* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Visual Responses of Pulvinar and Collicular Neurons During Eye
Metastability in the brain wikipedia , lookup
Neuroscience in space wikipedia , lookup
Psychoneuroimmunology wikipedia , lookup
Subventricular zone wikipedia , lookup
Time perception wikipedia , lookup
Neural coding wikipedia , lookup
Neuroesthetics wikipedia , lookup
Neuroanatomy wikipedia , lookup
Response priming wikipedia , lookup
Development of the nervous system wikipedia , lookup
Neuropsychopharmacology wikipedia , lookup
Psychophysics wikipedia , lookup
Optogenetics wikipedia , lookup
C1 and P1 (neuroscience) wikipedia , lookup
Premovement neuronal activity wikipedia , lookup
Transsaccadic memory wikipedia , lookup
Stimulus (physiology) wikipedia , lookup
Neural correlates of consciousness wikipedia , lookup
Channelrhodopsin wikipedia , lookup
Process tracing wikipedia , lookup
JOURNALOF NEUROPHYSIOLOGY Vol. 66, No. 2, August 199 1. Printed in U.S.A. Visual Responses of Pulvinar and Collicular Neurons During Eye Movements of Awake, Trained Macaques DAVID LEE ROBINSON, JOHN W. McCLURKIN, CAROLINE KERTZMAN, AND Section on Visual Behavior, Laboratory ofSensorimotor Research, National Eye Institute, National Institutes of Health, Bethesda, Maryland 20892 SUMMARY AND CONCLUSIONS 1. We recorded from single neurons in awake, trained rhesus monkeys in a lighted environment and compared responses to stimulus movement during periods of fixation with those to motion caused by saccadic or pursuit eye movements. Neurons in the inferior pulvinar (PI), lateral pulvinar (PL), and superior colliculus were tested. 2. Cells in PI and PL respond to stimulus movement over a wide range of speeds. Some of these cells do not respond to comparable stimulus motion, or discharge only weakly, when it is generated by saccadic or pursuit eye movements. Other neurons respond equivalently to both types of motion. Cells in the superficial layers of the superior colliculus have similar properties to those in PI and PL. 3. When tested in the dark to reduce visual stimulation from the background, cells in PI and PL still do not respond to motion generated by eye movements. Some of these cells have a suppression of activity after saccadic eye movements made in total darkness. These data suggest that an extraretinal signal suppresses responses to visual stimuli during eye movements. 4. The suppression of responses to stimuli during eye movements is not an absolute effect. Images brighter than 2.0 log units above background illumination evoke responses from cells in PI and PL. The suppression appears stronger in the superior colliculus than in PI and PL. 5. These experiments demonstrate that many cells in PI and PL have a suppression of their responses to stimuli that cross their receptive fields during eye movements. These cells are probably suppressed by an extraretinal signal. Comparable effects are present in the superficial layers of the superior colliculus. These properties in PI and PL may reflect the function of the ascending tectopulvinar system. INTRODUCTION Motion can be produced by the movement of visual stimuli in the external environment or it can be generated by saccadic or smooth pursuit eye movements. Central visual pathways must distinguish between these two sources of visual stimulation (von Helmholtz 1925). There are several contexts in which the brain must know how stimulus motion is produced. First, it is important for perception. Organisms need to know if an object moved or if visual field motion is derived from eye movements. Second, neural mechanisms that initiate responses to movement should not be falsely triggered by image motion that is self induced. Finally, those parts of the brain that control smooth pursuit and optokinetic eye movements need to deal differently with image motion, depending on its origin. Errors in mo- STEVEN E. PETERSEN tion discrimination would lead to unstable oculomotor control. Some have proposed that there is a corollary discharge from the oculomotor system to the visual system that indicates the presence of reafferent stimulation (von Holst and Mittelstaedt 1950; Sperry 1950; Teuber 1960). Others suggested purely visual processes or visual masking as mechanisms for eliminating vision during saccadic eye movements (Campbell and Wurtz 1978; MacKay 1970; Matin et al. 1972). It seems likely that both types of mechanisms (visual and extraretinal) contribute to conscious perception, oculomotor control, and other aspects of visual behavior. Several cortical areas have been investigated to understand visual responsiveness around the time of eye movements. Visual activity during saccadic and smooth pursuit eye movements has been tested in areas 17, 18, 19, and 7 (Battaglini et al. 1986; Fischer et al. 198 1; Galletti et al. 1984, 1986, 1988; Judge et al. 1980; Sakata et al. 1985; Wurtz 1969). Each of these regions has varying concentrations of neurons that respond differentially for real and selfinduced stimulus movement. We have shown that cells in the superior colliculus respond differentially to rapid stimulus movement (Robinson and Wurtz 1976). In the present studies we have found that cells in the retinotopitally organized regions of the inferior pulvinar (PI) and lateral pulvinar (PL), which are targets of collicular projections, also respond differentially to stimulus movement (Benevento and Fallon 1975; Benevento and Standage 1983; Marrocco et al. 1981; Partlow et al. 1977). We hypothesize that an extraretinal signal reduces responses to stimuli during eye movements. These data suggest that such processes may originate in the colliculus, be transmitted to PI and PL, and from there influence visual cortex. Brief reports of these studies have appeared previously (McClurkin and Robinson 1987; Robinson and Petersen 1985). METHODS Four rhesus monkeys (3 males, 1 female; 4.0-M kg) were used in these experiments. Initially each was trained to fixate a small spot of light and to release a bar when the fixation point dimmed to receive a water reward. After the monkeys had achieved a performance criterion, they were implanted with a cranial pedestal and a scleral search coil using sterile surgical procedures (Petersen et al. 1985). For the surgery, each monkey was premeditated with atropine sulfate (0.05 mg/kg im) to reduce respiratory tract secretions and 485 486 ROBINSON with cephalothin sodium (50 mg/kg im) to combat infection. Initial anesthesia was induced with ketamine HC1(4-60 mg/kg im). After the ketamine, pentobarbital sodium (20-33 mg/kg) was given