Adobe Acrobat 4.0 User Guide - CIHR Group in Sensory
Contents
1. 574 MOTOR SYSTEMS B CO QD ad rightward saccade FIGURE 40 2 Brain systems for the generation of saccades and fixation A Outline of the pathway from the visual input through cerebral cortex to oculomotor centers controlling e muscles The schematic drawing shows a lateral view of a monkey brain The afferent visual pathway is indicated by the dotted line arrows passing from the retina to the lateral genicu late nucleus LGN of the thalamus to the striate cortex V7 and then by a series of steps not shown to the parietal cortex the lateral intraparietal LIP area and the frontal cortex the frontal eye field FEF The projections from these areas to the superior colliculus SC are indicated by the solid line ar rows Omitted from the drawing are several areas including the supplementary eye field in frontal cortex reciprocal con nections between areas and the indirect pathway through the basal ganglia caudate and substantia nigra pars reticulata The output from the colliculus to the midbrain and pons ocu lomotor centers and then to the eye muscles also are indicated by dotted lines B The schematic drawing of the organization of the SC on the right and left sides of the brain Small sac cades are represented in the rostral SC 0 and large saccades are represented in the caudal SC 40 The activity before a saccade is across a population of neurons which is represented by a mound in the SC on the side2. THE NEW COGNITIVE NEUROSCIENCES Second Edition Michael S Gazzaniga Editor in Chief 0 262 07195 9 A Bradford Book The MIT Press Cambridge Massachusetts London England 40 The Superior Colliculus and the Cognitive Control of Movement ROBERT H WURTZ MICHELE A BASSO MARTIN PARE AND MARC A SOMMER ABSTRACT Normal vision consists of an alternation of sac adic eye movements and periods of visual fixation This is a relatively simple alternation of movement and nonmovement but the control of this alternation involves many of the cogni tive issues underlying the performance of more complex ac tions The superior colliculus SC is a nexus in the brain system controlling these saccades and fixations and recent ex periments have shown how this structure and its cortical inputs contribute to their control Activity on the SC movement map shows that the amplitude and direction of the impending sac cade is most likely to be represented by the vector average across a large population of neurons The control of the alter nation of movement and nonmovement which in the case of the saccadic system is the alternation of saccades and fixation involves the balance of competition between SC neurons toni cally active during visual fixation fixation neurons and those active long before saccades buildup neurons The prepara tion to move may be represented by the long lead or delay ac tivity of buildup neurons beca
3. J Comp Neurol 230 55 76 FUCHS A F 1967 Periodic eye tracking in the monkey J Physiol Lond 193 161 171 586 MOTOR SYSTEMS Fucus A F and D A ROBINSON 1966 A method for mea suring horizontal and vertical eye movement chronically in the monkey J Appl Physiol 21 1068 1070 GUMCHER P W and D L SPARKS 1992 Movement selec tion in advance of action in the superior colliculus Nature 355 542 545 Herr K V HENN T Vilas and B COHEN 1989 Brainstem regions related to saccade generation In The Neurobiology of Saccadic Eye Movements Reviews of Oculomotor Research Vol Il R H Wurtz and M E Goldberg eds Amsterdam Elsevier pp 105 212 HIKOSAKA O and R H WURTZ 1983 Visual and oculomo tor functions of monkey substantia nigra pars reticulata IV Relation of substantia nigra to superior colliculus J Neuro physiol 49 1285 1301 HUERTA M F L A KRUBITZER and J H KAAS 1986 Fron tal eye field as defined by intracortical microstimulation in squirrel monkeys owl monkeys and macaque monkeys Subcortical connections J Comp Neurol 253 415 439 Juper S J B J RICHMOND and F C CHU 1980 Implanta tion of magnetic search coils for measurement of eye posi tion An improved method Vision Res 20 535 538 KRAUZLIS R J M A BASSO and R H WURTZ 1997 Shared motor error for multiple eye movements Science 276 1603 1695 KusTov A A and D L ROBINSON 1996 Shar
4. K E HORN 1994 New roanatomy of saccadic omnipause neurons in nucleus raphe interpositus In Contemporary Ocular Motor and Vestib ular Research A Tribute to David A Robinson A F Fuchs T Brandt U B ttner and D Zee eds Stuttgart Thieme pp 488 495 COLBY C L J R DUHAMEL and M E GOLDBERG 1995 Oculocentric spatial representation in parietal cortex Cereb Cortex 5 470 481 CoLsy C L JR DUHAMEL and M E GOLDBERG 1996 Visual presaccadic and cognitive activation of single new rons in monkey lateral intraparietal area J Neurophysiol 76 2841 2852 DORRIS M C M PARE and D P MUNOZ 1997 Neuronal activity in monkey superior colliculus related to the initia tion of saccadic eye movements J Neurosei 17 8566 8579 EDELMAN J A and E L KELLER 1996 Activity of visuomo tor burst neurons in the superior colliculus i express saccades J Neurophysiol 76 908 926 EVARTS E V 1966 Methods for recording activity of indi vidual neurons in moving animals In Methods in Medical Re search R F Rushmer ed Chicago Year Book pp 241 250 FISCHER B and R BOCH 1983 Saccadic eye movements af ter extremely short reaction times in the monkey Brain Res 260 21 26 FISCHER B and H WEBER 1993 Express saccades and visual attention Bekan Brain Sci 16 553 567 Fries W 1984 Cortical projections to the superior colliculus in the macaque monkey A le study using horserad ish peroxidase
5. When the target dimmed the activity increased This neuron had a burst of action potentials associated with the onset of the saccade that did not differ sub stantially between the probability conditions PARIETAL CORTEX INPUTS TO SUPERIOR COLLICU LUS The lateral intraparietal LIP area is the area in the inferior parietal lobule where saccade related activ ity has been described by both the Andersen and Gold berg groups Andersen and Gnadt 1989 Andersen Essick and Siegel 1987 Barash et al 1991 Colby Du hamel and Goldberg 1995 Colby Duhamel and Goldberg 1996 Mazzoni et al 1996 Pare and Wurtz 1997 first identified the LIP neurons that can be acti vated antidromically from electrical stimulation within the SC The projection neurons act primarily on the in termediate layers of the SC because antidromic activa tion thresholds reached minimum values at depths where neurons showed saccade related activity The LIP projection to the SC appears to be organized topograph ically because the most efficient antidromic stimulation typically was obtained at SC sites with movement fields similar to those of the LIP neurons Approximately 75 of the identified LIP efferent neu rons responded to the onset of the visual target and ap proximately 60 of the neurons maintained their activity during the delay period of either the visual or memory delayed saccade tasks Several LIP neurons also exhibited a modest saccade relat
6. an increase in activity on one part of the SC map and this activity develops over time during the delay pe riod We believe that what we refer to loosely as prepa ration to move might best be regarded as the gradual specification over time of one part of the SC map as the center of activity for the next saccade This localized ac tivity specifies the vector for the next saccade and the narrowing of activity over time is a narrowing of the se lection of this vector Cortical input to superior colliculus Many of the fixation buildup and burst neurons are likely to receive input from the cerebral cortex This opens the possibility of seeing the transition in activity between the cerebral cortex and the superior collicu lus To explore this transition the cortical neurons that project to the SC must be identified It is not sufficient to compare the activity of any neuron in the cortical areas that project to the SC with the SC activity be cause many of the cortical neurons might not project to the SC We need to study specifically those neurons that project to the SC and have used antidromic stimu lation to do so In this technique a cortical neuron is identified as projecting to the SC if electrical stimula tion of the SC produces spikes in the cortical neuron with short consistent latencies along with other crite ria Lemon 1984 We have concentrated on the neu rons in the parietal and frontal cortex because neurons
