* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Visual and Oculomotor Functions of Monkey Subthalamic Nucleus
Biological neuron model wikipedia , lookup
Multielectrode array wikipedia , lookup
Electrophysiology wikipedia , lookup
Convolutional neural network wikipedia , lookup
Brain–computer interface wikipedia , lookup
Single-unit recording wikipedia , lookup
Functional magnetic resonance imaging wikipedia , lookup
Visual selective attention in dementia wikipedia , lookup
Eyeblink conditioning wikipedia , lookup
Response priming wikipedia , lookup
Executive functions wikipedia , lookup
Caridoid escape reaction wikipedia , lookup
Molecular neuroscience wikipedia , lookup
Activity-dependent plasticity wikipedia , lookup
Neuroesthetics wikipedia , lookup
Stimulus (physiology) wikipedia , lookup
Neuroeconomics wikipedia , lookup
Neuroplasticity wikipedia , lookup
Mirror neuron wikipedia , lookup
Central pattern generator wikipedia , lookup
Development of the nervous system wikipedia , lookup
Clinical neurochemistry wikipedia , lookup
Neuroanatomy wikipedia , lookup
Neural coding wikipedia , lookup
Basal ganglia wikipedia , lookup
Metastability in the brain wikipedia , lookup
Circumventricular organs wikipedia , lookup
Neural oscillation wikipedia , lookup
Nervous system network models wikipedia , lookup
Process tracing wikipedia , lookup
Pre-Bötzinger complex wikipedia , lookup
Channelrhodopsin wikipedia , lookup
Optogenetics wikipedia , lookup
Neuropsychopharmacology wikipedia , lookup
Neural correlates of consciousness wikipedia , lookup
Synaptic gating wikipedia , lookup
Premovement neuronal activity wikipedia , lookup
JOURNALOF NEUROPHYSIOLOGY Vol. 67, No. 6, June 1992. Printed in U.S.A. Visual and Oculomotor Functions of Monkey Subthalamic Nucleus MASARU MATSUMURA, JUN KOJIMA, THOMAS W. GARDINER, AND OKIHIDE HIKOSAKA Laboratory of Neural Control, National Institute for Physiological Sciences, Okazaki, 444, and Department of Neurosurgery, Gunma University School of Medicine, Maebashi, 371, Japan SUMMARY AND CONCLUSION 1. Single-unit recordingswere obtained from the subthalamic nuclei of three monkeystrained to perform a seriesof visuooculomotor tasks.The monkeysweretrained to fixate on a spotof light on the screen(fixation task). When the spot wasturned off and a target spot came on, they were required to fixate on the target quickly by making a saccade.Visually guidedsaccades were elicited when the target cameon without a time gap(saccadetask). Memory-guided saccadeswere elicited by delivering a brief cue stimuluswhile the monkey wasfixating; after a delay, the fixation spotwasturned off and the monkey madea saccadeto the rememberedtarget (delayed saccadetask). 2. Of 265 neuronstested, 95 showedspike activity that was relatedto someaspectsof the visuooculomotor tasks,whereas66 neuronsrespondedto active or passivelimb or body movements. The task-relatedactivities wereclassifiedinto the following categories: eyefixation-related, saccade-related,visual stimulus-related, target- and reward-related,and lever release-related. 3. Activity relatedto eye fixation ( YI= 22) consistedof a sustained spikedischargethat occurredwhile the animal wasfixating on a target light during the tasks.The activity increasedafter the animal startedfixating on the target and abruptly ceasedwhenthe target went off. The activity wasunrelated to eyeposition. It was not elicited during eye fixation outsidethe tasks.The activity decreasedwhen the target spot wasremoved. 4. Activity related to saccades(~1= 22) consistedof a phasic increasein spikefrequency that wastime locked with a saccade madeduring the tasks.The greatestincreasesoccurred predominantly after saccadeonset.This activity usually wasunrelatedto spontaneous saccades madeoutsidethe task. The changesin activity typically were optimal in one direction, generally toward the contralateralside. 5. Visual responses (~2= 14) consistedof a phasicexcitation in responseto a visual probe stimulusor target. Responselatencies usuallywere 70-120 ms. The receptive fieldsgenerallywere centeredin the contralateralhemifield, sometimesextending into the ipsilateralfield. The receptive fieldsincluded the fovea1region in sevenneurons;most of theseneuronsrespondedbestto parafoveal stimulation. Peripheralstimuli sometimessuppressed the activity of visually responsiveneurons. 6. Activity related to target and reward ( YI= 29) consistedof sustainedspike dischargethat occurred only when the monkey could expect a reward by detectingthe dimming of the light spot that he was fixating. This activity was present regardlessof whetherthe monkey obtainedthe rewardby releasinga lever or by just maintaining fixation on the target. Eye fixation on a light stimulusmay not beprerequisite:sometimesit startedevenbefore the target point cameon. 7. Among neuronsexhibiting activity that wasrelatedto lever releaseduring the task ( YI= 49)) four different types of responses were recorded: excitation before lever release,excitation after lever release,inhibition before lever release,and a combination of excitation and inhibition. Theseneurons showedno changein activity in relation to limb movementsmadeoutsidethe tasks. 8. Neuronsrelatedto visuooculomotortaskswerelocated primarily in the ventral part of the subthalamicnucleus,whereas neuronsresponsiveto skeletomotormovementswere found predominantly in the dorsalpart. 9. Theseresultsare consideredin relation to the current view that the subthalamicnucleussendsexcitatory signalsto the substantia nigra parsreticulata, which in turn exertstonic inhibition of saccadicburst cellsin the superiorcolliculus. In contrastto the caudate-nigralconnection, which would play a role in the initiation of saccades by a disinhibitory mechanism,the subthalamic nucleusmight act to suppresseye movements.The suppressive mechanismwould act to maintain eye position on an object of interest,prevent unwantedeyemovementsunder specificcircumstances,or recover eyefixation once a saccadeis executed. INTRODUCTION The subthalamic nucleus (STN) is in a crucial position to influence the output of the basal ganglia. It projects to the internal and external segments of the globus pallidus (Carpenter et al. 198 1a; Groenewegen and Berendse 1990; Kita and Kitai 1987; Nauta and Cole 1978; Parent and Smith 1987; Smith et al. 1990) and the substantia nigra pars reticulata (SNr, Canteras et al. 1990; Carpenter et al. 1968, 198 lb; Kanazawa et al. 1976; Kita and Kitai 1987; Nauta and Cole 1978; Parent and Smith 1987; Smith et al. 1990), the major output structures of the basal ganglia system. It also projects to the pedunculopontine nucleus (Nauta and Cole 1978; Parent and Smith 1987; Smith et al. 1990). Inputs to STN are multiple. Signals arising from the caudate and putamen are conveyed to STN via the external segment of the globus pallidus (Kim et al. 1976; Kita et al. 1983; Moriizumi et al. 1987; Percheron et al. 1984). The cerebral cortex is another source of input. STN receives direct projections from several frontal cortical areas ( Afsharpour 1985; Hartmann-von Monakow et al. 1978; Kiinzle 1977, 1978; Kiinzle and Akert 1977; Petras 1969). Motor functions of STN have been suggested by both clinical and physiological studies. A lesion localized to STN produces hemiballismus (Carpenter 198 1). This has been confirmed experimentally by making lesions in STN in the monkey (Carpenter et al. 1950; Crossman et al. 1984; Whittier and Mettler 1949). Abnormal neural activity in STN may underlie different types of movement disorders (Crossman 1987; Mitchell et al. 1985a,b, 1989). Cells are found in STN that change discharge rates during skeletal movements and in response to somatosensory stimulation (DeLong et al. 1985; Georgopoulos et al. 1983). In addition to skeletomotor functions, the basal ganglia are thought to be involved in both oculomotor and cogni- 0022-3077192 $2.00 Copyright 0 1992 The American Physiological Society 1615 MATSUMURA 1616 tive functions (Hikosaka and Wurtz 1989; Wurtz and Hikosaka 1986 ) . Oculomotor functions are controlled in part by the caudate-nigro-collicular pathway. SNr exerts a tonic y-aminobutyric acid (GABA)ergic inhibition on the superior colliculus (Hikosaka and Wurtz 1983d, 1985). During saccadic eye movements, this inhibition appears to be suppressed transiently by GABAergic inhibition originating from the caudate nucleus (Hikosaka et al. 1989a-c) . Such a disinhibition occurs when the animal attempts to make a saccade volitionally to a visual or remembered target. Presaccadic activity has been shown to occur in caudate neurons and is believed to suppress the tonic spike discharge of SNr neurons, leading to a disinhibition of output neurons in the superior colliculus (Hikosaka et al. 1989a). In addition, cells that are related to eye fixation, visual, auditory, cognitive, and reward-related functions are found in the caudate nucleus as well as in SNr (Hikosaka and Wurtz 1983a-c; Hikosaka et al. 1989b,c). Although the responses of STN neurons have been studied during limb movements, their activity has not been studied in relation to visuooculomotor functions. Possible oculomotor functions for STN are suggested by its afferent connections from the cortical eye fields (Huerta and Kaas 1990; Huerta et al. 1986; Stanton et al. 1988) and its efferent connections to SNr (Nauta and Cole 1978; Parent and Smith 1987; Smith et al. 1990). Because STN appears to play an important role in motor control, and damage to STN has been implicated in many movement disorders, its potential role in oculomotor functions is important to establish. Therefore in the present study we have investigated the activities of STN neurons in relation to visuooculomotor behavior in alert monkeys. We have found that nearly 20% of the STN neurons examined were indeed related to visuomotor functions. The majority of these neurons were found in the ventral part of STN. A preliminary report of this work appeared elsewhere in abstract form (Matsumura et al. 1991a,b). METHODS We studiedsingle-cellactivities in STN of three alert monkeys (Macaca fuscata, adult male, Je-8.4 kg, Er-1 1.O kg, Iv-9.0 kg) while they were performing visuooculomotor tasks.In one monkey, both hemispheres were surveyed.In the other two monkeys, only the left