through a transcutaneous catheter inserted into the saphenous vein, and the monkey was placed in a stereotaxic frame. Heart rate, respiratory rate, and body temperature were monitored and body temperature was controlled with a heating pad. After the surgery, the monkey was allowed to recover in an incubator in which the temperature was maintained at 80’F. When the monkey was alert, it was returned to its home cage. For 2 days after the surgery, each monkey was administered pentazocine lactate (1 S-3.0 mg/kg im) to relieve any postoperative discomfort. The monkeys were given food and water ad lib, were checked daily, and were given prophylactic cephalothin sodium for 1 wk. All procedures were conducted under a protocol approved by the Animal Care and Use Committee of the National Eye Institute and complied with Public Health Service Policy on the humane care and use of laboratory animals. Behavioral tasks After the monkeys had fully recovered from the surgery, they were retrained on the fixation task while their heads were restrained. To train the monkeys to maintain precise fixation, we monitored eye position and terminated trials if the eye moved outside of a computer-generated, 2O-square window surrounding the fixation point (Fig. IA). Once they were able to fixate correctly, they were trained to make saccadic eye movements from one stimulus to another on simultaneous termination of the fixation point and appearance of a target. They also learned to make smooth pursuit eye movements by following the slow, continuous movement of the fixation point. Both of these eye movement tasks required that eye position be maintained within 2’ of the desired location. In all conditions, the monkey’s task was to detect the dimming of a target light. The following conditions were used to assess the response characteristics of cells in relation to several types of stimulus movement. All testing was done binocularly. During repeated fixations, the visual receptive field was determined with a hand-held ophthalmoscope. Then the speed preference of a cell was measured by sweeping a stimulus through the receptive field at eight speeds ranging from 10 to 64O”/s in a pseudorandom order (Fig. IA). The stimuli consisted of circles or bars with a luminance of 1.5 log units above the background of 1 cd/m2. The size, shape, and direction of movement of the stimulus was selected to elicit a consistent response from the cell. On other blocks of trials the monkey made pursuit eye movements at 2O”/s or saccadic eye movements to peripheral targets; both were configured to sweep the cell’s receptive field across a stationary stimulus (Fig. 1B). The stimulus was the same one used in the speed preference experiments. Stimuli were selected and positioned to ensure that the visual receptive field actually crossed them even with slight variations in the trajectory of the eye movement. To measure the strength of observed effects, we increased the brightness of the stationary stimulus up to a maximum of 3.1 log units above the intensity of the background. The pursuit targets traveled a distance of 20’ from the center of the screen to the periphery, and the saccade targets were presented at an eccentricity of 20’ in either the horizontal or vertical direction. Finally, the same eye movements were made without a stationary stimulus (Fig. 1C). The monkeys also were trained to make saccades to remembered targets in total darkness (Hikosaka and Wurtz 1983). In this paradigm the saccade target was flashed on for 75 ms; after an interval ranging pseudorandomly from 50 to 500 ms, the fixation point was turned off, signalling the monkey to make a saccade to the place where the target had been. This situation eliminated the possibility of visual interactions. The monkey was rewarded if its ET AL. Moving Fixation Light /-----~~~ Stimulus (IpI Receptive Field (-j-j-t Rec~c;~ Fi;ld Eye Position Window 1 B Eye Position Window#2 Saccade Target ~~7~~~~~~~~u,us Fixation Light /. Eye Position Window #1 C 1 FIG. 1. Experimental conditions. A: tangent screen in the real stimulus movement condition. The monkey was required to fixate a central light, keeping his eye within the eye position window. The receptive field of a cell was plotted by hand and a stimulus was swept through it. In the self-induced image motion condition (R), the monkey began the trial by fixating a central spot and then a stationary stimulus was presented on the tangent screen. Next, the fixation point was extinguished, and simultaneously a target was turned on in the periphery: the monkey made a saccade to fixate this target. In the pursuit condition, after the stationary stimulus was presented, the fixation point was moved slowly, and the monkey was required to follow it, keeping his eye within a window. In both of these conditions, the resulting eye movements swept the receptive field across the stationary stimulus. C: no-stimulus condition. The monkey was required to make either pursuit or saccadic eye movements as before, but no receptive-field stimulus was present. eye position ended inside a 4’ window the previously illuminated target. surrounding the location of Recording locations Recordings were made from the two retinotopically organized regions of the pulvinar [inferior map (PI) and lateral map (PL)] and the superficial layers of the superior colliculus. The techniques for locating these structures have been described in detail previously (Petersen et al. 1985). Guide tubes were positioned in various parts of the maps of each of these structures and several penetrations were made at each location with tungsten microelectrodes (Crist et al. 1988). We tested a total of 20 guide-tube loca- REAL AND SELF-INDUCED tions in PI and PL (6 definitely localizedin PI, 7 unambiguouslyin PL, and 7 at the vertical meridian betweenPI andPL) and 6 in the superiorcolliculus. We made an averageof nine penetrationsin eachguide tube (range:2-19). Collicular penetrationswere continued until we encounteredneuronsthat dischargedbefore eye movements.Previousresearchindicatesthat suchcellsarelocated in the intermediatelayers(Wurtz and Goldberg 1972).Wedid not sampleenoughcells unequivocally located in PI or PL to make quantitative distinctionsbetweenthem for the propertiesstudied here. However, at leastthree cellswith any particular qualitative property were unequivocally located in each map. The bulk of cellshad fields located in the lower visual quadrant with eccentricities from 0” to 20”. At the sitesof interestingcells,we passed6 PA of anodalcurrent for 30-60 s to makemarking lesions,which