7. are more medial and downward are more lateral NEURON TPES The activity of all the neurons within the intermediate layers is not identical however and to systematically study them different types have been identified The classification we use Munoz and Wurtz 19950 is based largely on the activity of these neurons that comes after any initial visual response to the sac cade target but before any burst of activity preceding saccade onset This intervening activity is best re vealed in two behavioral paradigms the visual and memory delayed saccade tasks In these delay tasks there is a period of active fixation during which we measure the resting discharge rate of the neuron Then a visual target is presented in the center of the neu ron s movement field but the monkey is required to delay making a saccade to it the delay period until the fixation stimulus is extinguished the cue to move In the visual task the target remains until the end of each trial whereas in the memory task itis only flashed and the monkey has to make a saccade to the remembered location Burst neurons Munoz and Wurtz 1995a are the sac cade related neurons that have been most extensively studied in the SC and those that also have been referred to as saccade related burst neurons Sparks 1978 In the delayed saccade tasks these neurons frequently display a brief response time locked to the onset ofthe visual stim ulus but their salient response i
8. in the system through which much although not all of the information from the cor tex flows to the brainstem oculomotor structures the su perior colliculus SC The SC receives inputs from the parietal and frontal areas of the cerebral cortex known to be related to saccade generation both directly and through the basal ganglia Fries 1984 Hikosaka and Wurtz 1983 figure 40 2A and projects directly to the pontine and midbrain oculomotor areas The SC can be viewed as a location for the coordination of the varied inputs from the forebrain and as the location for the transformation of these into outputs for the control of movement Itis the last step in the system in which both a visual and a motor map clearly are evident Robinson 1972 Schiller and Stryker 1972 Wurtz and Goldberg 1972 In this chapter we concentrate on several recent de velopments of our understanding of the saccadic fixa tion system and its relation to the control of movement We first describe how the neuronal elements in the SC are organized including how only a few neuron types can be organized across a map to provide different out put signals Then we consider how the SC contributes to the alternating pattern of saccades and fixations Finally we consider the delay activity between the neuronal response to the stimulus and the burst before a saccade and consider the transition leading to this activity in the cortical neurons that project to the SC
9. is controlled by a system within the brain that includes the SC but that the observations just described show how the neuronal elements are organized to carry out this interaction The basic observations on the effects of activating or inactivating the rostral SC still hold Sac cades are suppressed by the activation of the SC fixa tion region by either electrical stimulation or injection of the chemical agent bicuculine a gamma aminobu tyric acid GABA antagonist Munoz and Wurtz 1993b Conversely saccade production is facilitated by the inactivation of the fixation neurons by injection of the GABA agonist muscimol Munoz and Wurtz 1993 Both observations show that alteration of the rostral SC affects when saccades are generated WURTZ BASSO PARE AND SOMMER SUPERIOR COLLICULUS 577 FIXATION BUILDUP NEURONS ma a NEURONS SMALL LARGE ERROR ERROR FIGURE 40 5 Mutual inhibition of fixation and buildup new rons The arrows indicate inhibitory interactions The fixation neurons in the rostral SC inhibit the saccade related neurons in the caudal SC whereas the saccade related neurons inhibit both other saccade neurons and the fixation neurons Not shown are the inhibitory inputs from the basal ganglia or the presumed excitatory inputs from cerebral cortex This leads to a hypothesis of the interaction between fixation and saccades shown schematically in figure 40 5 The basic assumption is that when any popula tion of SC neuro
10. on an intact SC Schiller Sandell and Maunsell 1987 These exper iments illustrate most clearly the interactions we have considered One of the most prominent ideas about the shift in fix ation caused by a saccade which usually is addressed with respect to the accompanying shift of attention is that there must be an initial disengagement from the current fixation stimulus This disengagement also was proposed to be critical in the reduction of saccadic reac tion time in the gap saccade paradigm and the genera tion of express saccades Fischer and Weber 1993 the latencies of which 80 ms approach the conduction time from the retina to the eye muscles Fischer and Boch 1983 With the discovery of the SC fixation neu rons a neural correlate for the fixation disengagement was obvious the disengagement occurs when the fixa tion neurons reduce their discharge after the disappear ance of the fixation stimulus or pause before saccades The increased incidence of express saccades following rostral SC inactivation Munoz and Wurtz 1991 pro vided experimental support for this hypothesis as did the finding that a decrease in fixation activity during the gap task correlates with the reduction in saccade reac tion time Dorris Par and Munoz 1997 Recording of the activity of SC burst neurons re vealed that the saccade related burst of activity is indis tinguishable from the target related responses of these neuron
11. 5a Second they both have movement fields most fixation neurons like buildup neurons were shown to have a burst of activity with small contralaterally directed saccades Munoz and Wurtz 19958 One possibility is that the fixation neurons have delay activity just as buildup neurons do and that the in creased activity during fixation essentially is this delay activity To investigate this Krauzlis and associates 1997 recorded from neurons in the rostral SC that had the characteristics of fixation neurons including contin ued discharge when the monkey fixated in the absence Too 8 amp 550 E gor Es 54321012345 ipsi Target Steps deg contra A 100 3 ge kog 0 TA 43 2 4 012 3 4 5 B ipsi Target Steps deg FIGURE 40 4 Fixation neuron activity with small changes in target position A Changes in firing rate of a neuron after small steps of the fixation target into the visual field either ipsilateral or contralateral from the SC in which the neuron was recorded Symbols show the mean firing rate over an interval beginning 100 ms after the target step and lasting for 100 ms or until 8 ms before the saccade onset Error bars indicate 1 SD for the 12 trials The dotted line indicates the percentage of trials on which each target step elicited a saccade Dashed line indicates the mean firing rate with no target steps B Change in activity of four fixation neurons from one monkey after the sa