sidewassurveyed.The monkeysweretrained to perform a seriesof behavioral paradigms(seebelow). The monkeys were kept in separateprimate cagesin an air-conditioned room. At each experimental session,usually twice a week, they were brought to the experiment room. The monkeys were given restricted fluid during periods of training and recording. Their healthconditions,suchasbody weightand appetite,werechecked daily. Supplementarywater and fruit wereprovided daily. Surgical procedures The monkeysweresedatedwith ketamine(4.6-6.0 mg/ kg) and xylazine ( 1.8-2.4 mg/ kg) given intramuscularly,and then general anesthesiawasinduced by intravenousinjection of pentobarbital sodium (4.5-6.0 mg kg-’ h-l). Surgical procedureswere conducted under asepticconditions in an operating room. After the skull wasexposed,holesfor lo- 15acrylic screwsweredrilled in it. The screwswere bolted into the holes and fixed with a dental acrylic resin.The screwsactedasanchorsby which a headholder l l ET AL. and a chamber,both madeof delrin, werefixed to the skull. Useof metal in the headpiecewasavoidedto permit magneticresonance imaging( seebelow). The chambersweretilted 40” laterally in the coronal planeand aimedat the center of STN [ 15mm anterior, 6 mm lateral, and 2 mm in height, basedon the atlasof M. fuscata (Kusama and Mabuchi, 1970)]. An eyecoil wasimplanted over oneeyefor measurementof eye position with the use of a method modified from Judge et al. ( 1980). The eyecoil consistedof Teflon-insulatedmultistranded stainlesssteel wire (Cooner Sales,0.3 mm OD) that was preformed into a 16-to l&mm diam coil with three completeturns. The coil wasleft in a pouch madeunder the conjunctiva -5 mm posterior to the lateral limbal margin. The twisted lead wasthen threadedthrough a hole madeon the lateral wall of the orbit. The twisted lead wasthreadedthrough a silicon tube (5 mm OD) to allow the eyeto rotate freely. The silicon tube with the leadinside wascoveredwith a dentalacrylic resinandfixed to a smallconnector mounted on the skull. The animalsreceivedantibiotics (sodium ampicillin, 23-30 mg/kg im eachday) after the operation. Behavioral tasks The monkey satin a primate chair in a dimly-lit and sound-attenuatedroom. The monkey’sheadwasfixed, and in front of him wasa tangent screen(57 cm from his face) onto which smallred spotsof light werebackprojected.The fixation point and the target point wereback-projectedspotsof light from light-emitting diodes (Toshiba, TLRA 150-C) and had a diameter of 0.2”. The projection systemconsistedof an LED light source,a pinhole aperture, and a projector lens.Three LED projectorswereused:the first one wasfor the fixation point, and its light wasprojecteddirectly onto the screen.The secondand the third oneswerefor the target point and another irrelevant stimulus,and their lightswerereflectedvia two galvanomirrorsthat controlled the horizontal and vertical positionsof the light spots. The basictask of the monkey wasto fixate hisgazeon a spotof light until it becamedim. If the spotsteppedfrom one location to another, the monkey was required to move his line of sight by making a saccadiceye movementto refixate the spot. To differentiatethe characteristicsof STN neurons,we usedtwo setsof paradigms.The first set required both eye fixation and manipulationof a lever (lever tasks); the secondsetrequired only eye fixation (eye tasks). Each setincluded different types of eye movementtasks. LEVER TASKS. Figure 1 illustratesthe paradigmsincluded in this category.Although different typesof eyemovementswereelicited, theseparadigmsall required manipulation of a lever. When the monkey depresseda lever on the chair, a spot of light (fixation point) cameon at the center of the screen.After a random period of time, the spotdimmed. If the monkey releasedthe lever within a shortperiod of time (0.5-0.7 s) after the dimming of the light, he wasrewardedwith a drop of water. If he releasedthe lever earlier or later, the trial terminatedwith neither reward nor punishment. The trial alsoterminated without reward when eye position deviated from the fixation point by more than a prefixed value (usually 3” in a horizontal or vertical direction) during the fixation period (except for the first 0.7 s,which wasallowedfor gazeto settledown). Successivetrials were separatedby an inter-trial interval of l-3 s (randomized). In the fixation task, the central fixation point (F) becamedim after 1.5-3.5 s with no other visual stimulusbeingpresented.In the fixation task with stimulus,a secondspotof light (S) appeared briefly (50 ms) during the fixation period ( 1.5-3.5 s). The monkey wasrequiredto maintain fixation without makinga saccadeto the stimulus.This task wasusedto study the visual responses of STN neurons.The saccadetask was designedto evoke visually guidedsaccades. The fixation point wasturned off after a random VISUOOCULOMOTOR FIXATION TASK FIXATION TASK WITH STIMULUS FSACCADE TASK DELAYED 3 SACCADE T TASK / ACTIVITY IN SUBTHALAMUS 1617 EXPERIMENTAL PROCEDURE. In most ofthe experiments,the location of the target point wasselectedfrom 32 points (8 directions with 5, 10, 20, and 30” eccentricity, Fig. 1, bottom). When spike activity of a singleneuron was isolated,we performed an initial screeningprocedure to test whether the neuron had any visuooculomotor properties. This consistedof the saccadetask and the delayedsaccadetask, both requiring the useof the lever; the targetpoint waschosenrandomly out of four directions(right, left, up, and down) with the eccentricity of 20”. If task-related activities were observed,we ran a seriesof experiment in blocks usingdifferent taskswhile recordingthe unitary spikeactivity and eyemovements.For eachblock of experiments(usually consisting of 40 trials), target pointswereusedthat hadthe sameeccentricity but different directions (8 or 4 directions). The target location alsocould be moved manually. This permitted usto determinethe visual receptive fieldsand movement fieldsof neurons. Recording procedures To determinethe location of STN, we obtainedmagneticresonanceimagesof the brain beforeunit recording(Hitachi Laboratory MRIS, 2.11 tesla). The magneticresonanceimagesprovided the approximatelocation of the STN in relation to the XYZ coordinate of the chamber. Glass-insulatedElgiloy microelectrodes (Suzuki andAzuma 1976, lo- 15MQ measuredat 10Hz, exposed tips 15-20 pm) were usedfor extracellular recordingof singlecell activities in STN. The electrodesweredriven by a hydraulic X- Y coordinate micromanipulator (MO-95, Narishige, Tokyo) that was attached to the chamber. In one hemisphere,we surveyed FIG. 1. Top: behavioral paradigms. F: fixation point (a central spot of STN widely to study the topography of the responses within STN. light that the monkey must fixate to start a trial). E: schematic eye posiIn other hemispheres, we implanted guidetubes(Teflon tube, 1.2 tion. S: stimulation point [a spot of light that was used to detect visual mm OD) one at a time for repeatedelectrodepenetration. Inserresponse (fixation task with stimulus)]. T: target point [a peripheral spot tion of the guide tube wasperformedunder anesthesiawith ketathat monkey must fixate after the central fixation point went off (saccade task)]. In the delayed saccade task, this spot was also turned on briefly as mine hydrochloride. Eye movementswere recordedwith the use the cue of a future target while the monkey was fixating. The depression at of the magneticsearch-coiltechnique ( Robinson 1963) . The behavioraltasksaswell asstorageand display of data were the end of the filled area represents dimming of the spot. Bottom : locations of the fixation point and target points (o ) . In most experiments, the target controlled by computer (PC 9801 RA2 1, NEC, Tokyo). Eyeposipoint was chosen from 32 points at 5, 10,20, and 30° from the center in 8 tions were digitized at 500 Hz and stored continuously during directions. The directions of the targets are represented by the numbers of each block of trials. The unitary action potentials were passed filled dots. In the raster displays of the following figures, filled dots are used through a window discriminator, and the times of their occurto indicate the direction of target in each trial. renceswere storedwith a resolution of 1 ms.At the time of each experiment, spike rastersand eye position data were displayed period of time ( 1S-2.5 s), and another spot of light [target point on-line on the computer monitor. (T)] cameon at the sametime. The target point dimmed after a The monkey’sbehavior wascontinuously monitored by an inrandom period of time (0.5-2.0 s). The monkey madea saccade frared camera. The camerasystem,with the use of mirrors, alto the target point and releasedthe lever in responseto its dim- lowed us to observebody movementsof the monkey in two perming. The delayed saccadetask wasdesignedto evoke memory- spectives:first, to seegrossmovements(hand/arm movements guided saccades.While the monkey was fixating, another light associatedwith manipulation of the lever, postural changes,and spot(cue stimulus)cameon briefly (50 ms), which indicated the foot /leg movements)and second,to seemovementsin the face location of a future target point. The monkey was required to (eye movements,eyeblinks, and oral movementsassociatedwith maintain fixation for 2-3 s, until the fixation point went off. The water intake). To examine somatosensoryand skeletomotorremonkey then wasrequiredto initiate a saccadeto the remembered sponses, body partsof the monkey weremanipulatedby the experlocation of the cue stimulusbeforean actualtarget point cameon, imenter. Eachmonkey wasconditioned to relax during thesema600 mslater. When the target cameon, the monkey then fixated nipulations. Spike discharge during these manipulations was on the target to detect its dimming and obtain reward. monitored with the oscilloscopeand audiomonitor. The responses EYE TASKS. The samesetof eyemovement taskswereincluded wereevaluatedby at leasttwo observers. in this category,which, however,did not require manipulation of a lever. The monkey wasrewardedwhen he maintained fixation Histology on a target spotuntil it becamedim. Appropriate fixation of gaze Severalmarking lesionswereplacedby passingpositive currents within the eyeposition window wasrequired (seeabove); the size of this window wasmadelarger for more eccentrictargets.When (2 PA for 200 s) through the recording electrodes.The marks eyepositionwasmaintainedwithin the fixation window, a reward madefor eachpenetration usually wereplacedat similar depths, wasgiven 300 ms after the dimming; failure to maintain fixation 2-3 mm aboveSTN. This produceda distinct pattern in the histoterminatedthe trial. When the task waschangedfrom a lever task logical sections.A few marks were alsomade at the siteswhere to an eye task, the monkey ceasedto usehis hand during several STN neuronswere recorded(Figs. 2 and 3). trials. Thus eye taskshelpedto differentiate visuooculomotor reAt the end of the experiments, each monkey was perfused sponses from limb movement-related responses. transcardially with saline followed by 4% paraformaldehyde, 1618 MATSUMURA ET AL. plotted alongthe identified tracks;their depthswerecalculatedby the distancesto the depth of the marks that were read on the electrodemanipulator at the time of experiment. Here we took into account the extent of tissueshrinkagedue to fixation, which wasestimatedto be 10%on the basisof the distancesbetweenthe marks. The location of the guide tube wasobvious in the histological sections.The depths of the recording siteswere determined by comparing the manipulator readoutsof the electrodesand the identified marking lesions. RESULTS FIG. 2. Photomicrograph of a histologtcal section showing the left STN of monkey Je, stained with cresyl violet. Arrow: an iron deposit made by passing positive currents through the recording electrode, showing the position where a saccade-related neuron was recorded. Calibration bar: 1 mm. under deeppentobarbital sodiumanesthesia.The headthen was fixed to the stereotaxicdevice,and the brain wascut in the coronal planeinto blocksof 12 mm thickness.The brain waspostfixed for 2 wk. Serialcoronal sections(40 pm) werecut on a cryotome and stainedwith cresylviolet. Reconstruction qf electrode penetrations The marking lesionswere visualizedon histologicalsectionsas round iron deposits.On the sidewithout any guidetube for electrodes,among33 penetrations,all but two of the electrodetracks could be identified. The recordingsitesof STN neuronsthus were A total of 265 neurons were recorded from STN in four hemispheres of three monkeys, their locations later being confirmed histologically. As shown in Table 1, 95 neurons responded during the task; 66 neurons responded during the animal’s active movements or in response to passive manipulation of the body. No responses were observed in the remaining 104 neurons. General characteristics of STN neurons Background discharge rate of neurons in STN ranged from 5 to 48 spikes/s (mean 18.47, SD 8.8 >and presented a unimodal distribution. Unit discharge patterns were usually irregular, sometimes occurring in doublets or triplets. No significant differences were found in background discharge rate between responsive cells and unresponsive cells. Visual responses were frequently found adjacent to STN in the zona incerta, where unit discharges usually were regular and tonic. As expected, visuomotor-related activity was also commonly found in SNr. Classification of task-related activities Neurons related to the behavioral tasks were further classified as fixation-related, saccade-related, visual-, targetand reward-related, and lever release-related. Note from Table 1 that individual neurons sometimes exhibited more than one type of activity. However, the neurons that preTABLE 1. ClussiJicationof activities qfsubthalamic neurons Types of Activities Task-related responses Saccade Visual Fixation Target and reward Lever-release Skeletomotor responses Orofacial Neck and trunk Soulder Arm and hand Leg Nonspecific Unresponsive All Tested FIG. 3. Location of the left STN and the surrounding structures at the same level shown in Fig. 2. This corresponds to the rostrocaudal level, A14.5, as illustrated in Fig. 4. CP, cerebral peduncle: GPi, internal segment of the globus pallidus: IC, internal capsule: OT, optic tract: STN, subthalamic nucleus: SNc, substantia nigra pars compacta: SNr, substantia nigra pars reticulata: ZI, zona incerta. Number of Units 22 14 22 29 49 21 16 8 12 4 5 104 265 Many neurons showed more than one type of response and thus are included in more than one category. Thus the number of responses exceeds the number of neurons tested. Skeletomotor responses included both activities related to the animal’s active movements and those related to passive movements. VLSUOOCULOMOTOR ACTIVITY IN SUBT ‘HALAMUS 1619 FIG. 4. Locations of neurons with visuooculomotor activities ( l ) and neurons with skeletomotor activities (o ) that were recorded from the left STN of monkey Je. Bars: locations of neurons that showed no response. They are plotted along the reconstructed electrode penetrations obtained from the coronal histological sections. The sections are arranged in caudaI-to-rostra1 sequence [ from A 13.5 to A 16.0, on the basis of the atlas of the brain of M. jimala (Kusama and Mabuchi 1970 )] , each separated by OS mm. The identified electrode tracks were projected on the nearest section. Symbols along the electrode tracks indicate the types of response: F, fixation; S, saccade; V, visual; T, target and reward. Neurons with 2 types of activity are indicated by conjugated symbols (e.g.. S, V). Calibration bar: 1 mm. sented activity related to visuomotor usually not accompanied by skeletomotor behaviour activity. were Topogru ph I1~ Figure 4 illustrates the locations of visuooculomotor neurons and other (skeletomotor responsive) neurons recorded in one hemisphere. The majority of visuooculomotor-related neurons were located in the ventral part of STN, and they tended to form clusters. No obvious segregation was found between visuooculomotor neurons that exhibited different types of responses during testing. In contrast, skeletomotor neurons tended to be located in the dorsal part of the nucleus. Although we found no evidence of a clear somatotopic organization in our sample, there was some tendency for orofacial responses to be found in the dorsolateral part of STN, trunk-related responses in the dorsal area, and limb-related responses in its central area. These responses were recorded from neurons that were in- 120 0 120 0 I I 11OOms 7LO7NOl,Ol/SAC 5. A left STN neuron with activity related to eye fixation ( lever task; saccade task). This neuron discharged tonically while the monkey was fixating on either the fixation point (F) or the target point (T). 7’011 to hottrm:durations of fixation point ( F) and of target point (T), horizontal eye positions [ H ( up: rightward, down: leftward )] , vertical eye positions [ V up: upward, down: downward], spike frequency histogram, and raster display of spike discharges. The records are aligned with respect to of&s of saccadcs made to the fixation point ( /P$), fixation point offset ( WIW), and target point offset (I?~@). Because the durations of these events were different across trials, the records are shown as interrupted. Each line in the raster display represents the activity of the neuron during a single trial, each dot indicating a single spike discharge. The target points were located at 20” from the fixation point in 4 directions. Their directions are indicated by the dots at the left of the raster line ( see Fig. 1 ). The raster lines were reordered with respect to the direction of the target point, though direction was selected randomly at the time of the experiment. Calibration for the histogram indicates a discharge rate of 40 imp/s, calibration for eye traces 20°, and time scale represents 100 ms. All subsequent figures use these conventions unless otherwise stated. FIG. I I I I I I I I I I ,I t II I t, 1 I I I t I I,, ItI I, II I I I t I I III II I I I II I 11’ I , : I l,,?ll I III I,, IIII I, tl 111 1 . III II ,III,Irn II 1 I II III I11 I I t III I? I l I ttt mt11tmt1 .I, I I 1.1 I I I I I , I , , . I,,. , I, I t 1 I ,I I I I III II I,1 11.1 I, I, tt 8 t I I .,t..l1.l.,“t.l’, III Ii l I I 1 t Irn1 I, I I I? il ,, I, I I t II I III I t I 1 I Ihl ImIt I II I I f I mm I ,t,,t,.t, ‘III III,, 1 - I Iomtml I I II I I t I, IHI t I11 .,I, 11. t,,,t:lt I I tt I I t I III t I I t t t II I I I’ 110 t II I I 111 I If111 1 III # II , l ltt I I1111 I II I irnitrl ‘8 II III I I I mm II I n a II II I III I I :I ‘.:I t’Xm ? :I I I , t ,,a& :I II LB tm Ital I t ?I I I:, II t I I II I,,, I II 111 I11 t lir I I 111mtt t I 81 I 7 ” II t I ? I Ill t’l” I t I I IIM tt I? t t ttml I l I’ ,??,i,?,l,:,?, I Ill I 120" 1I I I i I I I I t I I I I I I I lOOms I I I tmtIII.111 I II III I II 111 I I II 1 I Im I I I l , 111 I t I III I ttl I I I ?I t ,‘I’ I Ii I In? I I I II I tm ; I , ?I : , I I I11 I I I8 I c I I I II’ I11 ?a1 \l ! ‘I I I? : I?* I , 11, I I , I , , R,il It ,111 II1 I I I I I I 111 t~ttmtttttt II I I I 120" 111 II? II 111 I ?I ttt hI 111.n IIIIIIIIII 1111 Ii ?I my ml. l It I I I’ 1111 ’ #I II : :I t I I I,’ I i?t III I I I I?? I? III I I? ? l t I? I mm I I ttlmm’ II It I? I ;I’ IIO? n oam~mimI airno m Ii ?I I I Ii 111 H??rn II I t II Ii 11.11 it It 11111 II 1 tmm I I I I I I? I I ? l t I tu? ?I t Ii I I Y I ?ala? :‘~“I::‘: I I I II .‘I? I 1 I I I II I II I I I t ‘;A?lloy I II rnII :::i: I: II FIG. 6. Effects of fovea1 light stimulation on neural activity related to eye fixation. When the fixation point was turned off briefly (600 ms) while the monkey was fixating center panel, spike activity of this STN neuron decreased transiently (lever task; fixation task). Figures are aligned with respect to onset (lcrfi), blink-off (~~~ntcr), and offset ( rQ#) of the fixation point. vo 120” II t I 11 CONTRA IPSI 3!f!f zffrf I I 1 I I I I I t t t t fiOOms FTG. % A left STN neuron that discharged during saccades to both visual targets [ /~JJ)(eye task; saccade task)] and remembered targets [ httr1n7 (eye task; delayed saccade task)]. Trials were realigned and grouped into contralateral (CONTRA ) and ipsilatera1 ( IPSI ) ones. each with 3 directions ( horizontal. oblique up, and oblique down ). Contralateral saccades. but not ipsilateral saccades. were accompanied by increases in spike activity. Spike activity was greater for memory-guided saccades than for visually guided saccades (Y < 0.00 1. Mann-Whitney C’ test ). All the figures are aligned with respect to saccade onset. The eccentricity of the targets was W. H------/bt= ‘2oo V.