were later located on histological sections.The brains were pre- OBJECT MOTION 487 paredfor histologicalprocessingafter the animal, under deepbarbiturate anesthesia(250 mg/kg ip), was perfusedwith salinefollowedby formaldehyde.Figure 2 showsa microelectrodelesionin PI aswell asreconstructionsof a penetration in PI and one in PL. Response quantijcation All activity wasaligned on a trigger event that wasspecificfor eachtask. In the fixation task,the event wasthe start of the stimulus movement. In the saccadetask, the event wasthe moment at which the eyeleft the 2” window surroundingthe fixation point. In the pursuit task, the trigger wasthe start ofthe movementofthe pursuit target. Cell dischargesevoked during pursuit were Iate in the display, comparedwith other conditions, becauseof the slow speedof the target. Activity wassummedby replacingeachspike B FIG. 2. Reconstruction of penetrations into PI and PL. The outline drawing in A shows the path of a penetration into PI (long, vertical line) with tick marks indicating the locations of 5 recording sites. B: receptive-field sizes and locations for the neurons studied during the penetration in A. Dashed lines connect the field centers. C: another penetration into PL from which 6 neurons were studied, the corresponding receptive-field data are in D. E: photomicrograph of section through the pulvinar showing a penetration marked with an electrolytic lesion. The section was stained with cresyl violet. For the penetration into PI (A and B), at site 1 multi-unit activity was localized; site 2 was a neuron that responded to real but not self-induced movement in the light and dark; sites 3 and 4 were cells that responded to all speeds tested; site 5 was a neuron responding best to low-speed motion. From the penetration into PL (Cand D), site 6 was a multi-unit recording; site 7 was a cell responding well to all speeds; sites 8 and 10 were cells responding equally well to real and self-induced movement; site 9 was a cell activated by real but not self-induced motion; site 11 was a neuron activated best by low-speed movement. Positive and negative signs in A and C indicate the upper and lower visual fields, respectively. LGN, lateral geniculate nucleus; MGN, medial geniculate nucleus; FP, fixation point, thus fovea; PDM, dorsomedial portion of the lateral pulvinar (Petersen et al. 1985); PM, medial pulvinar; BSC, brachium of the superior colliculus; H, hippocampus. 488 ROBINSON TABLE Distribution 1. of response characteristics Saccade n Pulvinar Superior colliculus Neurons movement qualitatively ET AL. Suppressed Pursuit Task Facilitated No effect n Suppressed Task Saccade/Pursuit Facilitated No effect n Similar effects Relations Different effects* 77 35 10 32 34 10 8 16 22 9 13 24 18 0 6 29 22 1 6 22 14 8 were considered to be suppressed or facilitated if the difference between the response during the fixation condition and the response in the eye condition had a P < 0.05 (t test). *Different effects were noted when the response with saccadic eye movements (e.g., response facilitation) was different from that with pursuit eye movements (e.g., response suppression). peripheral bins (affected neurons) of Fig. 5. For the neurons in each structure we evaluated the strength and distribution of effects for the whole population by the use of standard errors (Fig. 5). We used the following formula to calculate the number of standard errors of the mean by which the responses differed: (X, - &,)/ (SEf + SE,,), where X, is mean response to stimulus movement during fixation, X,, is mean response to stimuli during eye movement condition, SE,, is standard error of the mean during eye movement condition, and SE, is standard error during fixation. Only cells with P < 0.05 were considered significantly modulated by eye movements, and positive values indicate that the response in the fixation condition was greater than in the eye movement situation; negative values represent the opposite. For examining the distribution of differential responses, we used a t test that compared one population of neurons against a normal distribution of data with the same number of neurons and the same standard deviation. in each trial with a gaussian kernel having a standard deviation of 8 ms, thereby producing a spike density function for each trial (Richmond et al. 1987). These spike density functions were summed within experiments to obtain the summed responses for each condition. Responses were quantified by first setting an analysis window for each condition. The locations and widths of the analysis windows were adjusted so as to capture the whole response observed in the summed spike density functions. Examples of window placements are illustrated in Fig. 4. When responses of the same cell in different tasks were to be compared, the same analysis window width was used. In the speed series, the widths of the analysis windows became smaller with increasing speed. In the pursuit eye movement condition, the width of the analysis window was set to be the same as that used for stimulus movement of 2O”/s. In the saccadic eye movement condition, the width of the analysis window was set to be the same as that used with stimulus movement of 64O”/s. The spikes in each analysis window then were counted, and the mean, standard deviation, and standard error of the mean were computed to obtain a response measure that was independent of the width of the analysis window and the number of trials in each condition. This quantification utilized actual spike data contained within the windows rather than the spike density functions. When comparing the responses of a cell in the fixation and eye movement conditions, we used a Student’s t test for each neuron. Only cells for which the difference between the means was significant at the P < 0.05 level were considered modulated. This was the criterion used for categorizing cells for Table 1 and also the criterion for placing neurons in the center bins (unaffected cells) or the A RESULTS We recorded from a total of 243 cells in the two retinotopically organized areas of the pulvinar (PI and PL) (Bender 198 1; Petersen et al. 1985) and 7 1 neurons in the superficial layers of the superior colliculus. Of these neurons, we were able to isolate 83 in PI and PL and 35 in the superior colliculus long enough to obtain complete quantitative data (Table 1). B 60 40 LD P s 1 1 I 200 I I 400 I I 2 = e n 1 200 600 DEGREE S/SEC P P 1 1 400 I 1 600 FIG. 3. Speed-tuning characteristics. Each graph shows the speed data of a different cell. Each circle represents the mean