12. First both the fixation and buildup neurons increase their activity for a target at some eccentricity in the contralateral visual field the fixation neurons for very small target steps and the buildup neurons for larger steps Second the delay activity is present even if no saccade is made as indicated for the fixation neurons by Krauzlis and associates 1997 and for the buildup neurons by Munoz and Wurtz 1995a Thus the fixation neurons might be regarded most parsimoniously as a rostral continuation of the buildup neurons The hypothesis that emerges is that the activity of these fixation and buildup neurons indicates an error between where the eye is and where the target is and the size of the error represented by the neuron de pends on the location of the neuron on the SC movement map This simplification eliminates the problem of decid ing where in the rostral SC the fixation cells end and where buildup neurons begin The different effects of ac vating these neurons depends on their interactions within the SC and probably their differential connections outside the SC as we consider in the next section FIXATION AND SACCADE INTERACTION How does this proposed continuity of fixation and buildup neu rons affect the original hypothesis Munoz and Wurtz 1993a 1993b that the activity of monkey fixation neu rons suppresses the generation of saccades We believe that the interaction between saccades and fixation still
13. accade task in which the fixation stimulus FP is extinguished 200 ms before the target 7 presentation as indicated by the bars along the bot tom Rasters and spike density functions are aligned on target onset In the bottom panel the spike density functions of the express solid line and the regular dashed line trials are su perimposed A The activity of burst neurons displayed two bursts before regular saccades one time locked to the target onset and a later one time locked to the saccade onset In con trast express saccades were preceded by only one robust Because this delay period activity is not tied to the occurrence of the saccade Munoz and Wurtz 1995a proposed that it represents the preparation to make a saccade Consistent with this motor preparation hy pothesis Dorris and colleagues 1997 demonstrated that the level of delay activity for a significant propor tion of buildup neurons 41 predicts motor perfor mance They showed that the level of buildup activity during a gap period between the time the fixation 580 MOTOR SYSTEMS burst time locked to both target and saccade onsets B The discharge rate of buildup neurons during the gap period be fore the target onset was greater for express saccade trials compared with regular saccades After the target presentation the buildup neuron showed activity similar to the burst neu ron C The activity of fixation neurons decreased during the
14. an be defined as those whose discharge rate during the delay period is significantly higher P lt 01 than the resting rate While the buildup neu rons clearly are different from the burst neurons they may form a continuum with them the buildup neurons also may be a more heterogeneous as suggested by the broad range of discharge behaviors exhibited in response to the visual stim ulus and before the saccade C An example of a fixation neuron with a high rate of discharge while the monkey is fix ating om a target Interaction of saccades and fixation One of the striking observations on the SC is the identi fication of neurons that are related to both the genera tion of saccades and the maintenance of fixation CONTINUITY OF FIXATION AND BUILDUP NEURONS Fixation neurons have been described as a separate class of neurons because of their high discharge rate during fixation and understanding how they are related to other SC neurons depends on determining what con trols their high rate of discharge during active visual fix ation One possibility is that the discharge of these fixation neurons could be influenced by factors similar to those influencing the buildup neurons because al though we have described the buildup and fixation neu rons as behaving very differently figure 40 3 they have at least two characteristics in common First they both lie somewhat deeper in the SC than the burst neurons Munoz and Wartz 199
15. cades made in response to the cue presentation This neuron was virtually silent during Nogo trials To quantify the level of delay activity the discharge rate was measured during a 300 ms epoch ending at the time of the cue presenta tion This neuron s delay activity was 33 3 sp s and 02 sp s in Go and Nogo trials respectively WURTZ BASSO PARE AND SOMMER SUPERIOR COLLICULUS 583 A LIP efferent neuron TTL Instruction B FEF efferent neuron Stimulus Onset Instruction FIGURE 40 9 Neuronal activity of cortical efferent neurons in a Go Nogo delayed saccade Responses of lateral in traparietal LIP top and frontal eye field FEF bottom neu rons that were activated antidromically by SC stimulation In the Go trials the neurons started discharging after the stimulus delay activity in Go trials was only 1 3 times greater than that in the Nogo trials The delay activity of the LIP neu rons started at the time of the visual target and continued for most neurons regardless of whether there was prepa ration to make the saccade This is in contrast to the SC neurons where activity in approximately two thirds of the SC neurons had activity only after the Go instruction and across the sample there was 2 4 times greater activity on the Go trials Thus the SC neurons were substantially more related to the signal to move and using our inter 584 MOTOR SYSTEMS Stimulus Onset Z
16. cessary for normal vision why should such a relatively mecha nistic system be of any relevance to cognitive neuro science There are at least three reasons First saccades are relatively simple movements and they therefore offer the opportunity to understand a sim ple system within the brain for generating actions The saccades involve rotation of each eye by the coordinated activation of only six muscles have no variation of load and therefore have no need for load compensa tion and have no complexity introduced by the move ment of joints Robinson 1968 These relatively simple movements do however involve most of the cognitive issues underlying the performance of more complex ac tions For example a saccadic eye movement to a visual stimulus requires shifting attention and selecting a target transforming input from a sensory map to the output on a movement map and then coordinating the appropri ate muscles to execute the movement and hold the eye in the new position Second we have a superb animal model of the human saccadic fixation system in the old world monkey which allows us to study the brain mechanisms controlling the movement Since the introduction of the now standard techniques for recording neuronal activity from awake behaving monkeys Evarts 1966 the accurate recording of eye movements Fuchs and Robinson 1966 Judge Richmond and Chu 1980 the training of monkeys to control eye movements Fuchs 1967 a