-t! ! I I t 120” I I I I I 1 ) I t I t I I t t t 10Oms VISUOOCULOMOTOR ACTIVITY MEMORY-GUIDED VISUALLY-GUIDED - - 1621 SACCADES SACCADES -E -400 IN SUBTHALAMUS - 0 FIG. 8. Time course of saccade-related activities in STN neurons. The start and end of each line indicates the onset and offset respectively, of saccade-related activity for each neuron in relation to saccade onset. No statistical difference was observed between the times of the onsets or offsets of the neuronal responses during visually guided saccades(saccade task) vs. memory-guided saccades (delayed saccade task). Saccade-related activity usually occurred after the saccade onset. - 4OOmsec I -400 I I I 0 I termingled and appeared in clusters over the whole rostrocaudal extent of the nucleus. Visuooculomotor activities Twenty-twoneurons exhibited sustained spike activity during attentive fixation, either on the fixation point or target point. However, these changes in firing rate were not observed during spontaneous fixation outside the task. Examples of such responses are presented in Figs. 5 and 6. In most cases (Fig. 5), discharge rate increased gradually after the fixation point came on and the monkey made a saccade to it. This fixation-related activity was sometimes preceded by an anticipatory discharge. The activity could be interrupted if the ACTIVITYRELATEDTOEYEFIXATION. I I I 400msec visual target was removed from the central visual field. This was typically observed during the saccade task, when the fixation point was replaced with a peripheral target point (Fig. 5 ) ; 14 of 2 1 tested neurons showed a transient decrease of firing, and this occurred during the period when the monkey made a saccade to the peripheral target. The activity also ceased abruptly when the target went off at the termination of the trial. A similar transient decrease of firing was observed when the fixation point was blinked off for a short period while the monkey was fixating (Fig. 6). The latencies of the off responses ranged from 107 to 174 ms (mean 130.8 ms, SD 19.6 ms). In the seven other neurons, decreases in firing rate at the time of fixation off were unclear. ‘8 ‘0/- /-- ‘,88 FIG. 9. Direction selectivity of saccade-related activity. For each neuron, response indexes for 8 directions of saccades (eccentricity: 20 deg) are shown as a polar plot. Solid and hatched lines: response indexes for visually guided and memoryguided saccades, respectively. The polar plots are shown only for those neurons in which sufficient data were obtained for all of the 8 directions. The index was defined by the equation T/C; C and T: mean discharge rates within a control period and a test period, respectively. The control period was a 1,OOO-msperiod starting 1,000 ms before offset of the fixation point; the test period was set by eye to cover the total period of the responsive phase of saccadic activity. Hatched circle: response index of 1, indicating no change in spike activity; points outside and inside the circle indicate increase and decrease of spike activity, respectively. The side ipsilateral to the neuron is shown on the left and the contralateral side on the right. 16’3-- MATSUMURA ET AL. The fixation-related activity was unrelated to the position of the fixation target. As illustrated in Fig. 5, discharge rates were virtually the same whether the monkey was fixating at the central fixation point ( F) or one of the four peripheral target points (T) 20° eccentric to the center. None of the neurons recorded in STN showed significant relation to the position of the eye during fixation. Fixation-related neurons usually exhibited irregular changes in spike frequency during fixation. There was no indication, however, that the changes were related to the accuracy of fixation or to small fluctuations of eye position. RELATED TO SACCADES. Twenty-two neurons exhibited saccade-related spike activity. This activity was associated with saccades made to either visual or remembered targets during the task; it appeared not to be related to spontaneous saccades made outside the task. Among 18 neurons fullvd tested, saccade-related activity was greater for mem- ACTIVITY ory-guided saccades than for visually guided saccades for seven neurons, as shown in Fig. 7. In four neurons, it was greater for visually guided saccades; the responses of the other seven neurons were equally related to both types of saccades. The onsets of saccade-related neural activity usually ( 1 I/ 16 visually guided; 12/ 17 memory guided) preceded the onsets of the saccades (Fig. 8). This activity outlasted the saccades, however, usually by 100-400 ms, and the greatest changes in activity occurred predominantly after saccade onset. Saccade-related neurons had optimal saccade directions. In Fig. 9 the polar plot of saccadic response indexes for each neuron is shown. The majority of the cells exhibited greater responses during contralateral saccades. A decrease of activity was sometimes observed with ipsilateral saccades (e.g., neurons 2, 3, and 9). Ten of 22 saccade-related neurons also exhibited visual k 30 . I I,, t IL. . -ma . Ill -I ,I II mm I t t I .I I * . . III I 1 II, L I I II II1 I I I I I, I . I t I I I II I II IL 0 I II I I II I 1 II 1 t0mI 1 III I II I II FziY 11m1111 ’ III I I1 II I I II I 1 I III t I, I it1 II I I I 7R12N03. I II mm111 111lII III I I I I mmmml II il.11 11011~11 IDHI 0 III ,-mu l 00Bl0 -111 II- ’ 1 I I t II FIG. 10. Combination of visual and saccade-related activities. G/J: in the saccade task (lever task), this right STN neuron showed a burst of spike activity in response to the onset of the target point to which the monkey made a saccade. The location of target point was 3” from center ( imct ). The h istogram and raster arc aligned with respect to the onsot of the target point. U~NWK in the delayed saccade task (eye task ), the neuron exhibited both a visual response ( first burst) and a saccade-related response ( second burst). The sum of these responses, however. was smaller than the activity seen during the saccadc task. The histogram and raster are aligned with respect to the onset of the target cue ( /L$) and onset of saccade ( righf ). 111m11, t I W/SAC I I I I I : I I !- j 4 ; I I 4 I looms VISUOOCULOMOTOR ACTIVITY 1623 IN SUBTHALAMUS A IPSI n=14 --- 30 n=t4 20 Latencies of visual msec responses FIG. 1 f . Visual receptive fields (A ) and latencies of visual responses (B) of STN neurons. In A, the numbers along the abscissa indicate degrees from the central fixation point. Circular areas: field in which visual stimuli evoked responses from each neuron. responses. The neuron in Fig. 10, for example, showed a burst of spikes before a small upward saccade to a visual target (top). The delayed saccade task (bottom) indicated that this was a combination of visual and saccadic responses. The combined presaccadic activity (top) was greater than the sum of the individual responses (bottom). Such enhancement rons. Vis ml rc3pomc~~s Fourteen neurons responded to visual stimuli presented while the monkey was actively fixating the central fixation E20 / / I \ \ + ‘5D32N07.12/SACD was seen in three of nine tested neu- E30/,$,, -1 ’ \ / \ / I 5D32N07 :*/SACD FIG. 12. Visual response of a left STN neuron. LX/~,: while the monkey was fixating, a spot of light presented at 3” to the right produced a phasic visual response ( lever task; delayed saccade task ). Figures are aligned on the onset of target cue. The duration of the visual stimulus was 50 ms. Right: polar plots of visual response indexes for eccentricities ranging from 3 (E3) to 30 deg (E30). Response index was calculated for each of 8 stimulus directions. The index was defined as the ratio of the mean discharge rate within a test period to the mean discharge rate within a control period (see Fig. 9 ). The control period was a 500-ms period before stimulus onset; the test period was from 80 to 180 msafter stimulus onset, which was set by eye to cover the total period of the excitatory visual response. The increase in spike frequency was limited to a small region in the upper right quadrant near the center. Spike activity was suppressed by more peripheral stimuli; this suppression was nearly complete for stimuli IO0 (see HO) from the center. 1624 MATSUMURA point. As presented in Fig. 1 1 A, the receptive fields of these neurons were centered predominantly contralaterally to the side of the recording site. These receptive fields sometimes extended into the ipsilateral side as well. The receptive field had a gradient, with the response magnitude reducing gradually as the location of the stimulus was shifted away from the center of the receptive field. Seven out of 14 neurons had receptive fields that included the central fovea1 region; these neurons typically were more responsive to parafoveal stimuli. The latencies of visual responses ranged from 70 to 118 ms (mean 98.0, SD 18.1, Fig. 11 B). In nine neurons, firing rate decreased when a visual stimulus was presented outside its excitatory receptive field, either on the side contralateral to the excitatory receptive field (n = 6) or in the region surrounding the excitatory receptive field (n = 3). An example of the latter type (surround inhibition type) is shown in Fig. 12. This neuron increased its activity in response to contralateral perifoveal stimuli (I[$), but its activity was strongly suppressed by more peripheral stimuli in all directions (see ElO). This suppression was even greater for stimuli presented in the same direction as the excitatory receptive field (see E30). In two cells, the activity evoked by a visual stimulus persisted until the monkey made a saccade to the remembered location of the visual stimulus (Fig. 13). This sustained activity was direction selective, corresponding to the receptive field of the initial visual response; no significant activity was evoked for stimuli that elicited memory-guided saccades in other directions. ilT I ET AL. RELATEDTOTARGETANDREWARD. Twenty-nine neurons presented activity that was related to the monkey’s fixation on the target point, which was followed by the delivery of reward; these neurons were not activated