figure 1.1 Thearchitectureofaneural circuit.Reproductionofa RamnyCajaloriginal drawingshowinganeural circuitformedbymany neurons.Onesingleneuron anditscellularspecializationsarehighlighted.In general,dendritesserve asthemainneuronal specializationreceiving synapsesfromother neurons.

Axonterminals establishtheneurons synapseswithotherbrain

cells.(Cajals drawing from Histology of the Nervous System was reproduced with permission of the Cajal Legacy, Instituto Cajal [CSIC], Madrid, Spain.)

figure 1.2 Use of graph theory to study the distribution of pathways linking

(Courtesy of Dr. Miguel Nicolelis; redrawn by Dr. Nathan Fitzsimmons, Duke University.)

pairsofneurons.Ontop,asquarematrixisusedtorepresentthedirect,monosynapticconnectivityofasmallbraincircuit.Inthismatrix,1representstheexistence ofadirectconnectionbetweenapairofbrainstructures,while0depictsitsabsence. Nexttothematrix,agraphisusedtorepresentthecircuit.Circleswithnumbers representthestructuresanddirectionalarrowsrepresentthedirectconnectivity informationcontainedinthesquarematrix.Thehistogrambelowdepictsthetotal numberofpathwayslinkingtwostructures(carotidbaroreceptorandcerebellum) that did not share a direct, monosynaptic connection. The X axis represents the numberofsynapsesofthepathwaysandtheYaxisdepictsthenumberofpathways found. Notice that millions of pathways ere found for this par ic ar example. w t ul

Youngandagraphicrepre en a ionofhistheoryontheright.Noticethatanycolor s t t stimulus(P,Q,R,orSintheXaxisoftheplotonthebottomleft)canberepresented by the graded response of three distinct color receptors, which responds maximally to red, green, and blue respectively, but can also respond submaximally to different colors. (Youngs portrait was reproduced with permission of the National Portrait Gallery, London. The figure was originally published in M.A.L. Nicolelis, Brain-Machine Interfaces to Restore Motor Function and Probe Neural Circuits. NatureReviewsNeuroscience4 [2003]: 41722.)

figure 2.1 ThomasYoungstrichromatictheoryofcolor.AportraitofThomas

figure 2.2 uttingtheneuronssupremacyinink.DonSantiagoRamnyCajal P

tology of the Ner vous System and the photo were reproduced with permission of the Cajal Legacy, Instituto Cajal [CSIC], Madrid, Spain.)

athisfavoriteplace,infrontofamicroscope,andafewofhismasterpiecedrawings ofdifferentportionsofthecentralner oussystem.(Cajals three drawings from Hisv

figure 3.1 orbinianBrodmannscorticalcytoarchitectonicmapandthesixlayK

with permission of the Cajal Legacy, Instituto Cajal [CSIC], Madrid, Spain. Brodmanns Areas, originally published in 1910, are in the public domain.)

ersofthecortex.AsideviewofthehumanbrainwiththeoriginalnumeraldesignationsofcorticalareascreatedbyBrodmannisontheright.Acomparisonofthe sixlayersofasectionthroughtheprimarymotor(M1)andvisual(V1)corticesis shownontheleft.Cytoarchitectonicallyspeaking,theM1ischaracterizedbythe presenceoflargepyramidal-ookingneurons(Betzcells)inlayerV,whileV1exhibl itsaverydenseclusterofneuronsinthebottompartoflayerIVandupperpartof layerVI.(Cajals two drawings from Histology of the Nervous System were reproduced

figure 3.2 The homunculus meets the rattunculus. The drawing depicts an impossiblemeetingbetweenthereconstructionofthecorticalhomunculus,the distortedrepre en a ionofahumansbodyintheprimarysomatosensorycortex, s t t basedonWilderPenfieldsstudies,andacorticalrattunculus,theequivalentdistorted repre en a ions of the rats body, in the rodent primary somatosensory s t t cortex.Noticetheoverrepre en a ionofthelipsandhandsinthehomunculusand s t t the facial whis ers, snout, and forepaws in the rattunculus. The cheese is Swiss. k

(Illustrated by Dr. Nathan Fitzsimmons, Duke University.)

