 NoninvasiveBrain-ComputerInterfaceResearchattheWadsworthCenter 31 EricW.Sellers,DeanJ.Krusienski,DennisJ.McFarland, andJonathanR.Wolpaw 3 Brain-Computer Interfaces for Communication in Paralysis: A Clinical ExperimentalApproach . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 ThiloHinterberger,FemkeNijboer,AndreaKu¨bler, TamaraMatuz,AdrianFurdea,UrsulaMochty,MiguelJordan, ThomasNavinLal,N.JeremyHill,Ju¨rgenMellinger, MichaelBensch,MichaelTangermann,GuidoWidman, ChristianE.Elger,WolfgangRosenstiel,Bernhard Scho¨lkopf,andNielsBirbaumer 4 Graz-Brain-ComputerInterface:StateofResearch . . . . . . . . . . . . 65 GertPfurtscheller,GernotR.Mu¨ller-Putz,AloisSchlo¨gl, BernhardGraimann,ReinholdScherer,RobertLeeb, ClemensBrunner,ClaudiaKeinrath,GeorgeTownsend, CarmenVidaurre,MuhammadNaeem,FelixY.Lee, SelinaWriessnegger,DorisZimmermann,EvaHo¨fler, andChristaNeuper vi Contents 5 The Berlin Brain-Computer Interface: Machine Learning-Based Detection ofUserSpecificBrainStates . . . . . . . . . . . . . . . . . . . . . . . . . 85 BenjaminBlankertz,GuidoDornhege,MatthiasKrauledat, VolkerKunzmann,FlorianLosch,GabrielCurio, andKlaus-RobertMu¨ller 6 The IDIAP Brain-Computer Interface: An Asynchronous Multiclass Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103 Jose´ delR.Milla´n,PierreW.Ferrez,andAnnaButtfield 7 BrainInterfaceDesignforAsynchronousControl . . . . . . . . . . . . . 111 JaimieF.Borisoff,SteveG.Mason,andGaryE.Birch II InvasiveBCIApproaches 123 8 ElectrocorticogramasaBrain-ComputerInterfaceSignalSource . . . . 129 JaneE.Huggins,BernhardGraimann,SeYoungChun, JeffreyA.Fessler,andSimonP.Levine 9 Probabilistically Modeling and Decoding Neural Population Activity in MotorCortex . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147 MichaelJ.BlackandJohnP.Donoghue 10 TheImportanceofOnlineErrorCorrectionandFeed-ForwardAdjustments inBrain-MachineInterfacesforRestorationofMovement . . . . . . . . . 161 DawnM.Taylor 11 Advances in Cognitive Neural Prosthesis: Recognition of Neural Data with anInformation-TheoreticObjective . . . . . . . . . . . . . . . . . . . . . 175 ZoranNenadic,DanielS.Rizzuto,RichardA.Andersen, andJoelW.Burdick 12 A Temporal Kernel-Based Model for Tracking Hand Movements from NeuralActivities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191 LaviShpigelman,KobyCrammer,RonyPaz,EilonVaadia, andYoramSinger III BCITechniques 203 13 GeneralSignalProcessingandMachineLearningToolsforBCIAnalysis 207 GuidoDornhege,MatthiasKrauledat,Klaus-RobertMu¨ller, andBenjaminBlankertz Contents vii 14 Classifying Event-Related Desynchronization in EEG, ECoG, and MEG Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 N.JeremyHill,ThomasNavinLal,MichaelTangermann, ThiloHinterberger,GuidoWidman,ChristianE.Elger, BernhardScho¨lkopf,andNielsBirbaumer 15 Classification of Time-Embedded EEG Using Short-Time Principal ComponentAnalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261 CharlesW.Anderson,MichaelJ.Kirby,DouglasR.Hundley, andJamesN.Knight 16 Noninvasive Estimates of Local Field Potentials for Brain-Computer Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 RolandoGravedePeraltaMenendez,SaraGonzalezAndino, PierreW.Ferrez,andJose´ delR.Milla´n 17 Error-RelatedEEGPotentialsinBrain-ComputerInterfaces . . . . . . . 291 PierreW.FerrezandJose´ delR.Milla´n 18 AdaptationinBrain-ComputerInterfaces . . . . . . . . . . . . . . . . . . 