17. ctivity in the rostral SC does directly inhibit saccade related neurons in the caudal SC Initially the saccade suppression effect caused by the activation of the SC fixation region logi 578 MOTOR SYSTEMS cally could be explained by an interaction between fixa tion and saccade related elements located downstream of the SC Indeed a pause in the activity of brainstem omnipause neurons is necessary to release the burst gen erator and produce a saccade Hepp et al 1989 Moschovakis and Highstein 1994 and these neurons receive preferential projection from the rostral SC Biit ner Ennever and Horn 1994 Par and Guitton 1994 The SC fixation neurons therefore could alter the gener ation of the saccade as an influence onto the activity of omnipause neurons that is without acting directly on the SC saccade related neurons Munoz and colleagues 1996 however showed that when electrical stimulation of the rostral SC interrupts saccades in mid flight there is a pause in the saccade related activity of both burst and buildup neurons in caudal SC when the eyes stop mo mentarily This indicates that the fixation neuron activity does act on the SC saccade related activity The mutual inhibitory effect of activity in the rostral and caudal SC also has received support from both physiological and anatomic experiments By electrically stimulating within the monkey SC Munoz and Istvan 1998 established the inhibitory effect of fixat
18. ed increase in activity but none had only a saccade related burst of activity like the SC burst neurons or some frontal eye field FEF ef ferent neurons Although the LIP neurons therefore were more similar to the SC buildup neurons than to the SC burst neurons the LIP neurons differed from the SC neurons in the Go Nogo task figure 40 94 Only about one third of the LIP efferent neurons had delay activity that occurred only after the Go instruction and the mean SC buildup neuron NOGO FP T Instruction FIGURE 40 8 Neuronal activity of a buildup neuron in a Go Nogo saccade paradigm Instructions were given by a change in color of the fixation stimulus FP before the peripheral stimulus 7 appeared A change from blue to red indicated a Nogo instruction whereas a change to green indicated a Go in struction After the peripheral stimulus was presented there was a delay period After this the fixation stimulus returned to its original blue color cueing the monkey to either maintain fixation for a prolonged duration Nogo or execute the sac cade Go Rasters and spike density functions for Go top and Stimulus Onset Cue 2003 Nogo bottom trials are aligned on the instruction lefi target middle and cue right This neuron started to discharge after the target appearance and exhibited delay activity only if the Go instruction had been given The discharge peaked at the onset of the sac
19. ed neural control of attentional shifts and eye movements Nature 384 74 77 LEICHNET2 G R and M E GOLDBERG 1988 Higher cen ters concerned with eye movement and visual attention Ce rebral cortex and thalamus In Neuroanatomy ofthe Oculomotor Sistem J A B ttner Ennever ed Amsterdam Elsevier pp 365 429 LEMON R 1984 Methods for neuronal recording in con scious animals In ZBRO Handbook Series Methods in the Neu rosciences Vol 4 New York J Wiley amp Sons pp 95 102 Lynch J C A M GRAYBIEL and L J LOBECK 1985 The differential projection of two cytoarchitectonic subregions of the inferior parietal lobule of macaque upon the deep layers of the superior colliculus J Comp Neurol 235 241 254 LYNCH J C J E HOOVER and P L STRICK 1994 Input to the primate frontal eye field from the substantia nigra supe rior colliculus and dentate nucleus demonstrated by trans neuronal rt Exp Brain Res 100 181 186 Mays L E and D L SPARKS 1980 Dissociation of visual and saccade related responses in superior colliculus neu rons J Neurophysiol 43 207 232 Mazon P R M BRACEWELL S BARASH and R A ANDERSEN 1996 Motor intention activity in the macaque s lateral intraparietal area I Dissociation of mo tor plan from sensory memory J Neurophysiol 76 1439 1456 MEREDITH M A and A S RAMOA 1998 Intrinsic circuitry of the superior colliculus Pharmacophysiological identifica ti
20. gap and paused just before either express or regular saccades However the level of activity associated with each type of sac cades did not differ Express saccades were defined as those with latencies between 70 to 120 ms and regular saccades had latencies between 130 and 180 ms After Dorris Par and Munoz 1997 stimulus was turned off but before presentation of the target was inversely correlated to the reaction time of the saccade the greater the discharge the shorter the If these early neuronal changes are related to prepara tion to make a saccade then such changes should be sensitive to the likelihood that a saccade will be made Richle and Requin 1993 Two recent experiments tested whether manipulating the prior knowledge of the monkey as to whether a saccade will be made alters the delay activity of the SC buildup neurons TARGET PROBABILITY PARADIGMS Basso and Wurtz 1997 1998 used two paradigms that required monkeys to make a saccade to a peripheral stimulus while chang ing the probability that a given stimulus would be the saccade target In the first paradigm varying the number of stimuli presented to the monkey changed the probability that the one located in the movement field of the neuron would become the target One two four or eight stim uli were presented and then later the monkey was cued as to which one was the saccade target figure 40 7 During the preselection period when the monkey did no
21. h the cortex and WURTZ BASSO PARE AND SOMMER SUPERIOR COLLICULUS 585 the SC and that there is simply a shift in the characteristics of the processing For example the pro cessing conveyed by buildup neurons is carried on in both the parietal cortex and the SC but analysis of the nature of the continuing activity shows that there is a shift in the dependence of that activity toward the prep aration to make a saccade REFERENCES ANDERSEN R and J W GNADT 1989 Posterior parietal cor tex In The Neurobiology of Saccadic Eye Movements Reviews of Oculomotor Research Vol III R H Wurtz and M E Gold berg eds Amsterdam Elsevier pp 315 336 ANDERSEN R A G K ESSICK and R M SIEGEL 1987 Neu rons of area 7 activated by both visual stimuli and oculomo tor behavior Exp Brain Res 67 316 322 BARASH S R M BRACEWELL L FOGASSI J W GNADT and R A ANDERSEN 1991 Saccade related activity in the lateral intraparietal area I Temporal properties J Neurophysiol 66 1095 1108 BASSO M A and R H WURTZ 1997 Modulation of neu ronal activity by target uncertainty Nature 389 66 69 BASSO M A and R H WURTZ 1998 Modulation of neu ronal activity in superior colliculus by changes in target probability J Neurosci 18 7519 7534 BEHAN M and N M KIME 1996 Intrinsic circuitry in the deep layers of the cat superior colliculus Vis Neurosci 13 1031 1042 BOTTNER ENNEVER J A and A
22. he buildup neu rons are found throughout the collicular map Our working hypothesis is that their buildup or delay activity represents the preparation to make a saccade and their activation may facilitate saccade production Fixation neurons become active tonically when the ani mal fixates a visual stimulus and pause when a saccade occurs figure 40 3C These neurons originally were de scribed in the cat SC Munoz and Guitton 1991 and then were identified in the monkey Munoz and Wurtz 1993a In the monkey the fixation neurons were shown further to sustain their discharge when the fixation stim ulus was removed and the monkey continued to fixate thereby ruling out the possibility that these neurons were simply visual neurons with a foveal receptive field excited by the fixation stimulus The fixation neurons are found in the rostral region of the SC on both sides of the brain Their discharge is related to the maintenance of fixation CONCLUSION SUPERIOR COLLICULUS ORGANIZA TION Its striking that there may be only a few neu ron types within the SC intermediate layers although there may be more variation particularly among the buildup neurons than we consider here The two points to be emphasized on the organization how ever are not new but rather have been recognized for a number of years First the map of saccadic vectors which we have described see review by Sparks and Hartwich Young 1989 conveys the vector f