figure 3.3 RevolutionaryplasticityexperimentinowlmonkeysbyJonKaasand

MichaelMerzenich.Ontheleft,JonKaasappearswithaclosecollaborator.Onthe right,thetopshelfdescribeshow,afteratraumaticamputationofthethirdfinger, theterritoryrepresentingthisfingerintheprimarysomatosensorycortexofan owlmonkeydoesnotremainsilent.Instead,theregionthatwasoccupiedbyfinger3 isnow invaded byenlarged repre en a ionsof fingers2and4.Thelower shelf s t t shows thatenlargementoftherepre en a ionsoffingers14canbeobtainedby s t t repetitive,selectivestimulationsofthesedigitsindetrimentoffinger5.Toseethe effect,justcomparethemaponthebottommiddle(beforeselectivestimulation) withthecorticalmaponthebottomright(afterstimulation).(Adapted from, M. M.
Merzenich, J. H. Kaas, J. Wall, R. I. Nelson, M. Sur, and D. Felleman, Topographic Reorganization of Somatosensory Cortical Areas 3B and 1 in Adult Monkeys Following Restricted Deafferentation, Neuroscience 8, no. 1 [1983]: 3355, with permission from Elsevier.)

figure 3.4 Thebrainadoptsarubberhand.Thedrawingdepictstheexperimental apparatus used to induce the rubber hand illusion. See the text for details.

(Illustrated by Dr. Nathan Fitzsimmons, Duke University.)

figure 4.1 ampleoffirstEEGrecordingeverobtainedbyHansBerger.Thetrace S

schen, Eu o e n Archives of Psychiatry and Clinical Neuroscience 87, no. 1 [1929]: r p a 52770, with permission from Springer.)

represents a few seconds of the electrical activity of Bergers own son, recorded usingscalpsensors. (From Hans Berger, ber das Elektrenkephalogramm des Men-

electrode,positionedintheextracellularspacenexttotwoneurons,iscapableof recordingtheextracellularactionpotentialsofbothcells.Inspectionoftherecordingtraceintheoscilloscoperevealsthattheactionpotentialsofthetwoneurons have different shapes and magnitudes, allowing the distinction between the two cells.(Illustrated by Dr. Nathan Fitzsimmons, Duke University.)

figure 4.2 esinglemicroelectroderecordingmethod.AsinglemetalmicroTh

W. H. Freeman, 1980], with permission from Mullard Radio Astronomy Observatory [MRAO] and Cavendish Laboratory, Cambridge.)

figure 4.3 -Dimageofasmallpatchoftheskyproducedbyanarrayofradio 3 telescopesinCambridge,En land.Peaksidentifygalaxiesthatemitradiosignals. g Theheightofeachpeakindicatesthemagnitudeoftheradiosignalproducedby eachgalaxy. (Originally published in Joseph Silk, TheBigBang[San Francisco, Calif.:

figure 4.4 Engineering a better way to listen to the brain. On the left, a high-

powermagnificationofamulti- lectrodearrayproducedattheDukeUniversity e CenterforNeuroengineering(DUCN)byGaryLehewandJimMeloy.Noticethat multiplethinmetalfilamentsareclusteredinamatrixformat.Suchfilamentsare flexibleandcanbechronicallyimplantedinthebrain,remainingactiveformonths toyears.Ontheright,asampleofdifferenttypesofmulti-electrodearrayscreated attheDUCNinthepastdecade.(Courtesy of Dr. Miguel Nicolelis.)

figure 5.1 Anexperimentalsetupdesignedfortestingtheabilityoffreelybehavingratstousetheirfacialwhis erstodiscriminatethediameterofanaperturein k thedark.Esheisseenintherightframeperformingthetaskwithgusto!(Originally

published in D. J. Krupa, M. C. Wiest, M. Laubach, and M.A.L. Nicolelis, Layer Specific Somatosensory Cortical Activation During Active Tactile Discrimination, Science 304 [2004]: 198992.)

electroderecordingmethods.Thetopgraphrelatesthespatialandtemporalresolutionofmosttechniquesusedtoinvestigatebrainfunction.Thelowergraphcompares the single and multi-electrode recording method using the same pa am ers. r et
(Adapted from A. Grinvald and R. Hildesheim, VSDI: A New Era in Functional Imaging of Cortical Dynamics. NatureReviewsNeuroscience 5 [2004]: 87485, with permission from Macmillan Publishers Ltd.)

figure 5.2 emporalandspatialresolutionofthesingleelectrodeversusmultiT

corner.(Courtesy of Dr. Miguel Nicolelis.)