303 Jose´ delR.Milla´n,AnnaButtfield,CarmenVidaurre, MatthiasKrauledat,AloisSchlo¨gl,PradeepShenoy, BenjaminBlankertz,RajeshP.N.Rao,RafaelCabeza, GertPfurtscheller,andKlaus-RobertMu¨ller 19 EvaluationCriteriaforBCIResearch . . . . . . . . . . . . . . . . . . . . 327 AloisSchlo¨gl,JulienKronegg,JaneE.Huggins,andSteveG.Mason IV BCISoftware 343 20 BioSig:AnOpen-SourceSoftwareLibraryforBCIResearch . . . . . . . 347 AloisSchlo¨gl,ClemensBrunner,ReinholdScherer, andAndreasGlatz 21 BCI2000:AGeneral-PurposeSoftwarePlatformforBCI . . . . . . . . . 359 Ju¨rgenMellingerandGerwinSchalk V Applications 369 22 Brain-Computer Interfaces for Communication and Motor Control—PerspectivesonClinicalApplications . . . . . . . . . . . . . . . 373 AndreaKu¨bler,FemkeNijboer,andNielsBirbaumer viii Contents 23 CombiningBCIandVirtualReality:ScoutingVirtualWorlds . . . . . . 393 RobertLeeb,ReinholdScherer,DoronFriedman, FelixY.Lee,ClaudiaKeinrath,HorstBischof, MelSlater,andGertPfurtscheller 24 Improving Human Performance in a Real Operating Environment through Real-TimeMentalWorkloadDetection . . . . . . . . . . . . . . . . . . . 409 JensKohlmorgen,GuidoDornhege,MikioL.Braun, BenjaminBlankertz,Klaus-RobertMu¨ller,GabrielCurio, KonradHagemann,AndreasBruns,MichaelSchrauf, andWilhelmE.Kincses 25 Single-TrialAnalysisofEEGduringRapidVisualDiscrimination:Enabling CorticallyCoupledComputerVision . . . . . . . . . . . . . . . . . . . . . 423 PaulSajda,AdamD.Gerson,MariosG.Philiastides, andLucasC.Parra References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 503 Foreword The advances in brain-computer interfaces in this book could have far-reaching conse- quences for how we interact with the world around us. A communications channel that bypasses the normal motor outflow from the brain will have an immediate benefit for paraplegic patients. Someday the same technology will allow humans to remotely con- trolagentsinexoticenvironments,whichwillopennewfrontiersthatwecanonlydimly imaginetoday. The earliest systems to be developed were based on noninvasive electroencephalo- graphic(EEG)recordings.Becausethesesystemsdonotrequireinvasivesurgicalimplants, they can be used for a wide range of applications. The disadvantage is the relatively low rate of signaling that can be achieved. Nonetheless, advances in signal processing tech- niquesandthedevelopmentofdryelectrodesmakethisanattractiveapproach. Three separate research areas have contributed to major advances in invasive brain- computer interfaces. First, the neural code for motor control was uncovered based on recordingsfromsingleneuronsindifferentcorticalareasofalertprimates.Thesecondwas the development of mathematical algorithms for converting the train of spikes recorded from populations of these neurons to an intended action, called the decoding problem. Third,itwasnecessarytoachievestable,long-termrecordingsfromsmall,corticalneurons inaharshaqueousenvironment. For both invasive and noninvasive BCIs interdisciplinary teams of scientists and engi- neersneededtoworkcloselytogethertocreatesuccessfulsystems. Successinbrain-computerinterfaceshasalsodependedontheremarkableabilityofthe braintoadapttounusualtasks,nonemorechallengingthan“mindcontrol”ofextracorpo- real space. We are still at an early stage of development, but the field is moving forward rapidlyandwecanconfidentlyexpectfurtheradvancesinthenearfuture. TerrenceJ.Sejnowski LaJolla,CA