response of a cell in spikes/second/trial. Vertical bars above and below each circle show 1 SE of the mean. The horizontal line through each plot depicts the mean response for that cell across all speeds tested. A: responses of a broadly tuned cell. B: cell tuned to low speeds. C: cell tuned to intermediate speeds. D: cell tuned to high speeds. FIG. 4. Responses to stimulus movement. Data in A and B illustrate the responses of 2 cells recorded in PL during fixation and saccade tasks. Left histogram-raster pair: response to stimulus movement during fixation (the circle and arrow depicting the stimulus and direction of motion). Stimulus speed was 640’1s. Right histogram-raster pair: activity with saccadic eye movements. Eye movement traces from 20” saccades are above the histogram, on which it was aligned. The cell in A responded well to real but not to self-induced motion. The cell in B responded well to real stimulus movement and better to stimulation generated during eye movements. Spike density functions were calculated only from the trials shown in the raster. The solid vertical line at the left of each raster-histogram pair indicates the beginning of the stimulus or eye movement, and its height indicates the same scale for the same cell. The pair of dashed lines in each set of data illustrates the placement of the analysis windows (see METHODS). Each horizontal row in the rasters represents a single trial and each dot corresponds to a single action potential. This data format will be used in subsequent figures. #I56 t 200 MSEC I SACCADE 40 c PULVINAR PURSUIT r A 20 B l-l 20 - IO 20 -C 20-D SUPERIOR coLLlcuLus I0 FIG. 5. Comparison of response differences. Left: graphs come from expenments comparing rapid stimulus movement during fixation with motion during saccadic eye movements. Right: data compare responses to image motion of 2O”/s during fixation and that induced with smooth pursuit eye movements. Each row of graphs comes from data collected from a different structure. Vertical axis indicates the number of neurons with a certain relationship between real and self-induced movement. Center column in each graph includes the cells for which responses in the 2 stimulus movement conditions did not reach the criterion of significance at the P < 0.05 level (t test). The first column to the right shows the number of neurons for which the responses to motion during fixation were significantly different and between 1 and 3 SE greater than the responses to motion with eye movements. The column to the far right represents those cells for which the responses differed by >3 SE. For the columns to the left of center, with negative values, responses to eye movement were significantly greater than to motion during fixation (t test). See METHODS for the procedures and formula that were used for calculating standard error differences. 489 490 ROBINSON ET AL. Self-induced image motion-pulvinar I i 400 MSEC B %+ 360 t 200 MSEC I FIG. 6. Similarity of responses in pursuit and saccade conditions. A: data illustrate the responses of a cell in PI to real movements at 2O”/s and self-induced motion in the pursuit condition at 2O”/s. B: responses of the same cell to real and self-induced movement in the saccade condition (64O”/s stimulus and 20” eye movement). In A, the triggers are the start of the stimulus movement or pursuit movement and the response is delayed. The vertical scales in A are the same even though the number of trials differs. Note that the time scales are very different for A and B. Responses to stimulus movement duringfixation-pulvinar We first measured the speed tuning of the neurons in our sample. Stimuli were swept across the receptive fields of neurons at each of eight speeds ranging from 10 to 64O”/s while the monkeys fixated. The mean response to the eight speeds was computed for each cell, and the responses for the individual speeds were compared with this mean. Cells were classified as being broadly tuned to all speeds or more narrowly tuned to low, intermediate, or high speeds. Broadly tuned cells were those in which discharges to each of the test speeds were within 1 SE of the mean response. Even those neurons that responded best to some speeds still discharged at all speeds. Low-speed cells responded to lower speeds 2 1 SE greater than the mean and to high speeds 2 1 SE less than the mean. Comparable criteria were used for intermediate- and high-speed cells. Examples of four types of speed tuning are illustrated in Fig. 3. Overall, we found far more broadly tuned than selectively tuned cells (x2 = 77.48, df = 3, P < 0.001). In PI and PL, 63% (52/83) were broadly tuned whereas 9% (S/83) were tuned to low speeds, 11% (9/83) to intermediate speeds, and 17% (14/83) to high speeds. We conclude from these data that cells in PI and PL respond over a very wide range of stimulus speeds and have marginal selectivity for speed. After demonstrating that cells in PI and PL respond to real image movement during periods of fixation, we tested whether the same cells responded to comparable motion when it was produced by eye movements. Neurons were considered to be able to distinguish real from self-induced movement if their responses in the saccade condition differed from their responses in the fixation condition at least at P < 0.05 (see METHODS). The responses of 77 neurons from PI and PL were examined quantitatively, and some gave smaller responses (Fig. 4A), whereas others gave larger responses (Fig. 4B) in the saccade condition (Table 1). Unequivocal examples of these properties were observed in penetrations in PI and also in penetrations in PL. Thus some neurons in PI and PL responded differently to image movement, depending on how the motion was generated. These experiments confirm our preliminary observations (Robinson and Petersen 1985). From our present sample we found no other property that correlated with the type of response to eye movements (suppression or facilitation). Figure 5A shows the distribution of differential responses in the two stimulus movement conditions. There is a collection of neurons that respond better to real than self-induced motion (positive standard error values), and this significantly shifts the mean to the right (t = 3.23, df = 76, P < 0.002) when the population is compared with a normal distribution with the same number of samples and the same standard deviation. For this statistical analysis we used the actual number of standard errors rather than the binned groupings in Fig. 5. It is clear from such an analysis that the retinotopically organized areas of