figure 5.3 Manysingleneurons. Computerscreenimage depictingtheaction potentialsproducedbya sampleof394cortical neuronsrecorded simultaneouslyinafreely behavingprimate.The leftmosthalfofthepicture showsfourdistinctfamilies ofactionpotentials, recordedsimultaneously fromasingle microelectrodeofanarray, demonstratingtheelectrical activityoffourdifferent corticalneuronssampled simultaneously.Thesefour distinctneuronsareshown inisolationinthelowerleft

figure 5.4 Thewhiskermapoftheratsface.Theleftpaneldepictsthedistributionoffacialwhis ersintheratsnoutonfourrowsandmanycolumns.Theright k halfofthefiguredepictsahorizontalsectionthroughlayerIVoftheratprimary somatosensorycortex(S1)thatcontainstheentirerattunculus,includingawhiskerrepre en a ion(barrelcortex),nose(N),lowerjaw(LJ),forepaw(FP),andhinds t t paw (HP), and is stained for the presence of a mitochondrial enzyme found in neurons. Dark clusters represent clusters of neurons in layer IV. Notice that the barrelcortexcontainsanisomorphicrepre en a ionofwhiskerrowsandcolumns. s t t CirclesidentifywhiskerC2bothintheratfaceandtheS1cortex.

(Courtesy of Drs. John Chapin and Rick Lin.)

figure 5.5 elefthalfofthefiguredepictstheconnectivityofsomeofthemain Th brainstructuresthatdefinetherattrigeminalsomatosensorysystem.Excitatory(+) andinhibitory(-)connectionsareshown.Themechanicalstimulationoffacialwhiskerstriggerselectricalresponsesofneuronsinthetrigeminalganglion(Vg).Vgneuronsprojecttotwodistincttrigeminalnucleiinthebrainstem:thespinal(SpV)and principal(PrV)nuclei.Thesetwosendnervepathwaystothreethalamicnuclei:the ventroposteriormedialnucleus(VPM),theposteriormedial(POM)nucleus,and thezonaincerta(ZI).Thethalamicreticularnucleus(RT)providesinhibitiontothe VPMandPOM.VPM,POM,andZIprovidethalamicnervefiberstotheprimary somatosensorycortex.Ofthose,theZIistheonlyonethatsendsinhibitoryafferents totheS1cortex.Intherighthalfofthefigure,astackof3-Dgraphsillustratesthe simultaneouslyrecordedtactile-evokedresponsesofpopulationsofindividualneuronsatdifferentlevelsofthetrigeminalsystem. (Adapted from M.A.L. Nicolelis, L. A.

Baccala, R.C.S. Lin, and J. K. Chapin, Sensorimotor Encoding by Synchronous Neural Ensemble Activity at Multiple Levels of the Somatosensory System. Science 268 [1995]: 135358; and from M.A.L. Nicolelis, A. A. Ghazanfar, B. Faggin, S. Votaw, and L.M.D. Oliveira, Reconstructing the Engram: Simultaneous, Multisite, Many Single Neuron Recordings.Neuron 18 [1997]: 52937, with permission from Elsevier.)

figure 5.6 PMperi- timulustimehistograms.Fourperi- timulustimehistoV s s

Duke University.)

gramsillustratetypicalaveragedelectricalresponsesofsingleVPMneuronsfollowingthedeflectionoffacialwhis ers.Ineachhistogram,theXaxisrepresents k the peri- timulus time, 0 indicates the time of whisker deflection. The Y axis s depictsthenumberofspikesproducedbythecell. (Courtesy of Dr. Miguel Vieira,

figure 5.7 patiotemporal RF and maps. (A) Spatiotemporal receptive S field(RF)ofasingleVPMneuron.Each3-Dgraphrepresentsthespatial domain (RF) of a single VPM neuron at a par ic ar post- timulus time t ul s interval(5 0ms,20 5ms,35 0ms).Ineach3-DgraphtheXandYaxes 1 2 5 depictthepositionofwhis ersintherowsandcolumnsfoundintherats k face. The Z axis represents the magnitude of the VPM neurons firing response when one par ic ar whisker is mechanically deflected. Notice t ul thatat5 0ms,whiskerE1elicitsthestrongestfiringresponseoftheVPM 1 neuron, while stimulation of other whis ers produces somewhat smaller k responses.Yet,at35 0msafterthewhiskerstimulus,whiskerE4triggers 5 thestrongestresponseofthesamecell.Thus,thespatialcenteroftheRFof thisVPMneuronshiftsasafunctionofpost-stimulustime.(B)SpatiotemporalhistogramdepictingthetactileresponsesofapopulationofsimultaneouslyrecordedVPMneurons.Post-stimulustimeisrepresentedintheX axis,with0markingtheonsetofwhiskerstimulation.TheYaxisdepictsa number of individual VPM neurons recorded simultaneously. The gray- shadedZaxisillustratesthemagnitudeoffiringoftheseVPMneuronsasa functionoftime.(C)SpatiotemporalRFofasingleneuronlocatedinthe ratprimarysomatosensorycortex.Ineachofthese3-Dgraphs,theXaxis representswhiskercolumns,theYaxisrepresentswhiskerrows,andtheZ axis(gray- cale)representsthemagnitudeofasinglecorticalneuronresponse. s Each3-Dgraphdepictsapar ic arpost- timulustimeinterval(8 2,12 t ul s 1 16, 16 0, 20 4, 24 8 ms). Notice that, like in the VPM, the spatial 2 2 2 domainoftheRFchangesasafunctionofpost-stimulustime.(Adapted from