23. hough the ratio of these signals differs between the two areas The efferents from the LIP area have not been found to show activity compa rable to the burst neurons in the SC whereas FEF ef ferents frequently do The efferents of both the LIP area and the FEF show delay activity but in the LIP area the activity is more pronounced Thus the delay activity is represented more heavily in LIP efferents than is the burst and for the FEF it is approximately DELAY BURST tie ACTIVITY ACTIVITY NEURONS vor Ii NEURONS SC BUILDUP sc A BURST FIGURE 40 10 Summary of the overlap of activity between neurons in the cerebral cortex LIP and FEF that project to the SC and the neurons in the SC burst and buildup Activity in the delay period is represented by the horizontal line the greater the activity the longer the line The burst is repre sented by the height of the vertical bar the higher the bar the larger the burst The delay activity is consistently present in LIP and SC buildup neurons but there is litle if any burst in the LIP neurons The delay activity is also present in the FEF neurons but these neurons also frequently have a burst of ac tivity as do the SC burst neurons There are therefore both dif ferences between the neurons in the two cortical areas Projecting to the SC and overlap between the cortical activity and SC activity the LIP having greater similarity to the SC buildup neurons than to the SC bur
24. in contrast to the LIP efferent neurons nearly two thirds of the FEF efferent neurons with delay activity had activity in Go trails that was higher than in Nogo trials figure 40 9B This activ ity on the Go trials although frequent was small only 1 5 times greater than that observed on Nogo trials Wartz and Sommer 1998 Thus although the FEF has buildup like activity as frequently as do the SC neurons the Go Nogo effect is smaller like that in the LIP The FEF appears to be a more heterogeneous area than is the LIP region and includes both burst and builduptype neurons While investigating the frontal cortex inputs to the SC there was an additional finding that raises ques tions about the unidirectional flow of information from cortex to brainstem outlined in figure 40 2A electrical stimulation of the SC Sommer and Wurtz 1998 showed that many FEF neurons receive inputs from the SC presumably by means of a thalamic synapse Lynch Hoover and Strick 1994 These FEF neu rons always had a phasic visual response and some also carried signals such as delay activity a presac cadic burst or fixation activity These results suggest that the processes underlying saccade generation in volve a bidirectional communication between the SC and cortical areas COMPARISON OF SUPERIOR COLLICULUS INPUTS The neurons in the LIP area and the FEF that project directly to the SC carry an amalgam of visual delay and saccade signals alt
25. in these regions have well established projections to the SC Huerta Krubitzer and Kaas 1986 Leichnetz and Goldberg 1988 Lynch Graybiel and Lobeck 1985 581 array on target dim movement field sds 001 g lg ganl114b MOTOR SYSTEMS 582 FIGURE 40 7 Buildup neuron activity modulated by shifts in target probability Increases in the number of potential targets for the impending saccade reduced buildup neuron activity The events of the task are indicated by the labeled periods of time across the top and the spatial arrangement is indicated by a schematic of the stimulus display in front of the monkey The first column shows activity on five trials in the rasters and the mean of these in the superimposed spike density trace when one A two B four C or eight D targets were presented during the preselection This example is taken from trials when the target was in the movement field of the neuron The first column is aligned vertical dashed lines on the onset of the possible targets preselection period the second on the dimming of one of the targets selection period and the third on the onset of the saccade eye movement period There is an initial visual response and subsequent sustained or delay period activity Both of these activities were reduced as the number of stimuli was increased
26. ion neu rons on burst and buildup neurons and of these sac cade related neurons back onto fixation neurons In addition these experiments showed that stimulation of saccade regions also inhibits other remotely located sac cade related neurons The only exception to the mu tual inhibition between collicular neurons is the connection between the fixation neurons in the two SCs stimulation of one fixation region excites rather than inhibits the fixation neurons in the other SC The presence of GABAergic connections throughout the SC Mize 1992 and the recent experiments in the ferret Meredith and Ramoa 1998 showing that stimulation at any point in the SC rostral or caudal produces an initial inhibition at distant points in the SC support the hypothesis of inhibitory interactions Behan and Kime 1996 showed in the cat that biocytin injections label cells within 0 5 mm up to 5 0 mm from the injection site which provides anatomic evidence for horizontal connections throughout the length of the SC Thus the mutual inhibition between different parts of the SC consistent with the fixation saccade interactions within the SC are supported by what is known about the phys iological and anatomic connections between these neu rons EXPRESS SACCADES The interaction between fixation and saccade related activity has been explored in a re Cent series of experiments on an express saccade a short latency saccade that is critically dependent
27. me small steps The four have been normalized so that the peaks of the curves are 100 The neuron with the darkened trace is the same as in A From Krauzlis and associates 1997 contra ofa fixation target They stepped the target on which the monkey was fixating to slightly eccentric locations and rewarded the monkey for making a saccade to the tar get Fixation neurons showed the largest increases in fir ing rates with small steps into the contralateral visual field the neuron in figure 40 4 showed the largest re sponse for steps of approximately 0 5 Different neu rons showed the maximal response for contralateral steps of different sizes figure 40 4B Saccades to large target steps were accompanied by a decrease in activity for both ipsilateral and contralateral saccades as had been reported previously Munoz and Wurtz 1993a What is particularly relevant is that the increase always was for small steps and that the increase in activity oc curred whether the step elicited a saccade or not Krauz lis and colleagues 1997 concluded that the fixation neurons are tonically active during visual fixation not because they are carrying a unique fixation signal but because they are indicating target locations very close to the fovea that usually do not elicit a saccade What these experiments show is that the activity of the fixation neurons during fixation has two striking similar ities to the delay activity in buildup neurons