the PI and PL, when considered as a summed, whole population, respond better to 6 I 200 MSEC FIG. 7. Responses of a collicular neuron to different forms of stimulus motion. Left: response of the collicular cell to rapid stimulus movement during periods of fixation. Right: complete lack of response to the same stimulus during saccadic eye movements. REAL AND t SELF-INDUCED 4 400 MSEC OBJECT MOTION 491 Superior colliculus Having demonstrated this differential activity for cells in PI and PL, we wished to know if such properties were present in another subcortical visual relay structure. We conducted comparable tests of speed selectivity on cells in the superficial layers of the superior colliculus. In the superior colliculus, 46% (16/35) of the neurons were broadly tuned whereas 29% (10/35) were tuned to low speeds, 11% (4/35) to intermediate speeds, and 14% (5/35) to high speeds. These distributions of tuning were not significantly different from those found in PI and PL. We also studied 24 neurons in the superior colliculus in the saccade and pursuit conditions. Some collicular neurons gave smaller responses in the saccade condition than during fixation (Figs. 5A and 7); this is similar to findings reported by Robinson and Wurtz (1976). The number of LIGHT 216 t 200 MSEC FIG. 8. Activity of collicular cell with pursuit and saccadic ments. A: cell responded clearly to a stimulus moved through receptive field at 2O”/s during periods of fixation (left). The sponded when the same stimulus was moved through the field pursuit eye movements (right). B: the same cell also discharged movement at 64O”/s during fixations (left) but did not respond motion was generated by saccadic eye movements (right). 1 eye movethe visual cell also reby smooth to stimulus when such stimulus movement during fixation than to stimulus movement generated by saccadic eye movements. Next we were interested in knowing whether cells in PI and PL were also modulated during smooth pursuit eye movements. We tested 34 cells in PI and PL, and some neurons had reduced responses in the pursuit condition (Fig. 6A, Table 1). Unambiguous examples of this characteristic were found in both retinotopically organized areas of the PI and PL. The responses occur later in the data displays because the speeds used were much slower than with saccadic eye movements. Some of these same cells also had reduced responses when tested in the saccade condition (Fig. 6B). Of the cells with response reduction during smooth pursuit, six failed to respond at all to the self-induced movement. Response enhancement during pursuit was also observed as was equivalent responding in both conditions. Figure 5B shows the distribution of differential responding in these two conditions, fixation and pursuit. There is no significant deviation of the mean from a normal distribution, and this suggests that PI and PL, when taken as a whole, summed population, do not distinguish between these two types of slow movement. Table 1 shows the distribution of response characteristics with saccadic and pursuit eve movements. LIGHT RY499 FIG. 9. - 200 MSEC Evidence of an extraretinal input. Response of a neuron from PI to real movement (top).Response to self-induced motion in the light (center) and the dark (bottom). This cell failed to respond to self-induced movement even when visual inputs were drasticallv reduced. 492 ROBINSON cells and the strength of their effect produced a very significant shift in the mean of the population (t = 5.7 1, df = 23, P 4 0.001) (Fig. 5C). PI and PL and the superior colliculus have activity that shifts the means of their pooled data toward a differentiation between real and self-induced movement. The superior colliculus population signal is strongly biased toward responses to real stimulus movement during fixation; the population signal from PI and PL is also biased but not as strongly as is the colliculus. Figure 5 also illustrates that the deviation of the mean for pursuit is significant for the superior colliculus (t = 4.7, df = 28, P 4 0.001) but not for PI and PL. Of 22 collicular cells tested in both saccadic and pursuit eye movement conditions, one had equal responses to real and self-induced movement in both conditions. The responses of 13 cells were suppressed in both pursuit and saccade conditions. Two cells showed a facilitated response in the pursuit condition only, whereas six cells had reduced responses in the saccade condition only. An example of this latter characteristic is illustrated in Fig. 8. This cell responded briskly to slow image movement, whether it was generated externally during fixation (Fig. 8A, left) or by smooth pursuit movements (Fig. 8A, right). The same neuron also responded to rapid image movement during fixation (Fig. 8B, left) but not when produced by saccadic eye movements (Fig. 8B, right). Table 1 shows that some superior collicular cells responded qualitatively differently in the two eye movement situations whereas others had similar responses. The type of response to self-induced movement did not depend on speed tuning. This was true for PI, PL, and the superior colliculus. We conclude from these data that the responses of many cells in PI and PL and the superior colliculus are modulated with eye movements. The strength of the population response is greatest in the superior colliculus and significant but less intense in PI and PL (Fig. 5). A B LIGHT ET AL. Source of difirences movement between real and selj?nduced There are at least two mechanisms that could alter the response of cells to visual stimuli during eye movements. One process would use visual interactions with images in the surround of the visual receptive field. The other mechanism is an extraretinal signal, a nonvisual input to these cells that would modulate their responses. We tried to distinguish between these possibilities by testing 13 cells in PI and PL and 23 collicular neurons while the monkey was in total darkness except for the test stimuli. We selected neurons that had a clear distinction between real and self-induced movement. The results of this type of experiment are shown in Fig. 9 for a cell recorded from PI. The neuron responded to image motion during fixation and did not discharge when comparable motion was caused by eye movements in the light and dark (Robinson and Petersen 1985). All 13 cells in PI and PL gave comparable effects when tested in the light and dark, and we obtained this result for 4 cells clearly located in PI and 3 in PL. The same result was found for all collicular cells tested here, as was demonstrated