while the animal was fixating on stimuli that were not directly associated with reward. We have categorized this type of response separately from fixation-related activity (described above) because of its clear dependence on reward. Examples of these responses are shown in Figs. 14 and 15. The neurons became active only when the monkey began to fixate on a target spot that was expected to dim. The activity was present irrespective of the location of this fixation point. When the task was changed such that the central spot became dim (Figure 14, hottom), after several trials the neuron became active while the monkey was fixating on it. The same result was obtained in all of the four neurons tested. This further indicates that the reward expectancy attached to the spot of light was critical for these neuronal responses. The reward-related activity was observed even when the eye tasks were performed without manipulating the lever (unlike the neuron shown in Fig. 17), suggesting that these neurons were not concerned with actual type of movement employed for obtaining reward. We monitored the monkey’s behavior during this period with the use of an infrared camera (see METHODS) but found no evidence of consistent body movements, including postural changes, licking movements, and other orofacial movements; rather, the monkey usually withheld gross movements. ACTIVITY I H V FIG. 13. this activity remembered absent when raster. The A right STN neuron that showed sustained activity after visual stimuli. In the delayed saccade task (lever task), was triggered by a target cue presented on the contralateral side and continued until a saccade was made to the location of the stimulus (straight left, as indicated with 3 dots on the left side of the raster). The activity was saccades to other directions were required. The directions are also indicated by the arrows on the right side of the records arc aligned on the onset of the target cue (I&) and the onset of saccades ( @II). VISUC)OCULOMOTOR ACTIVITY IN SUBTHALAMUS 7RlON02.07/ESAC FIG. 14. A right STN neuron showing activity related to target and reward. In the saccadc task [eye task ( try) ] - the neuron was activated while the monkey was fixating on the target point (T), but not on the central fixation point (F). No directional selectivity was observed. In the fixation task [eye task (hr~om )], in which the monkey was rewarded by fixating the central point (F), the same neuron became active during the fixation. Note that, in these experiments. the monkey was rewarded by maintaining fixation on the spot of light: no hand movement was required. Figures are aligned on the onset of the fixation point ( /I$), the of&et of fixation point (c~~cv), and the offset of the target point (right). Some neurons with fixation-related activity (2/22), or with the target- and reward-related activity (4/29), showed phasic increases in discharge rate just after the fixation point came on ( Fig. 15). To determine whether this activity represented a visual response with a receptive field, we examined eye positions at the time of fixation point onset. We found, however, no correlation between the retinal position of the fixation point and the magnitude of the phasic increase in firing rate. ACTIVITY RELATED TO LEVER RELEASE. Forty-nine neurons changed their activity in relation to the lever release. Neurons that had sensorimotor responses were excluded from this group. Both phasic responses ( Fig. 16) and preparatory responses (Fig. 17) were found. Within the phasic response category were four types: excitation before lever release (Fig. 16A, n = 14), excitation after lever release (Fig. 16& n = 1 I ), inhibition before lever release (Fig. 16C, n = 13), and a combination of excitation and inhibition (Fig. 16D, n = 7) . Figure 17 illustrates the response of neurons exhibiting preparatory activity (n = 4). This activity started > 1 s before lever release (Fig. 17, top). We found no evidence of consistent movements in any body parts during this period. These changes in activity were absent, however, during the eye task (Fig. 17, hotto~n) in which the monkey fixated the same target spots but did not release the lever. This result suggests that the activity was specific for the hand movements per se rather than being related to reward expectancy, unlike the neurons classified above as target- and reward-related ( see above). AC‘TIVITY RELATED -I’0 ANIMAL’S ACTIVE MOVEMENTS OR PASSIVE MANIPIJLATlONS OF THE BODY PART. A summary TO of responses related to active movements or passive manipulation is shown in Table 1. Effective stimuli were passive manipulation of joints and pressure on the muscles. Light touch and air-blow never evoked responses. The somatosensory responses usually had wide receptive fields, although these were confined to the contralateral side. Because our primary concern was with the visuomotor activity of these 1626 MATSUMURA ET AL. FIG. 15. Complex nature of the activity related to target and reward. Sustained activity during fixation on the target point was present during both the saccade task [lever task (top)] and the delayed saccade task [lever task (hoitmn)]. In addition, this neuron discharged phasically after onset of the central fixation point. It also showed spike discharges before the appearance of the target point, when the monkey was presumably expecting it. Figures are aligned on onset of fixation point ( /L$), onset of target point (caner), and offset of target point (righi). neurons, we did not examine the characteristics neurons in detail. of these DISCUSSION Nearly 20% of the neurons recorded from the STN in the present study became active during eye fixation or saccadic eye movements or in response to visual stimuli. The results suggest that STN participates in the control of visuooculomotor behavior in cooperation with other basal ganglia nuclei such as the caudate nucleus and SNr. Topographic organization of STN Neurons related to visuooculomotor behaviors were found primarily in the ventral part of STN, whereas neurons related to skeletomotor behaviors usually were located in the dorsal part. This pattern of responses is consistent with the differential fiber connections of the dorsal and ventral STN with the cerebral cortex and basal ganglia nuclei. The somatomotor cortical areas (areas 4 and 6) project in a somatotopic manner to the laterodorsal part of STN, with face area located laterally and leg area medially. In contrast, the ventral part of STN receives projections from the frontal association cortices (areas 8 and 9, Hartman-von Monakow et al. 1978) as well as from the frontal eye field (Huerta et al. 1986; Stanton et al. 1988) and supplementary eye field (Huerta and Kaas 1990). STN is known to project to several basal ganglia nuclei. The globus pallidus (internal and external segments) and SNr are its two major recipient areas. Efferents from STN seem to exhibit a ventrodorsal topography. The ventral part of STN projects to substantia nigra and the caudate nucleus, whereas the dorsal part projects to the internal segment of the globus pallidus and putamen (Nakano et al. 1990; Parent and Smith 1987 ). A mediolateral topography has also been suggested in the subthalamonigral connection (Smith et al. 1990). The subthalamonigral fibers terminate mostly within the pars reticulata (Carpenter et al. 198 1b). Cells related to visual and/or mnemonic-oculomotor behaviors are found in both the SNr and the caudate nucleus (Hikosaka and Wurtz 1983a-d; Hikosaka et al. 1989a-c). In the following sections we will consider how each of the VISUOOCIJLOMOTOR ACTIVITY l . . .’ . . . l . l . . l . l . 1627 IN SLJBTHALAMUS . m . l . m . . . - l . . l . l . l . . . . I .: l . . . . , . . l .*. .m. . . . l l l l * .* l ’ : . . . l . . . . I* l l : . . . 0’ l . 7rO9nOl.011sacd 7rOlnOl.Ol/sac C D .*a . . l . . l .* l n .* l . l , :. . . . . . . .: . . l ..m* **I*. l . . . l ..* l . . . . l . . . .**. .* . . l . . . .**I- l * , . 1 1. . : .: . .a . . . . . l l . m. . . . . : . l . l . .: .-mm. . . : . : . l :.*-.: . :*‘.. .:, . l l .:.::, I . . . . .: .‘.a i l ‘. . 7r03nO2.Oflsacd FIG. 16. STN neurons showing different types of activity related to lever release. Figures are aligned on the time of lever release. Increased activity before lever release (A : delayed saccade task), after lever release ( II: saccade task ) , decreased activity before lever release ( C’: delayed saccade task ), and a combination of excitation and inhibition at lever release ( II: saccade task ) . visuooculomotor signals might be reorganized through the afferent and efferent connections or integrated of the STN. A significant number of the neurons studied in STN showed sustained spike discharges during active eye fixation on task-specific targets. This activity appeared to be unrelated to eye position. It usually was not elicited during eye fixations made outside the tasks. The activity decreased when the target spot subsequently was removed. Although the origin of this activity is uncertain, both the caudate nucleus and frontal cortical areas may be two likely candidates by virtue of their projections to STN. Many neurons in the caudate nucleus show sustained spike activity while the monkey is fixating to obtain a reward (Hikosaka et al. 1989~). The activity is unrelated to eye position and is unrelated to spontaneous eye fixation, as in STN. Thus the sustained excitatory signals from the caudate nucleus could be transmitted to STN through serial inhibitory connections via the external segment of the globus pallidus, producing a similar sustained activity in STN. Whether the external pallidal neurons exhibit responses related to eye fixation remains to be examined. Different parts of the prefrontal association cortex contain neurons that show sustained increases or decreases during eye fixation on task-specific light targets. Suzuki et al. ( 1979) were the first to characterize the types of eye fixation-related neurons in the inferior dorsolateral prefrontal cortex. Some neurons were shown to be dependent on the presence of visible targets, whereas others maintained the spike activity even when the target was turned of?’ but the monkey was still fixating on the position of the target. Joseph and Barone ( 1987) found