M.A.L. Nicolelis, L. A. Baccal, R.C.S. Lin, and J. K. Chapin, Sensorimotor Encoding by Synchronous Neural Ensemble Activity at Multiple Levels of the Somatosensory System. Science 268 [1995]: 135358; from M.A.L. Nicolelis, A. A. Ghazanfar, B. Faggin, S. Votaw, and L.M.O. Oliveira, Reconstructing the Engram: Simultaneous, Multisite, Many Single Neuron Recordings. Neuron 18 [1997]: 52937, with permission from Elsevier; from M.A.L. Nicolelis, and J. K. Chapin, The Spatiotemporal Structure of Somatosensory Responses of Many-Neuron Ensembles in the Rat Ventral Posterior Medial Nucleus of the Thalamus.Journal ofNeuroscience14 [1994]: 351132, with permission; and from A. A. Ghazanfar and M.A.L. Nicolelis, Spatiotemporal Properties of Layer V Neurons in the Rat Primary Somatosensory Cortex.CerebralCortex9 [1999]: 34861, with permission from Oxford Journals.)

150 0 150

SpV

Whisker Twitching Onset

Barrel Cx Intracranial LFP Amplitude

Whisker Twitching Onset

VPM

VPM

Spikes/s

0 150 0 200

S1PC1

CP

S1PC2

HI

64

66

Time (s)

68

Time (s)

10

15

20

figure 5.8 Examplesof7 4Hzrhythmicmuoscillationsobservedintherattri1 geminalsomatosensorysystem.Intheleftpanel,differenttracesobtainedsimultaneouslyillustratethatmuoscillationsstartintheS1cortex,spreadtotheVPM, andlatertothespinalcomplexofthetrigeminalbrain-stemcomplex(SPV)before whiskertwitchingmovementsbegin.Intherightpanel,asimilarillustrationof therelationshipbetweenmurhythmandwhiskertwitchingismadebetweenthe barrelcortex(ratS1whiskerarea),theVPM,thebasalganglia(CP),andthehippocampus (HI). (Originally published in M.A.L. Nicolelis, L. A. Baccala, R.C.S. Lin,

and J. K. Chapin, Sensorimotor Encoding by Synchronous Neural Ensemble Activity at Multiple Levels of the Somatosensory System.Science268 [1995]: 135358.)

figure 5.9 eri- vent histograms depict the firing response pattern of a single P e cortical neuron in the rat primary somatosensory cortex under three different behavioral conditions: active tactile discrimination in a freely behaving animal (leftmostpanel),awakebutimmobilized(centerpanel),andimmobilizedandpassive discrimination (rightmost panel). Notice how the pattern of this neurons responsesistotallydifferentaccordingtotheanimalsbehavioralcontext.Foreach histogram, the X axis represents peri- vent time, with 0 indicating the onset of e facialwhiskermechanicalstimulation,andtheYaxisrepresentstheelectricalfiringresponseoftheneuroninspikespersecond.(Originally published in D. J. Krupa,

M. C. Wiest, M. Laubach, and M.A.L. Nicolelis, Layer Specific Somatosensory Cortical Activation During Active Tactile Discrimination.Science304 [19891992, 2004].)

ing depicting the general or a i a ion of a brain- achine interface. Multi- g nz t m electrodearraysandmicrochipsareusedtorecordlarge- calebrainactivity.Signal s pro essing techniques are then used to translate raw brain activity into digital c commands that can be employed to reproduce, in a robotic arm, the voluntary motorintentionsgeneratedbythebrain.Visual,tactile,andproprioceptivefeedback signals from the robotic actuator are then sent back to the subjects brain.
(Originally published in M.A.L. Nicolelis and M. A. Lebedev, Principles of Neural Ensemble Physiology Underlying the Operation of Brain-Machine Interfaces. Nature ReviewsNeuroscience10 [2009]: 53040.)