previously (Robinson and Wurtz 1976). To further exclude the possibility of visual interactions in the saccade condition, we tested 10 cells recorded in PI and PL while the monkey made saccades to remembered targets in total darkness (Hikosaka and Wurtz 1983). Figure 10 shows the activity of a neuron with real and self-induced image movement; there was both a lack of response to the visual stimulus after the eye movement and a suppression of activity after the saccade (Fig. 1OB). The suppression was also present after eye movements made to remembered targets in total darkness (Fig. IOC). These characteristics were observed in 4 of the 10 neurons tested and indicate that PI and PL receive an extraretinal signal. Similar properties are found in cells in the superficial layers of the superior colliculus (Goldberg and Wurtz 1972; Richmond and Wurtz C LIGHT TOTALDARKNESS FIG. 10. Activity with remembered eye movements in total darkness. A: response of this cell from PI to a stimulus swept across its receptive field during periods of fixation. As shown in B, this cell did not respond to the same stimulus when it was moved across the receptive field by saccadic eye movements. When the monkey made eye movements in total darkness, there was a suppression of activity after the saccade. The monkey was rewarded for making the eye movement to the point where a target light had appeared previously. TY478a 100 MSEC REAL AND SELF-INDUCED 1980; Robinson and Wurtz 1976). For the six remaining cells tested in this way, five discharged to the stimulus motion generated by eye movements and had no suppression with remembered eye movements in total darkness; the other neuron did not respond to self-induced motion and had no suppression in the dark. An additional 25 cells not investigated with real and self-induced movement were studied with eye movements in the dark; of all 35 cells tested, 16 had suppression with eye movements. For 13 neurons in this sample, we also flashed stimuli in the receptive field at varying times before and after eye movements to determine excitability changes (Robinson et al. 1990). Those cells that had a suppression of activity with A E+ 15. LOG OBJECT MOTION 493 remembered eye movements in total darkness also had a reduction in excitability (4/ 13, 3 1%) and conversely, those without suppression in the dark had no change in their excitability (6/ 13,46%). Three neurons with increased excitability after eye movements had either no suppression with remembered eye movements (n = 2) or a slight increase in activity (n = 1). Intensity ofmodulation In PI and PL, a number of the neurons completely failed to respond to stimuli that crossed their receptive fields during eye movements (Figs. 4A, 9, and IOB), To test the strength of the processes that prevent responses during eye movements, we increased the intensity of the stimuli for 23 cells in PI and PL and 12 cells in the superior colliculus that had no response to our standard stimuli. For 14 cells in PI and PL and 9 collicular neurons, responses with eye movements began to appear with stimuli 2.1 log units above background (Fig. 11). For the balance of the cells tested in this way, responses began to appear with stimuli 3.1 log units above the background. We conclude from these experiments that the suppressive process is not absolute. DISCUSSION We compared responses of cells to real stimulus movement during periods of fixation with motion that was produced by eye movements. We have demonstrated that many cells in PI and PL respond weakly, if at all, when motion is caused by eye movements. Similar properties are common in the superficial layers of the superior colliculus. When the same experiment is conducted in the dark, thereby eliminating visual stimulation of the surround of the visual receptive field, the result is similar. Also, some of these same cells have a reduction in activity after eye movements made in total darkness. Thus these cells receive an extraretinal signal. This input probably prevents cells in PI and PL from responding to visual stimulation produced by eye movements. In discussing our data we will first consider the functional contributions these neurons might make. Then we will relate our data to other experiments that have addressed the issue of vision during eye movements. 1.5 LOG 21. LOG Functional 31. LOG RY216 H FIG. 11. Effects of increasing stimulus intensity. -4: a cell recorded from PL. B: a superior collicular neuron. Top pair: responses of cells to standard stimuli swept through the receptive field during fixations. Second pair: response to self-induced motion when the intensity of the stimulus was equal to that of the moving stimulus (1.5 log units above background). Bottom two pairs: responses to self-induced movement when the intensity was increased to 2.1 and 3.1 log units, respectively. implications There are cells within PI and PL that incorporate visual and eye movement signals. Such neurons may be organized to encode stimulus salience, participate in the initiation of motor responses, function in motion perception, or reflect the reduction in vision during eye movements (Chalupa 1990). It is possible that signals from these cells are part of a system for selecting interesting objects from visual space, for encoding the salience of the visual stimuli. These neurons are visual in nature, although they are not very selective (Bender 1982; Petersen et al. 1985). As we have demonstrated here, they do receive suppressive signals that eliminate self-induced stimulation. We have also reported a facilitation of these neurons before eye movements (Petersen et al. 1985). Taken together these properties would be helpful for those systems using salience, those operating to choose the next image to be processed by fovea1 vision. 