that neurons in the area superior to the principal sulcus showed sustained activity during eye fixation, but in a manner highly dependent on the behavioral contexts. These prefrontal cortical neurons thus share some features in common with STN neurons. In contrast, the possibility that the direct inputs from the parietal cortex to STN cause the sustained STN activity is less likely, because the activitv of parietal fixation neurons is clearly related to eye posit&, unlike STN neurons (Andersen and Mountcastlc 1983; Lynch et al. 1977: Mountcastle et al. 1975; Sakata et al. 1980). The eye fixation-related signals in STN mav be convevcd to SNr. Although initially considered to be inhibitory (beLong and Georgopoulos 198 1; Hammond et al. 1983b: Larsen and Sutin 1978; Perkins and Stone 1980), STN efferent connections are now thought to be at least partlv excitatory and may use glutamate as a neurotransmitte; (Hammond et al. 1978, 1983a; Kitai and Kita 1987: Na- I628 MATSUMURA ET AL. 120” c V- \ m I 2o” L 1 III III I III I I I I I I I I I t I I I I I I I I I I I I I IlOOms g4 7RlON01,03/SACD 120 “I I I I I I I I l l l l FIG. 17. Activity related to preparation for lever release. Top: activity increased after onset of the target point and ceased 200 ms before lever release (lever task; delayed saccade task). Rasters are aligned on onset of target point (1&) and time of lever-release (rig&). Bottom: activity disappeared in the eye task in which monkey was required neither to press nor to release the lever (eye task: delayed saccade task). Rasters are aligned on onset (/<fi) and offset (C&Y) of target point. 0 1OOms 7RlONOl.O5/ESACD kanishi et al. 1987; Robledo and Fi3ger 1990; Rouzaire-Dubois et al. 1983, 1984; Smith and Parent 1988). The excitatory signals would be used to enhance the tonic spike activity of nigral cells, thus help maintaining eye fixation. Although most visuooculomotor neurons in SNr show a clear-cut decrease in firing during eye fixation, some neurons show a sustained, albeit less clear-cut, increase in firing (unpublished observations). Smmdic activity I in STN A group of neurons in STN showed an increase of spike activity that was time-locked to task-specific saccades. The changes in activity were unrelated to spontaneous saccades made outside the task. One third of the neurons were active predominantly during saccades made to visual targets, and another third were active predominantly during saccades to remembered targets. A similar feature was found in the caudate nucleus (Hikosaka et al. 1989a). For more than half of the STN neurons, the onset of the saccadic activity preceded saccade onset, but not to a great extent. These results suggest that saccadic information in STN could play a role in the initiation of saccades. It seems unlikely, however, that initiation is a major function of STN neurons, because a dominant portion of their activity occurs after a saccade. The saccadic activity in STN could be derived directly from the frontal cortex or indirectly from the caudate nucleus. The frontal eye field (Huerta et al. 1986; Stanton et al. 1988) and supplementary eye field (Huerta and Kaas 1990) are the two primary candidate cortical sites. Neurons that have presaccadic and postsaccadic activities are found in both areas. Neurons in these two cortical eye fields become active before purposive saccades to visual or remembered targets (Bruce and Goldbern 1985; Schlag and VISUOOCULOMOTOR ACTIVITY Schlag-Rey 1987 ) . These signals may reach STN, although the outputs from STN may occur too late to play a role in saccade initiation. Postsaccadic activity in these cortical areas could also contribute to the STN activity. This appears less likely, however, because postsaccadic activity in these regions of cortex is usually less selective, is sometimes omnidirectional, and may occur during spontaneous saccades, unlike neurons in STN. The caudate nucleus contains many neurons that fire before task-specific voluntary saccades (Hikosaka et al. 1989a). This signal could be transmitted to STN through the external segment of the globus pallidus (Groenewegen and Berendse 1990; Kita et al. 1983; Ohye et al. 1976; Parent 1986). The conditional nature of STN saccadic activity appears to be very similar to the responses of caudate neurons, including those responses contingent on visually guided saccades and those contingent on memory-guided saccades. If the saccadic signals present in STN are conveyed to SNr as discussed above, we would expect a phasic increase of SNr cell activity. However, the saccadic activity in SNr is usually characterized by decreases in firing rate (Hikosaka and Wurtz 1983a). There is no easy way to interpret the discrepancy. Perhaps the summation of influences from the STN with stronger inhibitory influences from the striatum always results in a net inhibition of activity in the SNr. This explanation would imply that input from the STN served primarily to alter the excitability of SNr neurons but usually did not result in clear increase in firing rate of most SNr neurons. Of course, it also remains possible that the saccade-related activity of STN neurons may not be transmitted directly to the SNr, or that perhaps these particular projections are of an inhibitory nature. The suppressive role of the STN in saccades may be supported by the paucity of saccadic eye movements in parkinsonian patients (Hikosaka and Wurtz 1989) and MPTPtreated primates (Usui et al. 1990), presumably both associated with increased STN activity (Bergman et al. 1990). In contrast, the lesion of the STN would be expected to produce an opposite effect-hyperkinesia of the eye. However, oculomotor symptoms have been reported only in a limited number of cases (Bedwell 1960; Martin 1927). In our preliminary experiments, injection of muscimol (a GABA agonist) into the monkey STN induced a strong eye deviation to the contralateral side in addition to ballistic movements in the contralateral body parts. Whether this is due to the blockade of the STN excitatory outflow remains to be determined. Visual response Visually responsive neurons appear to be always present in brain areas containing eye movement-related neurons (Bruce and Goldberg 1985; Bushnell et al. 198 1; Hikosaka et al. 1989b; Hikosaka and Wurtz 1983a; Mohler et al. 1973; Robinson et al. 1978, 1986; Schlag and Schlag-Rey 1984, 1987; Wurtz and Mohler 1976), and STN is no exception to this rule. Receptive fields of visually responsive neurons in STN were usually centered in the contralateral hemifield, sometimes extending into the ipsilateral field. This feature is similar to visual neurons in the caudate nucleus (Hikosaka et al. 1989b), frontal eye field (Bruce and IN SUBTHALAMUS 1629 Goldberg 1985 ) , supplementary eye field ( Schlag and Schlag-Rey 1987 ), and intralaminar nuclei of the thalamus (Schlag and Schlag-Rey 1984). As discussed for eye fixation-related activity and saccadic activity (see above), visual signals may also be directly derived from the cortical eye fields or indirectly from the caudate nucleus. Latencies of visual responses are usually 70- 120 ms in STN (present study), 60- 100 ms in the frontal eye field (Goldberg and Bushnell 198 1 ), and loo-250 ms in the caudate nucleus (Hikosaka et al. 1989b). This would seem to preclude the caudate nucleus as a major site of origin of STN visual responses. Visual signals could instead reach the caudate by the reverse connection (Nakano et al. 1990). Many of the STN visual neurons responded best to parafovea1 stimulation, and these changes in firing rate were almost always increases in activity. In contrast, peripheral stimuli, either contralateral or ipsilateral, could suppress these neurons (e.g., Fig. 12). This effect requires wholefield integration of visual information, as postulated for neurons in a prestriate visual area ( V4, Desimone et al. 1985 ) . This type of STN neuron could act to enhance SNr cell activity if a visual stimulus was present close to the line of sight, but would reduce the facilitatory effect if the stimulus was present in the periphery. In other words, such STN neurons would decrease or increase the probability of saccades, depending on the position of the saccade-triggering stimuli within the visual field. Activity related to target and reward Target- and reward-related activity occurred only when the monkey could expect to gain a reward by detecting the dimming of the light spot that he was fixating. This type of neuron showed no change in activity while the monkey was fixating the central fixation point to start a trial if the fixation point itself was not associated with reward. Eye fixation on a light stimulus may not be prerequisite: sometimes it started even before the target point came on (Fig. 15 ). Similar responses were commonly seen in the caudate nucleus (Hikosaka et al. 1989~). The sustained spike discharge of caudate neurons was present regardless of whether the monkey obtained reward by releasing a lever, by just maintaining fixation on the target, or by making an appropriate saccade, all features that were also observed for STN neurons (Figs. 14 and 15). Thus both STN and caudate neurons appear to carry information related to expectation of reward. In the dorsomedial part of the prefrontal cortex, Rizzolatti et al. ( 1990) found neurons that showed sustained spike activity only when the monkey fixated on an object that he was ready to reach and grasp. No change in activity was seen when the monkey fixated an object but refused to reach it (e.g., a syringe). Such reward contingency appears to be similar to that observed for STN neurons, although the paradigms used to study the neurons were different. Furthermore, the prefrontal activity was unrelated to eye position, as seen in STN. It is uncertain whether this type of activity has a motor role. It could be related to expectation of reward, contributing to the sequence of neural events that are not directly coupled with m otor outputs. Its motor role cannot be denied, however. It may help the eyes focused on a visual 1630 MATSUMURA object by suppressing saccades. This would be the same effect expected for eye fixation-related activity, but more emphasis here would be placed on a visual object that was directly associated