figure 6.1 edebutofthebrain- achineinterface.SystemsengineeringdrawTh m

figure 6.2 eneralalgorithmfortranslatingrawneuronalelectricalactivityinto G

Dr. Nathan Fitzsimmons,Duke University.)

digitalcommandsthatcanbeemployedtopredictkinematicpa am ersandconr et trolartificialtoolsbasedonbrainactivityalone(seetextfordetails).(Illustrated by

tions,derivedfromBellesandCarmensbrainactivity,andhowwelltheyreproducetheactualpositionofthesetwoowlmonkeyshands.Thebottompanelshows thatthesamekinematicpredictionscanbeusedtocontrolsimultaneouslyarobot armlocatednexttotheanimalsatDukeUniversityorremotely,atMIT.(Originally

published in J. Wessberg, C. R. Stambaugh, J. D. Kralik, P. D. Beck, J. K. Chapin, J. Kim, S. J. Biggs, M. A. Srinivasan, and M.A.L. Nicolelis, Real-Time Prediction of Hand Trajectory by Ensembles of Cortical Neurons in Primates.Nature408 [2000]: 361 65.)

figure 6.3 Where will Belles wrist go? The graphs illustrate real- ime predict

figure 6.4 uroralosesherjoystickandfreeshermind.Ontop,theexperimenA talsetupemployedforAuroratouseherbrainactivityalonetocontrolthemovementsofaroboticarm.Onthebottomleft,asampleoftheelectricalactivityof 96corticalneuronsfromAurorasbrain.OntheYaxis,eachverticalbarrepresents anactionpotentialproducedbyasinglecorticalneuron.TheXaxisrepresents time(10seconds).Intherightpanel,agraphicrepre en a ionofthetaskAurora s t t performed and examples of predictions of arm movements based on combined brain activity. (Originally published in J. M. Carmena, M. A. Lebedev, R. E. Crist,

J.E.ODoherty, D. M. Santucci, D. R. Dimitrov, P. G. Patil, C. S. Henriquez, and M.A.L. Nicolelis, Learning to Control a Brain-Machine Interface for Reaching and Grasping by Primates.PublicLibraryofScience1 [2003]: 193208.)

figure 7.1 owmanyneuronsdoesittake?Twoneuron- roppingcurvescorreH d

Crist, J. E. ODoherty, D. M. Santucci, D. R. Dimitrov, P. G. Patil, C. S. Henriquez, and M.A.L. Nicolelis, Learning to Control a Brain-Machine Interface for Reaching and Grasping by Primates.PublicLibraryofScience1 [2003]: 193208.)

late the number of neurons (X axis) with the accuracy of real- ime prediction t obtainedfortwodistinctpa am ers(handpositionandhandgrippingforce).The r et samesampleofneuronsfromtheprimarymotorcortexofamonkeywasusedto constructthesecurves. (Originally published in J. M. Carmena, M. A. Lebedev, R. E.

Position

Gripping Force

0.8
Motor Cortex (M1)

0.8

Motor Cortex Posterior Parietal Cortex (PP)

R2 of Prediction

0.4

R2 of Prediction
Posterior Parietal Cortex

0.6

0.6

0.4

0.2

0.2

20

40

60

20

40

60

Number of Neurons

Number of Neurons

figure 7.2 athering motor commands all over the brain. Neuron dropping G curvesareusedtocomparetheaccuracyofpredictionofpopulationsofneurons simultaneously recorded in the primary motor cortex (M1, traced line) and the posteriorparietalcortex(PP,dottedline)fortwopa am ers(handpositionand r et grippingforce).Noticethatinformationaboutbothpa am ersisavailableinboth r et corticalregions,butthatM1containsmoreinformation,forequalneuronalpopulations,forhandposition.Yet,bothM1andPPyieldsimilarlevelsofaccuratepredictions for gripping force, considering neuronal populations of the same size.