494 ROBINSON The visual processing we have studied may be part of a system leading to the initiation of movement. Those parts of the brain that start eye movements to salient images and those that initiate somatic movements to such targets need signals that are uncontaminated by self-generated motion. Because cells in PI and PL respond -64 ms after visual stimulation (Petersen et al. 1985), there is adequate time for their signals to be transmitted to other brain structures. Previous studies in the cat have shown that stimulation of the pulvinar can affect oculomotor neurons (Wilson and Goldberg 1980). However, because eye movements can be started after lesions of the pulvinar, this part of the brain cannot be the sole site of initiation (Bender 1988; Ungerleider and Christensen 1977, 1979). When one class of cells in PI and PL discharges (Fig. 4A), this indicates the presence of images in the external environment. Because the cell responses are suppressed with eye movements, they are not contaminated with self-induced stimulation. At the simplest level, these cells might help produce the conscious perception of motion. However, there are several limits to this interpretation. PI and PL neurons are not well tuned for the direction of motion nor for orientation (Bender 1982; Petersen et al. 1985). We have demonstrated here that these cells are not well tuned for stimulus speed (Fig. 3). Given the modest visual properties of these cells, it is unlikely that these individual neurons encode motion perception. However, a population signal might encode more precise data (McIlwain 1975, 1976). Furthermore, if signals from the present cells were combined with those in the middle temporal area that are well tuned for motion (Baker et al. 198 1; Maunsell and Van Essen 1983), then a more appropriate construct might be present. Cells in PI and PL could reflect the suppression of vision during saccadic eye movements. Although there are qualitative similarities between the physiology and perception, there are clear differences. Saccadic suppression appears much less intense (0.5 log units) than the effect we report here (Riggs et al. 1982). Another difficulty is that some studies have suggested that purely visual factors can account for the perceptual effect, whereas our cells receive an extraretinal signal. MacKay (1970) showed that rapid movement of the visual scene produces an elevation of threshold comparable with that found in saccadic suppression experiments. Others have shown that visual stimulation just before or after an eye movement will eliminate any vision during the saccade. This is true for perception as well as for activity of cells in striate cortex (Campbell and Wurtz 1978; Judge et al. 1980). Other studies ofvision during eye movements in primates Previously we demonstrated that cells in the superficial layers of the superior colliculus, like those in PI and PL, respond over a wide range of stimulus speeds, do not discharge to visual stimuli during saccadic eye movements, and can be induced to respond to bright images during eye movements (Robinson and Wurtz 1976). In the superior colliculus there is an inhibition with eye movements in the dark, and there too this input is most likely the source of response reduction with saccades. The superficial layers of ET AL. the superior colliculus project to PI and PL (Benevento and Fallon 1975; Harting et al. 1980; Marrocco et al. 198 1; Partlow et al. 1977; Raczkowski and Diamond 1978). We propose that these collicular properties are conferred onto the neurons we have studied in PI and PL. This pathway might seem functionally quiescent, because collicular lesions have no effect on the visual responses of neurons in PI, whereas striate lesions completely silence the area (Bender 1983). However, striate cortex may provide the visual inputs to PI cells whereas the colliculus provides the nonvisual mechanisms that suppress responses with eye movements. Many studies of the primate visual system have shown eye movement-related effects on visual responses. In the squirrel monkey, there is a decrease in excitability in the magnocellular layers of the lateral geniculate nucleus with eye movements in the dark (Bartlett et al. 1976). Depression followed by facilitation was also observed in the visual cortex (Bartlett et al. 1976). There were minimal changes in geniculate activity related to eye movements in the dark (Biittner and Fuchs 1973), although stimulation of the reticular formation significantly influenced transmission through the lateral geniculate (Cohen et al. 1969). Cells in striate cortex can respond equivalently to motion whether it is caused by real image movement or by saccadic eye movements (Fischer et al. 198 1; Wurtz 1969). However, others have described cells in striate cortex that respond differently to real and self-induced motion (Battaglini et al. 1986). PI and PL project to the most superficial layers of the striate cortex (Benevento and Rezak 1976; Ogren and Hendrickson 1976, 1977); if striate cells in layer I receive eye movement modulations they may not have been sampled because of the difficulty of isolating these neurons. Initial studies of area 17 reported equivalent responses for real and pursuit-generated motion (Fischer et al. 198 1). Recent experiments have reported that a small fraction of cells respond differently to self-induced movement when it is generated by tracking (Galletti et al. 1984). A subset of cells in V2 and area 19 responded differently to real and self-induced movement. This was true for both saccadic and smooth pursuit movements (Battaglini et al. 1986; Fischer et al. 198 1; Galletti et al. 1986, 1988). Overview Our work and that of others has shown that cells in PI and PL are visual in nature, although they are not very selective for stimulus features (Bender 1982; Petersen et al. 1985). The present and previous studies show that these same neurons are modulated by various eye movement conditions (Acuna et al. 1983; Perryman et al. 1980; Petersen et al. 1985; Robinson and Petersen 1985; Robinson et al. 1990). Their visual responsescan be suppressedafter eye movements, making them lessaccessibleto visual stimulation, and their activity can be adjusted by eye position. Many of these properties are present in the cells of the superficial layers of the superior colliculus, which are afferents to both PI and PL (Benevento and Fallon 1975; Benevento and Standage 1983; Marrocco et al. 198 1; Partlow et al. 1977; Raczkowski and Diamond 1978). We hypothesize that the modulatory influences in PI and PL come from the REAL AND SELF-INDUCED superficial layers of the superior colliculus, although the visual properties come from the cortex (Bender 1983). The dominant efferents from PI and PL go to striate and prestriate cortical areas (Benevento and Rezak 1976; Kennedy and Bullier 1985; Lin and Kaas 1979). We propose that the modulated visual signals in PI and PL are projected to these visual cortical areas and might be used for establishing visual salience. Such signals might be used eventually for attentional purposes or for triggering motor responses. We appreciate the technical help of G. Creswell, L. Cooper, C. Crist, T. Ruffner, A. Ziminsky, and C. Scoutaris; we are also grateful for the help of J. Steinberg and B. Zoltick in preparing this manuscript. Present address of S. E. Petersen: Dept. of Neurology and Neurological Surgery, Washington University School of Medicine, St. Louis, MO 63 110. Address for reprint requests: D. L. Robinson, Section on Visual Behavior, Laboratory of Sensorimotor Research, National Eye Institute, Bldg. 10, Rm. lOClO1, Bethesda, MD 20892. Received 19 November 1990; accepted in final form 16 April 199 1. REFERENCES ACUNA, C., GONZALES, F., AND DOMINGUEZ, R. Sensorimotor unit activity related to intention in the pulvinar of behaving Cebus apeZZamonkeys. Exp. Brain Res. 52: 41 l-422, 1983. BAKER, J. F., PETERSEN, S. E., NEWSOME, W. T., AND ALLMAN, J. M. Visual response properties of neurons in four extrastriate visual areas of the owl monkey (Aotus trivirgatus): a quantitative comparison of the medial, dorsomedial, dorsolateral, and middle temporal areas. J. Neurophysiol. 