28. nd the visual stimuli that guide them Wurtz 1969 the regions of the monkey brain active before visually guided saccades have been identified The visual pathway to the striate cortex to extrastriate areas and then to the regions of the parietal and frontal cortices have been identified and are included in figure 40 2A Just as the elaboration of the changes in visual processing at each step in this path way has contributed to our understanding of the visual input on which further cognitive processing is built the extensive understanding of the brainstem oculomotor centers for saccade generation Hepp et al 1989 Moschovakis and Highstein 1994 has led to greater un derstanding of what output is required of cognitive 573 READING FROM LEFT TO RIGHT FIGURE 40 1 Interplay between rapid eye movements sac cades and the pauses between these movements fixation In this example of the successive eye positions during the reading of a single line of text saccades vertical lines move the eye across the page but are separated by periods of visual fixation horizontal lines Almost all of visual pes occurs during these periods of visual fixation Modified from Yarbus 1967 processing for movement generation Understanding the spinal cord for the saccadic system allows us to step backward gradually from the basic mechanics of move ment to consider the control of these actions at higher levels Third there is a nexus
29. ns is active anywhere on the SC map they suppress the activity of all other SC neurons When buildup neurons in the caudal SC become ac tive they act to inhibit all other neurons in the SC in cluding the most rostral fixation type buildup neurons They also provide an error signal according to their po sition on the SC indicating which saccadic vector would be required to bring the eye to the target to eliminate the error The output of these neurons may be to adjacent burst neurons as well as directly to neu rons in the pons and midbrain outside the SC When the fixation neurons in the rostral SC become active they also inhibit all other SC neurons The fixation ac tivity also indicates that a position error exists but this error signal is so small that no eye movement is elic ited giving incidentally a physiological basis for the concept of a dead zone for saccade initiation but con siderably larger than that in humans Wyman and Steinman 1973 The outcome of this competition for fixation or saccade here and probably in other parts of the brain as well determines the initiation of saccades There may be differences in the strength of the inhibi tory effects in the rostral and caudal regions of the SC but the neural mechanisms can be regarded as funda mentally the same Several lines of evidence are consistent with the inter actions between rostral and caudal SC as outlined in fig ure 40 5 First it has become clear that a
30. of the brain opposite to the direction of the impending saccade Organization of the superior colliculus The SC lies on the roof of the midbrain and consists of successive gray and white layers Neurons in the superfi cial layers respond to visual stimuli and have receptive fields in the field contralateral to their location in the SC Neurons in the intermediate layers often also re spond to visual stimuli but their most vigorous dis charge is before the onset of saccadic eye movements We concentrate on these saccade related neurons be cause it is in them that we see the transition from sen sory to motor related activity SUPERIOR COLLICULUS MAP FOR SACCADES Just as superficial layer neurons have visual receptive fields the saccade related neurons in the intermediate layers have movement fields that is they increase their activity be fore saccades made only to one region of the visual field Wurtz and Goldberg 1972 There is a gradient of activ ity within each movement field and a neuron s maxi mum discharge is associated with saccades of specific amplitude and direction the neuron s optimal saccadic vector Different neurons have different vectors and the neurons are organized in a highly regular fashion to pro duce a neural map of saccadic vectors covering the con tralateral visual field figure 40 28 On this map large saccades are represented in the caudal portion small are in the rostral region upward
31. on of horizontally oriented inhibitory interneurons J Neu rephysiol 79 1597 1602 MILLER J 1988 Discrete and continuous models of human information processing Theoretical distinctions and empiri cal results Acta 67 191 257 MIZE R R 1992 The organization of GABAergic neurons MOHLER C W and R H WURTZ 1976 Organization of monkey superior colliculus Intermediate layer cells dis charging before eye movements J Neurophysiol 39 722 744 MOSCHOVAKIS A K and S M HIGHSTEIN 1994 The anat omy and physiology of primate neurons that control rapid eye movements Annu Rev Neurosci 17 465 488 MUNOZ D P and D GUITTON 1991 Control of orienting gaze shifts by the tectoreticulospinal system in the head free cat Il Sustained discharges during motor preparation and fixation J Neurophysiol 66 1624 1641 Munoz D P and P J ISTVAN 1998 Lateral inhibitory inter actions in the intermediate layers of the monkey superior colliculus J Neurophysiol 79 1193 1209 Munoz D P D M WAITZMAN and R H WURTZ 1996 Ac tivity of neurons in monkey superior colliculus during inter rupted saccades J Neurophysiol 75 2562 2580 Munoz D P and R H WURTZ 1991 Disruption of visual fixation following injection of GABAergic drugs into the fix ation zone of the primate superior colliculus Soc Neurosci Abs 17 544 MUNOZ D P and R H WURTZ 1993a Fixation cells in monkey superio
32. ooms Cue onset and maintained their discharge until the saccade was made in response to the cue In the Nogo trials less response was observed after the stimulus presentation Same organiza tion as in figure 40 8 pretation of the Go Nogo task were more closely related to the preparation to move whereas the LIP neurons were more dependent on the visual stimulus than the sig nal to make a saccade FRONTAL CORTEX INPUTS TO SUPERIOR COLLICU Lus The FEF is the area in the frontal lobe that to gether with the SC is necessary for saccade generation chiller True and Conway 1980 Segraves and Gold berg 1987 previously reported that FEF neurons that are antidromically activated by SC stimulation carry a saccade related burst signal that sometimes is combined with a response to visual stimulation We have con firmed this observation and find that approximately 15 of the FEF neurons have only presaccadic bursts Thus in contrast to the LIP area the FEF does have neurons projecting to the SC that have characteristics of the SC burst neurons As was the case with the LIP region the FEF projection to SC was topographically organized and directed to the intermediate layers In addition to these burst like neurons approximately one third of the FEF efferent neurons showed delay ac tivity between the onset of the stimulus and the saccade which made them similar to the SC buildup neurons Like the SC buildup neurons but
33. or the change in eye position produced by the saccade Sec ond a change in the activity of a population of neurons on the SC map has significant importance which we WURTZ BASSO PARE AND SOMMER SUPERIOR COLLICULUS 575 50 sp s have not yet emphasized Before the onset of a sac cade it is not just a few neurons that become active but roughly 25 of the entire population Munoz and Wurtz 1995b and it is the vector average of these ac tive neurons that appears to determine which saccade is made Sparks et al 1990 as represented by the mound of activity on the SC map in figure 40 2B 576 MOTOR SYSTEMS FIGURE 40 3 Classification of neurons in the intermediate layers of the SC into three categories burst buildup and fixation The top row of each example shows the eye posi tion traces and below these are rasters with each tick in the raster representing a single action potential and each row of ticks representing a single trial The superimposed spike density functions show the sum of the activity across the tri als All traces are aligned on the onset of the saccadic eye movement A An example of a burst neuron with a clear response to the onset of the visual target start of the line and a burst of activity before the saccade arrowhead under the line B An example of a buildup neuron with a slight visual response continuing activity and a burst of activi before the saccade Buildup neurons c