(Originally published in J. M. Carmena, M. A. Lebedev, R. E. Crist, J. E. ODoherty, D. M. Santucci, D. R. Dimitrov, P. G. Patil, C. S. Henriquez, and M.A.L. Nicolelis. Learning to Control a Brain-Machine Interface for Reaching and Grasping by Primates. Public LibraryofScience1 [2003]: 193208.)

figure 7.3 Fine-tuningofcorticalneuronsthatcontrolthemovements ofthebodyandmachines. Gray-scaledpolarplotsofthefiringrateofone M1 neuron asafunctionofbotharmvelocity (key, topleft)fordifferent lagswithrespecttoinstantofinstantaneousarmvelocitymeasurement (0ms).Velocity=0isatthecenterofeachcircleandmaximumvelocity(14 cms/second)isattheperimeterofthecircle.Firingpatternswereobtained duringdifferentmodesofoperation(polecontrolandbraincontrolwith and without hand movements), and during the use of different actuators (handorrobot movements,seelegends).Eachofthecirclesalsoencodes the neurons preferred direction (see scale). Gray shading indicates firing rate(minimumiswhite,maximumisdarkgray).(A)Aneighboringneuron thatexhibitsstrongvelocityanddirectiontuningonlywhentheanimalis usingitshandtoplaythevideogame,butnottherobot.(B)AsingleM1 neuronthatdisplaysbothvelocityanddirectiontuningduringbothpole andbraincontrolandwhentheanimalisusingitshandoronlytherobotto playthevideogame.(C)AneighboringM1neuronthatexhibitsenhanced velocity and direction (dashed arrows) tuning only when the monkey is preparingtouseitsbrainactivitytomovetherobotarm,butnotwhenit movesitsownbiologicalarm. (Originally published in M. A. Lebedev, J. M.

Carmena, J. E. ODoherty, M. Zacksenhouse, C. S. Henriquez, J. Principe, and M.A.L. Nicolelis. Cortical Ensemble Adaptation to Represent Velocity of an Artificial Actuator Controlled by a Brain Machine Interface. JournalofNeuroscience25 [2005]: 468193, with permission.)

M.A.L. Nicolelis, Extracting Kinematic Parameters for Monkey Bipedal Walking from Cortical Neuronal Ensemble Activity. FrontiersinIntegrativeNeuroscience 3 [2009]: 119.)

figure 8.1 Experimentalsetupemployedtocreateabrain- achineinterfacefor m locomotion. (Originally published in N. Fitzsimmons, M. A. Lebedev, I. Peikon, and

figure 8.2 inematicpredictions,derivedfromcombinedrawbrainactivity,for K differenttypesofbipedallocomotionbehaviors:slowforwardwalking(topshelf), fast forward walking (middle shelf), and variable speed (lower shelf). Black trace depictsactualpositionofIdoyaslegwhereasthegraytracesyieldreal- imepredict tionsofthiskinematicpa am erobtainedfromherbrainactivityalone.(Originally r et

published in N. Fitzsimmons, M. A. Lebedev, I. Peikon, and M.A.L. Nicolelis, Extracting Kinematic Parameters for Monkey Bipedal Walking from Cortical Neuronal Ensemble Activity. FrontiersinIntegrativeNeuroscience 3 [2009]: 119.)

figure 8.3 doyaandCB- sgreatleapacrosstheglobe.GeneralschematicrepreI 1 sen a ionoftheexperimentthatallowedamonkeyontheU.S.EastCoasttouseits t t brainactivitytocontrolthelegmovementsofahumanoidrobot(CB-1)inKyoto, Japan,whilereceivingvisualfeedbackbackinDurham,NorthCarolina,fromthe robotlocomotion.(Courtesy of Dr. Miguel Nicolelis.)

figure 9.1 lbertoSantos- umontandhisflyingmachine.Aphotographofthe A D

Space Museum, Smithsonian Institution Archives, Washington, D.C.)

aviator(ontheleft)andthehistoricmomentwhenhecircumnavigatedtheEiffel Tower on October 19, 1901. (Reprinted with permission from the National Air and

Distal-type neurons

sensory receptive eld

visual receptive eld

visual receptive eld

visual receptive eld

Before tool use

After tool use

Passive holding

Proximal-type neurons
sensory receptive eld visual receptive eld

visual receptive eld

Before tool use

After tool use

figure 9.2 SummaryoftheexperimentscarriedoutbyDr.AtsushiIrikiandcol-

Tools for the Body (Schema).TrendsinCognitiveSciences8, no. 2 [2004]: 7986, with permission from Elsevier.)