with reward. In addition, this target- and reward-related activity might contribute to suppression of skeletomotor behaviors, especially because the monkey usually withheld gross movements during this period. Function of the subthalamic nucleus We will first consider the possible outcomes of the neural activities in STN, given the presumed efferent connections, and then put forward a hypothesis for the function of STN in voluntary, purposive behaviors. Any argument about the functions of the subthalamic nucleus critically depends on whether its neurons are excitatory or inhibitory. On the assumption that STN neurons are excitatory, as most recent studies suggest (Hammond et al. 1978, 1983a; Kitai and Kita 1987; Nakanishi et al. 1987; Robledo and Feger 1990; Rouzaire-Dubois et al. 1983, 1984; Smith and Parent 1988), activation of STN neurons should lead to an increase of basal ganglia efferent activity, thus enhancing the inhibition of target structures such as the thalamus, superior colliculus, and pedunculopontine nucleus. The behavioral outcome of neural activity in STN should therefore be an inhibition of a motor outputs, at least for eye movements. Similar proposals have been made with respect to skeletal movements on the basis of lesions of STN, which produce hemiballismus, jerky involuntary movements in contralateral body parts (Bergman et al. 1990; Carpenter et al. 1950; Carpenter 198 1; DeLong and Georgopoulos 198 1) . STN-mediated suppression of motor outputs may be necessary for appropriate execution of purposive behaviors. Suppose a purposive behavior consists of a set of motor acts organized in a particular spatiotemporal manner. The central motor systems would set up motor programs for these motor acts, but allow only a portion of them to be activated at a time while suppressing the others. Triggering a movement would be carried out by the removal of the suppression in a selective and sequential manner. The subthalamic nucleus might act as an interface for these suppressive mechanisms. The mechanisms could be activated directly by the frontal cerebral cortices through corticosubthalamic connections, or indirectly through striato external pallidal subthalamic connections. The suppression should be specific, highly contingent on the behavioral contexts, and could be predictive on the basis of the motor plan. Interestingly, the basal ganglia also possess an opposing mechanism: striato-nigral/ internal pallidal connections, which would facilitate the target motor areas by disinhibition (Chevalier and Deniau 1990; Hikosaka and Wurtz 1989; Mink and Thach 199 1). The disinhibitory mechanism would be at work for the initiation of movements, as seen for presaccadic neurons in the caudate nucleus (Hikosaka et al. 1989a) and SNr (Hikosaka and Wurtz 1983a,b). The subthalamic suppressive mechanism would act to maintain eye position on an object of interest or recover eye fixation once a saccade is executed (this study). Note that the above argument is based on the assumption that STN neurons transmit signals by increasing their activ- ET AL. ity. In fact, we occasionally observed decreases in activity, which would lead to an opposite effect. Furthermore, the possibility still remains that neurons in the STN are heterogeneous in synaptic action and that some of them may exert inhibitory effects on their target neurons. We express our deep appreciation to Prof. Chihiro Ohye for discussing the earlier results of these experiments. We thank M. Nakanishi and Dr. H. Tokuno for their help with histology, L. Bertrand, M. Togawa, and 0. Nagata for their help in preparation of figures, Dr. M. Kato for designing the computer programs, and Drs. P. Bedard, M. Filion, and A. Parent for reviewing the manuscript. Thomas W. Gardiner was on leave from the University of Tennessee, Memphis, TN. Address for reprint requests: 0. Hikosaka, National Institute for Physiological Sciences, 38 Nishigonaka, Myodaijcho Okazaki, 444, Japan. Received 8 July 199 1; accepted in final form 17 January 1992. REFERENCES AFSHARPOUR, S. Topographical projections of the cerebral cortex to the subthalamic nucleus. J. Camp. Neural. 236: 14-28, 1985. ANDERSEN, R. A. AND MOUNTCASTLE, V. B. The influence of the angle of gaze upon the excitability of the light-sensitive neurons of the posterior parietal cortex. J. Neurosci. 3: 532-548, 1983. BEDWELL, S. F. Some observations on hemiballismus. NeuroZogy 10: 6 19622, 1960. BERGMAN, H., WICHMANN, T., AND DELONG, M.R.Reversalofexperimental Parkinsonism by lesions of the subthalamic nucleus. Science Wash. DC 249: 1436- 1438, 1990. BRUCE, C. J. AND GOLDBERG, M. E. Primate frontal eye fields. I. Single neurons discharging before saccades. J. Neurophysiol. 53: 603-635, 1985. BUSHNELL, M.C., GOLDBERG, M. E., AND ROBINSON, D. L. Behavioral enhancement of visual responses in monkey cerebral cortex. I. Modulation in posterior parietal cortex related to selective visual attention. J. Neurophysiol. 46: 755-772, 198 1. CANTERAS,N.S.,SHAMMAH-LAGNADO,~. J., SILVA, B. A., ANDRICARDO, J. A. Afferent connections of the subthalamic nucleus: a combined retrograde and antereograde horseradish peroxidase study in the rat. Brain Res. 5 13: 43-59, 1990. CARPENTER, M. B. Anatomy of the corpus striatum and brain stem integrating systems.In: Handbook of Physiology. The Nervous System. Motor Control. Bethesda, MD: Am. Physiol. Sot., 198 1, sect. 1, vol. II, part 2, chapt. 19, p. 947-995. CARPENTER, M.B., BATTON, R.R., CARLETON,S.C.,ANDKELLER, J.T. Interconnections and organization of pallidal and subthalamic neurons in the monkey. J. Comp. Neural. 197: 579-603, 198 la. CARPENTER,M.B., CARLETON,S.C.,KELLER, J.T., ANDCONTE, P.Connections of the subthalamic nucleus in the monkey. Brain Res. 224: l-29, 1981b. CARPENTER, M.B., FRASER,R. A. R., AND SHRIVER, J.E.Theorganization of the pallidosubthalamic fibers in the monkey. Brain Res. 11: 522-559, 1968. CARPENTER, M. B., WHITTIER, J. R., AND METTLER, F. A. Analysis of choreoid hyperkinesia in the rhesus monkey. Surgical and pharmacological analysis of hyperkinesia resulting from lesions of the subthalamic nucleus of Luys. J. Comp. Neural. 92: 293-33 1, 1950. CHEVALIER, G. AND DENIAU, J. M. Disinhibition as a basic process in the expression of striatal functions. Trends Neurosci. 13: 277-280, 1990. CROSSMAN, A. R. Primate models of dyskinesia: the experimental approach to the study of basal ganglia-related involuntary movement disorders. Neuroscience 2 1: l-40, 1987. CROSSMAN, A. R., SAMBROOK, M. A., AND JACKSON, A. Experimental hemichorea/ hemiballismus in the monkey. Brain 107: 579-596, 1984. DELONG, M.R., CRUTCHER, M.D., ANDGEORGOPOUPOS, A.P.Primate globus pallidus and subthalamic nucleus: functional organization. J. Neurophysiol. 53: 530-543, 1985. DELoNG,M. R. ANDGEORGOPOULOS, A.P.Motorfunctionsofthebasal ganglia. In: Handbook of Physiology. The Nervous System. A4otor Control. Bethesda, MD: Am. Physiol. Sot., 198 1, sect. 1, vol. II, part 2, chapt. 21, p. 1017-1061. DESIMONE, R., SCHEIN,~. J., MORAN, J., ANDUNGERLEIDER, L.G.Contour, color and shape analysis beyond the striate cortex. Vision Res. 25: 441-452, 1985. VISUOOCULOMOTOR ACTIVITY A. P., DELONG, M. R., AND CRUTCHER, M. D. Relations between parameters of step-tracking movements and single cell discharge in the globus pallidus and subthalamic nucleus of the behaving monkey. J. Neurosci. 3: 1586-1598, 1983. GOLDBERG, M. E. AND BUSHNELL, M. C. Behavioral enhancement of visual responses in monkey cerebral cortex. II. Modulation in frontal eye fields specifically related to saccades. J. Neurophysiol. 46: 773-787, 1981. GROENEWEGEN, H. J. AND BERENDSE, H. W. Connections of the subthalamic nucleus with ventral striatopallidal parts of the basal ganglia in the rat. J. Comp. Neural. 294: 607-622, 1990. HAMMOND, C., DENIAU, J. M., RIZK, A., AND F~GER, J. Electrophysiological demonstration of an excitatory subthalamonigral pathway in the rat. Brain Res. 15 1: 235-244, 1978. HAMMOND, C., ROUZAIRE-DUBOIS, B., F~GER, J., JACKSON, A., AND CROSSMAN, A. R. Anatomical and electrophysiological studies on the reciprocal projections between the subthalamic nucleus and nucleus tegmenti pedunculopontinus in the rat. Neuroscience 9: 4 l-52, 1983a. HAMMOND, C., SHIBAZAKI, T., AND ROUZAIRE-DUBOIS, B. Branched output neurons of the rat subthalamic nucleus: electrophysiological study of the synaptic effects on identified cells in the two main target nuclei, the entopeduncular nucleus and the substantia nigra. Neuroscience 9: 5 ll520, 1983b. HARTMANN-VON MONAKOW, K., AKERT, K., AND K~~NZLE, H. Projections of the precentral motor cortex and other cortical areas of the frontal lobe to the subthalamic nucleus in the monkey. Exp. Brain Res. 33: 395-403, 1978. HIKOSAKA, O., SAKAMOTO, M., AND USUI, S. Functional properties of monkey caudate neurons. I. Activities related to saccadic eye movements. J. Neurophysiol. 6 1: 780-798, 1989a. HIKOSAKA, O., SAKAMOTO, M., AND USUI, S. Functional properties of monkey caudate neurons. II. Visual and auditory responses. J. Neurophysiol. 6 1: 799-8 13, 1989b. HIKOSAKA, O., SAKAMOTO, M., AND USUI, S. Functional properties of monkey caudate neurons. III. Activities related to expectation of target and reward. J. Neurophysiol. 6 1: 8 14-832, 1989~. HIKOSAKA, 0. AND WURTZ, R. H. Visual and oculomotor functions of monkey substantia nigra pars reticulata. I. Relation of visual and auditory responses to saccades. J. Neurophysiol. 49: 1230-1253, 1983a. HIKOSAKA, 0. AND WURTZ, R. H. Visual and oculomotor functions of monkey substantia nigra pars reticulata. II. Visual responses related to fixation of gaze. J. Neurophysiol. 49: 1254-1267, 1983b. HIKOSAKA, 0. AND WURTZ, R. H. Visual and oculomotor functions of monkey substantia nigra pars reticulata. III. Memory-contingent visual and saccade responses. J. Neurophysiol. 49: 1268-1284, 1983~. HIKOSAKA, 0. AND WURTZ, R. H. Visual and oculomotor functions of monkey substantia nigra pars reticulata. IV. Relation of substantia nigra to superior colliculus. J. Neurophysiol. 49: 1285- 130 1, 1983d. HIKOSAKA, 0. AND WURTZ, R. H. Modification of saccadic eye movements by GABA-related substances. I. Effect of muscimol and bicuculline in the monkey superior colliculus. J. Neurophysiol. 