34. r colliculus I Characteristics of cell dis charge J Neurophysiol 70 559 575 Munoz D P and R H WURTZ 1993b Fixation cells in monkey superior colliculus II Reversible activation and de activation J Neurophysiol 70 576 589 Munoz D P and R H WURTZ 1995a Saccade related ac tivity in monkey superior colliculus L Characteristics of burst and buildup cells Neurophysiol 73 2313 2333 Munoz D P and R H WURTZ 1995b Saccade related ac tivity in monkey superior colliculus IL Spread of activity during saccades J Neurophysiol 73 2334 2348 PARE M and D GUITTON 1994 The fixation area of the cat superior colliculus Effects of electrical stimulation and di rect connection with brainstem omnipause neurons Exp Brain Res 101 109 122 PARE M and D P MUNOZ 1996 Saccadic reaction time in the monkey Advanced preparation of oculomotor pro grams is primarily responsible for express saccade occur rence J Neurophysiol 76 3666 3681 PARE M and R H WURTZ 1997 Monkey posterior parietal cortex neurons antidromically activated from superior colli culus J Neurophysiol 78 3493 3497 RIEHLE A and J REQUIN 1993 The predictive value for per formance speed of preparatory changes in neuronal activity of the monkey motor and premotor cortex Behan Brain Res 26 35 49 ROBINSON D A 1968 Eye movement control in primates Science 161 1219 1224 ROBINSON D A 1972 Eye movements evoked b
35. s 156 1 16 SPARKS D L and R HARTWICH YOUNG 1989 The deep lay ers of the superior colliculus In The Neurobiology of Saccadie Eye Movements Reviews of Oculomotor Research Vol II R H Wurtz and M E Goldberg eds Amsterdam Elsevier pp 213 256 SPARKS D L C LEE W H ROHRER 1990 Population cod ing of the direction amplitude and velocity of saccadic eye movements by neurons in the superior colliculus Cold Spring Harbor Symp Quant Biol 55 805 811 WURTZ R H 1969 Visual receptive fields of striate cortex neurons in awake monkeys J Neurophysiol 32 727 742 WURTZ R H and M E GOLDBERG 1972 Activity of superior colliculus in behaving monkey III Cells dis charging before eye movements J Neurophysiol 35 575 586 WURTZ R H and M A SOMMER 1998 Instructional depen dence of delay activity in the projection from frontal eye field to superior colliculus in macaque Soe Neurosci Abs 24 1146 WYMAN D and R M STEINMAN 1973 Small step tracking Implications for the oculomotor dead zone Vision Res 18 2165 2172 YARBUS A L 1967 Eye Movements and Vision New York Plenum WURTZ BASSO PARE AND SOMMER SUPERIOR COLLICULUS 587
36. s the vigorous burst of ac tivity before a saccade made into their movement field figure 40 3A In contrast to other saccade related neu rons they have next to no discharge during the delay pe riod The burst neurons are found throughout the rostral to caudal extent of the collicular map and it is assumed that their activity is a signal to make a saccade with a vec tor coded by the position of that neuron on the SC map Buildup neurons continue to show sustained activity during the delay period in a delayed saccade task figure 40 3B In contrast to the burst neurons these neurons do not necessarily show a discrete saccade related burst in activity and certainly must include many if not all of the saccade related neurons other than the burst new rons that have been described in many experiments be ginning with the earliest investigations Mays and Sparks 1980 Mohler and Wurtz 1976 Sparks 1978 Wurtz and Goldberg 1972 Because of their delay activ ity they originally were referred to as long lead neurons Sparks 1978 Munoz and Wurtz 1995a more recently categorized these neurons as buildup neurons because their delay activity often increases gradually as the sac cade onset approaches Note that this neuronal class also may include neurons that were described differently by other groups for example quasi visual neurons Mays and Sparks 1980 and prelude bursters Glimcher and Sparks 1992 Like the burst neurons t
37. s when an express saccade is produced Dorris Par and Munoz 1997 Edelman and Keller 1996 It was hypothesized that the release of fixation permits the target related responses of SC neurons to be strong enough to trigger shortlatency express saccades Edel man and Keller 1996 Sommer 1994 Fixation disengagement however can only be a part of the mechanism for express saccades in the monkey as indicated by recent behavioral experiments by Par and Munoz 1996 Using the gap saccade task they trained monkeys to make express saccades after present ing the saccade target repeatedly in the same part of the visual field After this training express saccades were made only to a restricted region of the visual field cen tered on the location of the target used for training The disappearance of the fixation stimulus before target pre sentation did not lead to express saccade production at the other target locations but simply reduced the mean saccadic reaction time Overall these results indicate that it is the neuronal activity associated with the prepa ration to make saccades to the training target that deter mines the occurrence of express saccades not just the release of fixation afforded by the disappearance of the fixation stimulus In subsequent physiological experi ments Dorris Par and Munoz 1997 showed that fixa tion neurons decrease their discharge before express saccades but that these changes are not predic
38. scribed emphasize several points First the interactions underlying the alternation of fixation and saccades all may depend on interactions between one nevronal class in the SC what we have re ferred to as the buildup neuron This illustrates a case in which one neuronal element can be used to produce quite different effects fixation or saccades depending on its relation to other neurons on a neuronal map Sec ond the movement generation is a result of the balance of competition between the activity of the neurons signaling movement the buildup neurons and probably the burst neurons as well and those signaling nonmovement the fixation neurons This interaction is based on compet ing activity on two parts of the SC map rostral fixation and caudal saccade and the outcome determines when a saccade will be made Superior colliculus delay activity The SC activity that we have concentrated on in con sidering fixation is the activity in the delay period that precedes the burst of activity accompanying saccades Following we consider factors that alter this delay activity WURTZ BASSO PARE AND SOMMER SUPERIOR COLLICULUS 579 Burst Neuron A Buildup Neuron Fixation Neuron FIGURE 40 6 Neuronal activity of burst A buildup B and fixation C neurons during the generation of short latency ex press saccades top and long latency regular saccades mid dle The paradigm used was the gap s
39. sig naling to the monkey that the upcoming peripheral stimulus should be either ignored Nogo trials or should be taken as the target for a saccade Go trials see stimulus bars in figure 40 8 Most buildup neurons tested in this paradigm 61 showed delay activity spe cific to the Go instruction as exemplified in figure 40 8 Sommer Par and Wurtz 1997 For these neurons the buildup activity was much higher after target pre WURTZ BASSO PARE AND SOMMER SUPERIOR COLLICULUS sentation if the Go instruction was provided that is if a saccade was to be executed For the sample of buildup neurons n 62 the delay activity in Go trials was 2 4 times greater than that observed in Nogo trials Other studies have demonstrated similar modulation of early low frequency discharge in SC neurons in tasks de signed to show activity changes with movement selec tion Glimcher and Sparks 1992 or shifts in attention Kustov and Robinson 1996 CONCLUSION DELAY ACTIVITY Both the target prob ability and the Go Nogo experiments show that a greater probability that the monkey will be required to make a saccade on a particular trial leads to a higher level of delay or buildup activity This increase is consis tent with viewing the buildup neuron activity as part of the preparation to make a saccade that precedes the burst of activity at the time of a saccade This increased activity that precedes the generation of a saccade repre sents