leaguesshowingthatthevisualreceptivefieldofaparietalcorticalneuronexpands whentheanimalemploysasimpletooltoperformatask.Onthetopshelf,asingle neuronwithbothatactileandvisualRFcenteredonthehandchangeditsvisual receptivefieldtoincludetheentiretoolusedbytheanimaltoretrievesomefood reward.Noticethatwhentheanimalmerelyholdsthetool,butdoesnotutilizeitto performatask,thevisualRFremainscenteredontheanimalshandalone.Onthe bottomshelf,anotherneuronwiththetactileRFcenteredontheanimalsshoulder andabroadvisualRFshowsthesameexpansionofthevisualRFwhentheanimal utilizesatoolina3-Dspace.NoticethatthevisualexpansionoftheRFincludes theentirespaceinwhichthetoolcanreach.(Adapted from A. Maravita and A. Iriki,

versinthe1960s.Ontheright,adrawingrepresentshowthetheoryproposedin thisbookpredictswhatPelssensorimotorcortexwouldlooklike.Accordingtothis view,thesoccerballwouldbeincorporatedintoPelsfootrepre en a ioninthe s t t cortex.(From Swedish press image in the public domain.)

figure 9.3 eleftpanelshowsPelinoneofhischaracteristicshootingmaneuTh

stimulation.(Photographs used with permission of Dr. Jos Delgado.)

figure 10.1 Theneuroscientistin thebullring.This sequenceofphotographsillustratesthe classicalexperiment carriedoutbyDr.Jos Delgadoinwhicha chargingbullwasmade tostopitschargeusing deepelectricalbrain

the leftpanel, Dr.John K. Chapin andhis creation.Below,roboratwalksoverametal mesh.(Courtesy of Dr. John Chapin.)

figure 10.2 ohnChapinandroborat.In J

figure 10.3 alking back to the brain. Top shelf illustrates the experimental T paradigmutilizedtodeliverelectricalmessagestothebrainofamonkey.Chronically implanted microelectrode arrays are employed to deliver spatiotemporal electricalpatternsthatrepresentdifferentmessages.Middleshelfillustratesthedifferent types of patterns utilized to deliver messages to the primate brain: basic amplitude discrimination, temporal discrimination, or spatiotemporal discrimination. Lower shelf illustrates learning curves for each of the three methods to delivermessages. (Adapted from N. Fitzsimmons, W. Drake, T. Hanson, M. Lebedev,

and M.A.L. Nicolelis, Primate Reaching Cued by Multichannel Spatiotemporal Cortical Microstimulation.JournalofNeuroscience27 [2007]: 5593 602.)

figure 10.4 DrawingdepictingafutureexperimentinwhichanR6- ratwillbe T implantedwithamagneticfieldsensorthatdeliverselectricalmicrostimulation totheanimalsprimarysomatosensorycortexproportionaltomagneticfieldsof differentmagnitudes.Eachofthesedifferentmagneticfieldsidentifiesdifferent objects,thingslikefood,water,andthelocationofatoyrat.(Illustrated by Dr. Nathan

Fitzsimmons, Duke University.)

figure 10.5 eexplorerratandthedecoderrat.Illustrationofatruebrain- o- Th t

Nathan Fitzsimmons, Duke University.)

braininterfacelinkinganexplorerrat,whichusesitsfacialwhis erstodiscrimik natethediameterofavariableaperture,withadecoderratwhosemainfunctionis toindicatethediameteroftheaperturebasedonthepatternofbrainactivitytransmittedbytheexplorerrat,withoutevertouchingtheaperturewithitswhis ers. k Thebrain- o- raininterfaceconnectsthebrainsofthetworats. (Illustrated by Dr. t b

figure 10.6 differentversionofabrain- o- raininterfaceinvolvinganinterA t b mediary layer of rats between the explorer and decoder rats. (Illustrated by Dr. Nathan Fitzsimmons, Duke University.)

figure 11.1 pectrograms(topfour3Dgraphs)andcorrespondingrawlocalfield S potentials(5graphsinthelowershelf)illustratethefrequencyandgeneraltime p attern of different types of brain oscillations observed under different types of behaviors: active exploration, quiet waking, whisker twitching, slow wave sleep, andparadoxicalREMsleep.Inthespectrogram(topshelf)graphs,Xaxisistime (withtheperiodofeachbehavioralstatemarked)andtheYaxisdepictsthefrequencyofoscillations.GrayscaleintheZaxisdepictsthemagnitudeorpowerofa par ic arfrequencyofoscillationforeachstate.(Adapted from D. Gervasoni, Shiht ul

Chieh L., S. Ribeiro, E. S. Soares, J. Pantoja, and M.A.L. Nicolelis, Global Forebrain Dynamics Predict Rat Behavioral States and Their Transitions.JournalofNeuroscience 24 [2004]: 11137 47.)