53: 266-29 1, 1985. HIKOSAKA, 0. AND WURTZ, R. H. The basal ganglia. In: The NeurobioZogy ofsaccadic Eye Movements, Reviews in Oculomotor Research, edited by R. H. Wurtz and M. E. Goldberg. Amsterdam: Elsevier, 1989, vol. 2, p. 257-281. HUERTA, M. F. AND KAAS, J. H. Supplementary eye field as defined by intracortical microstimulation: connections in macaques. J. Comp. Neural. 293: 299-330, 1990. HUERTA, M. F., KRUBITZER, L. A., AND KAAS, J. H. Frontal eye field as defined by intracortical microstimulation in squirrel monkeys, owl monkeys, and macaque monkeys. I. Subcortical connections. J. Comp. Neural. 253: 415-439, 1986. JOSEPH, J. P. AND BARONE, P. Prefrontal unit activity during a delayed oculomotor task in the monkey. Exp. Brain Res. 67: 460-468, 1987. JUDGE, S. J., RICHMOND, B. J., AND CHU, F. C. Implantation of magnetic search coils for measurement of eye position: an improved method. Vision Res. 20: 535-538, 1980. KANAZAWA, I., MARSHALL, G. R., AND KELLY, J. S. Afferents to the rat substantia nigra studied with horseradish peroxidase, with special reference to fibers from the subthalamic nucleus. Brain Res. 115: 485-49 1, 1976. KIM, R., NAKANO, K., JAYARAMAN, A., AND CARPENTER, M. B. Projections of the globus pallidus and adjacent structures: an autoradiographic study in the monkey. J. Comp. Neurol. 169: 263-289, 1976. GEORGOPOULOS, IN SUBTHALAMUS 1631 H., CHANG, H. T., AND KITAI, S. T. Pallidal inputs to subthalamus: intracellular analysis. Brain Res. 264: 255-265, 1983. KITA, H. AND KITAI, S. T. Efferent projections of the subthalamic nucleus in the rat: light and electron microscopic analysis with the PHA-L method. J. Comp. Neural. 260: 435-452, 1987. KITAI, S. T. AND KITA, H. Anatomy and physiology of the subthalamic nucleus: a driving force of the basal ganglia. In: The Basal Ganglia II, edited by M. B. Carpenter and A. Jayaraman. New York: Plenum 1987, p. 357-373. K~~NZLE, H. Projections from the primary somatosensory cortex to basal ganglia and thalamus in the monkey. Exp. Brain Res. 30: 48 l-492, 1977. K~~NZLE, H. An autoradiographic analysis of the efferent connections from premotor and adjacent prefrontal regions (area 6 and 9) in Macaca fascicularis. Brain Behav. Evol. 15: 185-234, 1978. K~~NZLE, H. AND AKERT, K. Efferent connections of cortical area 8 (frontal eye field) in Macaca fascicularis. A reinvestigation using the autoradiographic technique. J. Comp. Neural. 173: 147-164, 1977. KUSAMA, T. AND MABUCHI, M. Stereotaxic Atlas ofthe Brain of Macaca fuscata. Tokyo: Univ. of Tokyo Press, 1970. LARSEN, K. D. AND SUTIN, J. Output organization of the feline entopeduncular and subthalamic nuclei. Brain Res. 157: 2 l-3 1, 1978. LYNCH, J. C., MOUNTCASTLE, V. B., TALBOT, W. H., AND YIN, T. C. Parietal lobe mechanisms for directed attention. J. Neurophysiol. 40: 362-389, 1977. MARTIN, J. P. Hemichorea resulting from a local lesion of the brain. The syndrome of the body of Luys. Brain 50: 637-65 1, 1927. MATSUMURA, M., KOJIMA, J., GARDINER, T. W., AND HIKOSAKA, 0. Oculomotor activities in the monkey subthalamic nucleus. Sot. Neurosci. Abstr. 17: 1219, 199 la. MATSUMURA, M., KOJIMA, J., GARDINER, T. W., AND HIKOSAKA, 0. Visual and oculomotor function of the subthalamic nucleus in monkeys. Neurosci. Res. Suppl. 14: 78, 199 1b. MINK, J. W. AND THACH, W. T. Basal ganglia motor control. II. Late pallidal timing relative to movement onset and inconsistent pallidal coding of movement parameters. J. Neurophysiol. 65: 30 l-329, 199 1. MITCHELL, I. J., JACKSON, A., SAMBROOK, M. A., AND CROSSMAN, A. R. Common neural mechanisms in experimental chorea and hemiballismus in the monkey. Evidence from 2-deoxyglucose autoradiography. Brain Res. 339: 346-350, 1985a. MITCHELL, I. J., JACKSON, A., SAMBROOK, M. A., AND CROSSMAN, A. R. The role of the subthalamic nucleus in experimental chorea. Evidence from 2-deoxyglucose matabolic mapping and horseradish peroxidase tracing studies. Brain 112: 1533- 1548, 1989. MITCHELL, I. J., SAMBROOK, M. A., AND CROSSMAN, A. R. Subcortical changes in the regional uptake of [ 3H] -2-deoxyglucose in the brain of the monkey during experimental choreiform dyskinesia elicited by injection of a gamma-aminobutyric acid antagonist into the subthalamic nucleus. Brain 108: 405-422, 1985b. MOHLER, C. W., GOLDBERG, M. E., AND WURTZ, R. H. Visual receptive fields of frontal eye field neurons. Brain Res. 6 1: 385-389, 1973. MORIIZUMI, T., NAKAMURA, Y., KITAO, Y., KITAO, Y., AND KUDO, M. Ultrastructural analyses of afferent terminals in the subthalamic nucleus of the cat with a combined degeneration and horseradish peroxidase tracing method. J. Comp. Neural. 265: 159- 174, 1987. MOUNTCASTLE, V. B., LYNCH, J. C., GEORGOPOULOS, A., SAKATA, H., AND ACUNA, C. Posterior parietal association cortex of the monkey: command functions for operations within extrapersonal space. J. Neurophysiol. 38: 871-908, 1975. NAKANISHI, H., KITA, H., AND KITAI, S. T. Intracellular study of rat substantia nigra pars reticulata neurons in an in vitro slice preparation: electrical membrane properties and response characteristics to subthalamic stimulation. Brain Res. 437: 45-55, 1987. NAKANO, K., HASEGAWA, Y., TOKUSHIGE, A., NAKAGAWA, S., KAYAHARA, T., AND MIZUNO, N. Topographical projections from the thalamus, subthalamic nucleus and pedunculopontine tegmental nucleus to the striatum in the Japanese monkey, Macaca fuscata. Brain Res. 537: 54-68, 1990. NAUTA, H. J. W. AND COLE, M. Efferent projections of the subthalamic nucleus: an autoradiographic study in monkey and cat. J. Comp. Neural. 180: 1-16, 1978. OHYE, C., LE GUYADER, C., AND F~GER, J. Responses of subthalamic and pallidal neurons to striatal stimulation: an extracellular study on awake monkeys. Brain Res. 111: 241-252, 1976. PARENT, A. Comparative Neurobiology of the Basal Ganglia. New York: Wiley, 1986. KITA, 1632 MATSUMURA A. AND SMITH, Y. Organization of efferent projections of the subthalamic nucleus in the squirrel monkey as revealed by retrograde labeling methods. Brain Res. 436: 296-3 10, 1987. PERCHERON, G., YELNIK, J., AND FRANCOIS, C. A Golgi analysis of the primate globus pallidus. III. Spatial organization of the striato-pallidal complex. J. Comp. Neurol. 227: 2 14-227, 1984. PERKINS, M. N. AND STONE, T. W. Subthalamic projections to the globus pallidus: an electrophysiological study in the rat. Exp. Neurol. 68: 500511, 1980. PETRAS, J. M. Some efferent connections of the motor and somatosensory cortex of simian primates and felid, canid and procyonid carnivores. Ann. NY Acad. Sci. 167: 469-505, 1969. RIZZOLATTI, G., GENTILUCCI, M., CAMARDA, R., GALLESE, V., LUPPINO, G., MATELLI, M., AND FOGASSI, L. Neurons related to reaching-grasping arm movements in the rostra1 part of area 6 (area 6arB). Exp. Brain Res. 82: 337-350, 1990. ROBINSON, D. A. A method of measuring eye movement using a scleral search coil in a magnetic field. IEEE Trans. Biomed. Eng. 10: 137- 145, 1963. ROBINSON, D. L., GOLDBERG, M. E., AND STANTON, G. B. Parietal association cortex in the primate: sensory mechanisms and behavioral modulations. J. Neurophysiol. 41: 910-932, 1978. ROBINSON, D. L., PETERSON, S. E., AND KEYS, W. Saccade-related and visual activities in the pulvinar nuclei of the behaving rhesus monkey. Exp. Brain Res. 62: 625-634, 1986. ROBLEDO, P. AND F~GER, J. Excitatory influence of rat subthalamic nucleus to substantia nigra pars reticulata and the pallidal complex: electrophysiological data. Brain Res. 5 18: 47-54, 1990. ROUZAIRE-DUBOIS, B., HAMMOND, C., YELNIK, J., AND F~GER, J. Anatomy and neurophysiology of the subthalamic efferent neurons. In: The Basal Ganglia, edited by J. S. Mckenzie, R. E. Kemm, and L. N. Wilcock. New York: Plenum, 1984, p. 205-234. ROUZAIRE-DUBOIS, B., SCARNATTI, E., HAMMOND, C., CROSSMAN, A. R., AND SHIBAZAIU, T. Microiontophoretic studies on the nature of the neurotransmitter in the subthalamo-entopeduncular pathway of the rat. Brain Res. 27 1: 1l-20, 1983. SAKATA, H., SHIBUTANI, H., AND KAWANO, K. Spatial properties of visual fixation neurons in posterior parietal association cortex of the monkey. J. Neurophysiol. 43: 1654-1672, 1980. PARENT, ET AL. J. AND SCHLAG-REY, M. Visuomotor functions of central thalamus in monkey. II. Unit activity related to visual events, targeting, and fixation. J. Neurophysiol. 5 1: 1175-l 195, 1984. SCHLAG, J. AND SCHLAG-REY, M. Evidence for a supplementary eye field. J. Neurophysiol. 57: 179-200, 1987. SMITH, Y., HAZRATI, L.-N., AND PARENT, A. Efferent projections of the subthalamic nucleus in the squirrel monkey as studied by the PHA-L anterograde tracing method. J. Camp. Neurol. 294: 306-323, 1990. SMITH, Y. AND PARENT, A. Neurons of the subthalamic nucleus in primates display glutamate but not GABA immunoreactivity. Brain Res. 453: 353-356, 1988. STANTON, G. B., GOLDBERG, M. E., AND BRUCE, C. J. Frontal eye field efferents in the macaque monkey. I. Subcortical pathways and topography of striatal and thalamic terminal fields. J. Comp. Neurol. 27 1: 473492, 1988. SUZUKI, H. AND AZUMA, M. A glass insulated “Elgiloy” microelectrode for recording unit activity in chronic monkey experiments. Electroencephalogr. Clin. Neurophysiol. 4 1: 93-95, 1976. SUZUKI, H., A~UMA, M., AND YUMIYA, H. Stimulus and behavioral factors contributing to the activation of monkey prefrontal neurons during gazing. Jpn. J. Physiol. 29: 47 l-489, 1979. USUI, S., KATO, M., KORI, A., MATSUMURA, M., MIYASHITA, N., AND HIKOSAKA, 0. Deficits in spontaneous eye movements induced by unilateral infusion of MPTP in the monkey caudate nucleus. Sot. Neurosci. Abstr. 16: 235, 1990. WHITTIER, J. R. AND METTLER, F. A. Studies on the subthalamus of the rhesus monkey. II. Hyperkinesia and other physiologic effects of subthalamic lesions, with special reference to the subthalamic nucleus of Luys. J. Comp. Neurol. 90: 3 19-372, 1949. WURTZ, R. H. AND HIKOSAKA, 0. Role of the basal ganglia in the initiation of saccadic eye movements. In: Progress in Brain Research, The Oculomotor and Skeletalmotor Systems: Dtflerences and Similarities, edited by H.-J. Freund, U. Battner, B. Cohen, and J. Noth, Amsterdam: Elsevier, 1986, vol. 64, p. 175- 190. WURTZ, R. H. AND MOHLER, C. W. Enhancement of visual responses in monkey striate cortex and frontal eye fields. J. Neurophysiol. 39: 766772, 1976. SCHLAG,