40. st neurons and the FEF tending to have the reverse relationship the reverse Figure 40 10 compares schematically this overlap of the activity in the LIP efferents the FEF ef ferents and the SC buildup and burst neurons Whether this input is directed separately to the burst and buildup neurons in the SC remains to be deter mined Because signals also are transmitted from the SC to cortex and between the FEF and the LIP area as well the development of saccade related activity must result from the continuous interaction between multi ple brain structures including the FEF the LIP area and the SC CONCLUSION DISTRIBUTED PROCESSING The shifts in processing from one area to the next in the brain logi cally could be of substantially different types Miller 1988 On the one hand this processing could be dis crete the processing in one area is completed and the result is passed on to the next area where a new stage of processing begins Conversely the processing might be distributed the same type of processing in one area continues at some higher level in the next area Com parison of the activity of cortical and SC neurons of fers one of the few opportunities in sensorimotor systems to determine which type of transformation oc curs In the visual oculomotor processing that we have considered it is distributed processing One of our most striking observations is that the same type of neu ronal activity can be identified in bot
41. t know which of the stimuli would be the target in creasing the number of possible targets from one to eight decreased the activity When the target dimmed the activity in each condition increased to the level present when only a single target was present During the preselection period the mean discharge rate of all buildup neurons studied decreased as target probability decreased In a second paradigm the number of stimuli re mained constant so that no changes related to visual interaction could account for the results Presenting the eight targets in a series of blocked trials in which the same one of the eight stimuli became the target increased target probability After a number of trials the neurons developed a higher level of activity than in the case when any of the eight stimuli could be come the target Therefore changes in target probabil ity modulate the buildup activity during a delay period whether those changes result from varying the number of stimuli or varying the monkey s previous experience In contrast the burst of activity of burst neurons and the saccade related burst activity of buildup neurons did not change in either probability experiment GO NOGO PARADIGM In this experiment the fixa tion stimulus provided information about whether a sac cade would be required to a peripheral stimulus located in the movement field of the neuron In the two types of trials used the fixation stimulus first changed color
42. tive of ex press saccade production figure 40 6A In contrast the increased level of discharge of buildup neurons before target presentation was found to be correlated with the occurrence of express saccades when these were made into the movement field of the neurons figure 40 6B This increase in the early activity of buildup neurons thus may facilitate the excitability of the corresponding saccade region and thereby allow the target related re sponses of the SC neurons to trigger the short latency express saccades Additional evidence for this motor preparation hypothesis for express saccade generation comes from another behavioral study Using a scanning saccade paradigm Sommer 1997 compared the spatial attributes of saccades produced in response to suddenly appearing stimuli with those made while scanning to un changing stimuli He found that the occurrence of ex press saccades was related to the congruence between the location of the suddenly appearing target and the goal of voluntary saccade planning If we generalize the observations made on express saccade generation to all saccades the decision to make a saccade can be reduced to the competitive interaction between the signals related to planning a saccade and those related to maintaining fixation Each signal is nec essary but not sufficient for determining when a sac cade will be initiated CONCLUSION SACCADE AND FIXATION INTERAC TION The observations de
43. use as the probability that a saccade will be made increases this delay activity increases Whether the processing preceding movement initiation is dis crete for each area or is distributed across areas can be an swered by comparing neuronal activity in the SC with that of the neurons in frontal and parietal cortices that project to the SC The processing is distributed with overlap between neu ronal activity in cortex and SC and a shift in the SC toward movement preparation Rapid or saccadic eye movements shift our line of sight from one part of the visual field to another and allow us to direct the higher visual acuity provided by the fovea toward successive points in the field We make these eye movements as frequently as two times per second during such tasks as scanning the visual scene or reading Just as important as the saccades however is the lack of move ment between them because it is in these periods of fixa tion when almost all vision occurs Thus saccades and fixations can be regarded as an integrated system that both moves the eye rapidly to the next target and then holds the eye steadily on that target This alternation of saccades and fixation is particularly evident during read ing figure 40 1 ROBERT H WURTZ MICHELE A BASSO MARTIN PARE and MARC A SOMMER of Sensorimotor Research National Eye Institute National Institutes of Health Bethesda Md Even given that saccades and fixation are ne
44. y collicular stimulation in the alert monkey Vision Res 12 1795 1808 SCHILLER P H J H SANDELL and J H R MAUNSELL 1987 The effect of frontal eye field and superior colliculus lesions on saccadic latencies in the rhesus monkey J Neurophysiol 57 1033 1049 SCHILLER P H and M STRYKER 1972 Single unit recording and stimulation in superior colliculus of the alert rhesus monkey J Neurophysiol 35 915 924 SCHILLER P H S D TRUE and J L CONWAY 1980 Deficits in eye movements following frontal eye field and superior colliculus ablations J Neurophysiol 44 1175 1189 SEGRAVES M A and M E GOLDBERG 1987 Functional Properties of corticotectal neurons in the monkey s frontal eye field J Neurophysiol 58 1387 1419 SOMMER M A 1994 Express saccades elicited during visual scan in the monkey Vision Res 34 2023 2038 SOMMER M A 1997 The spatial relationship between scan ning saccades and express saccades Vision Res 37 2745 2756 SOMMER M A M PARE and R H WURTZ 1997 Instruc tional dependence of preparatory discharges of superior col liculus neurons Soc Neurosci Abs 23 843 SOMMER M A and R H WURTZ 1998 Frontal eye field neurons orthodromically activated from the superior collicu lus J Neurophysiol 80 3331 3335 SPARKS D L 1978 Functional properties of neurons in the monkey superior colliculus coupling of neuronal activity and saccade onset Brain Re
Download Pdf Manuals
Related Search