figure 11.2 Th eLochNessMonster,as depictedintheclassic fakephotograph,appears inthetopshelf.Itsbrain relativeisdepictedinthe state- paceofthelowerleft s shelf.Onthelowerright panel,thedifferentbrain statesthatcorrespondto differentlocationsinthe state- pacearemarkedby s ellipses.Atransitionstate betweenquietawakeand slow- avesleepisalso w

depicted.(Adapted from D. Gervasoni, Shih-Chieh L., S. Ribeiro, E. S. Soares, J. Pantoja, and M.A.L. Nicolelis, Global Forebrain Dynamics Predict Rat Behavioral States and Their Transitions.Journalof Neuroscience24 [2004]: 11137 47, with permission.)

eachoftheirmainbehavioralstates.TheXandYaxesofthegraphdepictthestate- spaceshowninFigure11.2,whilethegray- caledZaxisrepresentsthetimespent s by rats in each state. Notice that for most of their lives, our furry friends sleep, without dreaming, in the slow wave sleep state. Quiet awake is only the second mostcommonbrainstateinthelifeofrats.(Courtesy of Dr. Shih-Chieh Lin, National
Institute on Aging, NIH; and Dr. Damien Gervasoni, Claude Bernard University, Lyon, France.)

figure 11.3 ehypno-map.This3-Dgraphillustratesthetimespentbyratsin Th

Wire and Ball Model

Slow Wave Sleep

Active Exploration

Quiet Waking

Brain State-Space
figure 11.4 ewireandballmodel(seetextforexplanation). (Illustrated by Dr. Th Nathan Fitzsimmons, Duke University.)

and during the occurrence of a sensory stimulus (middle shelf). The refractory period,illustratedinthebottomshelf,definesthemaximumfiringratethataneuron can reach. Throughout these examples, the total firing rate produced by the w holebrainhastobemaintainedbelowamaximumcap.(Illustrated by Dr. Nathan
Fitzsimmons, Duke University.)

figure 11.5 atterns of single neuronal firing during normal firing (top shelf) P

figure 12.1 ross-modal pro essing in the primary somatosensory and visual C c cortices of rats. Peri- vent histograms display the isomodal and cross-modal e sensory- vokedresponsesofindividualS1andV1neurons.Intheleftpanel,the e traditionalvisuallyevokedresponsesofV1neuronsandtactile- vokedresponses e ofS1neuronsareshown.Intherightpanel,samplesofV1neuronsthatrespond toatactilestimulusandS1neuronsthatrespondtoavisualstimulusareplotted. KorbinianBrodmannwouldbeshocked! (Courtesy of Dr. Sidarta Robeiro, Interna-

tional Institute of Neuroscience of Natal, Brazil; and Dr. Damien Gervasoni, Claude Bernard University, Lyon, France.)

figure 13.1 orticalNeuroprosthesisforRestoringMotorFunctions.Adrawing C illustrateshowacorticalneuroprostheticdevicemayonedayhelppatientsparalyzedduetoaspinalcordlesion(seetextfordetails).(Adapted from M.A.L. Nicolelis,

Brain-Machine Interfaces to Restore Motor Function and Probe Neural Circuits.Nature ReviewsNeuroscience4 [2003]: 41722.)

figure 13.2 esign of the hole- ody exoskeleton to be utilized by the Walk D w b AgainProject.(Courtesy of Dr. Gordon Cheng, Technical University of Munich.)

stimulationofthespinalcord.Onthetopshelf,thestimulatingelectrodesandthe implantation approach to place them on the dorsal surface of the spinal cord. Middle shelf shows an implanted rat that exhibits signs of Parkinsons disease. Onthebottomshelftwocirclesareusedtoidentifyburstsoftheepilepticactivity observedinaspectrogramoftheratsbrainactivity,thatarecorrelatedwiththe akinesiaproducedbyParkinsonsdisease.At0time,theelectricalstimulationof the spinal cord started, using the implanted electrodes. Notice that the epileptic activity disappears and, as a result, the rat was able to walk freely again (not shown).ThespectrogramXaxisrepresentstime(0startofstimulation)andtheY axisrepresentsfrequencies.ThegrayscaleoftheZaxisrepresentspowerormagnitudeofbrainactivityatagivenfrequency(seescaleontheright).(Adapted from R.
Fuentes, P. Petersson, W. B. Siesser, M. G. Caron, and M.A.L. Nicolelis, Spinal Cord Stimulation Restores Locomotion in Animal Models of Parkinsons Disease.Science323 [2009]: 157882.)

figure 13.3 reatingaParkinsonsdisease- ikesyndromeinratsusingelectrical T l