45: 397-416, 1981. BARTLETT, J. R., DOTY, R. W., LEE, B. B., AND SAKAKURA, H. Influence of saccadic eye movements on geniculostriate excitability in normal monkeys. Exp. Brain Res. 25: 487-509, 1976. BATTAGLINI, P. P., GALLETTI, C., AICARDI, G., SQUARTRITO, S., AND MAIOLI, M. G. Effect of fast moving stimuli and saccadic eye movements on cell activity in visual areas V 1 and V2 of behaving monkey. Arch. Ital. Biol. 124: 11 I- 119, 1986. BENDER, D. B. Retinotopic organization of the macaque pulvinar. J. Neurophysiol. 46: 672-693, 198 1. BENDER, D. B. Receptive field properties of neurons in the macaque inferior pulvinar. J. Neurophysiol. 48: 1- 17, 1982. BENDER, D. B. Visual activation of neurons in the primate pulvinar depends on cortex but not colliculus. Brain Res. 279: 258-26 1, 1983. BENDER, D. B. Electrophysiological and behavioral experiments on the primate pulvinar. Prog. Brain Res. 75: 55-65, 1988. BENEVENTO, L. A. AND FALLON, J. H. The ascending projections of the superior colliculus in the rhesus monkey (Macaca mulatta). J. Comp. Neural. 160: 339-362, 1975. BENEVENTO, L. A. AND REZAK, M. The cortical projections of the inferior pulvinar and adjacent lateral pulvinar in the rhesus monkey (Macaca mulatta): an autoradiographic study. Brain Res. 108: l-24, 1976. BENEVENTO, L. A. AND STANDAGE, G. P. The organization of projections of the retino-recipient and nonretino-recipient nuclei of the pretectal complex and layers of the superior colliculus to the lateral pulvinar and medial pulvinar in the macaque monkey. J. Comp. Neural. 2 18: 307336,1983. B~~TTNER, U. AND FUCHS, A. F. Influence of saccadic eye movements on unit activity in simian lateral geniculate and pregeniculate nuclei. J. Neurophysiol. 36: 127-141, 1973. CAMPBELL, F. W. AND WURTZ, R. H. Saccadic omission: why we do not see a grey-out during a saccadic eye movement. Vision Res. 18: I2971303, 1978. CHALUPA, L. M. The visual function of the pulvinar. In: The Neural Basis of‘ Visual Function, edited by A. G. Leventhal. London: Macmillan, 1991, p. 140-159. COHEN, B., FELDMAN, M., AND DIAMOND, S. P. Eflects of eye movement, brain-stem stimulation, and alertness on transmission through lateral geniculate body of monkey. J. Neurophysiol. 32: 583-594, 1969. GRIST, C. F., YAMASAKI, D, S. G., KOMATSU, H., AND WURTZ, R. H. A grid system and a microsyringe for single cell recording. J. Neurosci. Metho& 26: 117-122, 1988. OBJECT MOTION 495 FISCHER, B., BOCH, R., AND BACH, M. Stimulus versus eye movements: comparison of neural activity in the striate and prelunate visual cortex (Al 7 and A19) of trained rhesus monkey. kp. Brain Res. 43: 69-77, 1981. GALLETTI, C., BATTAGLINI, I? P., AND AICARDI, 6. ‘Real-motion’ cells in visual area V2 of behaving macaque monkeys. Ekp. Brain Res. 69: 279288,1988. GALLETTI, C., BATTAGLINI, P. P., SQUATRITO, S., MAI~LI, M. G., AND AICARDJ G. Real motion cells in the posterior bank of the lunate sulcus of behaving monkeys. Behav. Brain Res. 20: 98-99, 1986. GALLETTI, C., SQUATRITO, S., BATTAGLINI, P. P., AND MAIOLI, M. G. ‘Real-motion’ cells in the primary visual cortex of macaque monkey. Brain Res. 30 1: 95- 1 10, 1984. GOLDBERG, M. E. AND WURTZ, R. H. Activity of superior colliculus in behaving monkey. I. Visual receptive fields of single neurons. J. Neurophysiol. 35: 542-559, 1972. HARTING, J. K., HUERTA, M. F., FRANKFURTER, A. J., STROMINGER, N. L., AND ROYCE, G. J. Ascending pathways from the monkey superior colliculus: an autoradiographic analysis. J. c’otnf). Neural. 192: 853882,198O. VON HELMHOLTZ, H. Treatise (?f‘P~zysiologiclal Qptics, edited by J. P. C. Southall. Menasha, WI: Optical Society of America, 1925. HIKOSAKA, 0. AND WURTZ, R. H. Visual and oculomotor functions of monkey substantia nigra pars reticulata. III. Memory-contingent visual and saccade responses. J. Neuroph~~siol. 49: 1268- 1284, 1983. VON HOLST, E. AND MITTELSTAEDT, H. Das Reafferenzprinzip. Wechselwirkungen zwischen Zentralnervensystem und Peripherie. Naturwissenschaften 37: 464-476, 1950. JUDGE, S. J., WURTZ, R. H., AND RICHMOND, B. J. Vision during saccadic eye movements. I. Visual interactions in striate cortex. J. Neurophysiol. 43: 1133-1155, 1980. KENNEDY, H. AND BULLIER, J. A double-labeling investigation of the afferent connectivity to cortical areas V 1 and V2 of the macaque monkey. J. Neurosci. 5: 28 15-2830, 1985. LIN, C.-S. AND KAAS, J. H. The inferior pulvinar complex in owl monkeys: architectonic subdivisions and patterns of input from the superior colliculus and subdivisions of the visual cortex, J. Camp. Neural. 187: 655677, 1979. MACKAY, D. M. Elevation of visual threshold by displacement of retinal image. Nature Land. 225: 90-92, 1970. MARROCCO, R. T., MCCIXRKIN, J. W., AND YOUNG, R. A. Spatial properties of superior colliculus cells projecting to the inferior pulvinar and parabigeminal nucleus of the monkey. Brain Rcs. 222: 150- 154, 198 1. MATIN, E., CLYMER, A. B., AND MATIN, L. Metacontrast and saccadic suppression. Science Wash. DC 178: 170-l 8 1, 1972. MAUNSELL, J. H. R. AND VAN ESSEN, D. C. Functional properties of neurons in middle temporal visual area of the macaque monkey. I. Selectivity for stimulus direction, speed, and orientation. J. Neurophysiol. 49: 1127-1147,1983. MCCLURKIN, J. W. AND ROBINSON, D. L. Subcortical visual signals during eye movements. Invest. Ophthalmol. Visual Sci. 28: 124, 1987. MCILWAIN, J. T. Visual receptive fields and their images in superior colliculus of the cat. J. Neurophysiol. 38: 2 19-230, 1975. MCILWAIN, J. T. Large receptive fields and spatial transformations in visual system. Int. Rev. Physiol. 2: 223-248, 1976. OGREN, M. P. AND HENDRICKSON, A. E. Pathways between striate cortex and subcortical regions in Macacu mulatta and Suimiri sciureus: evidence for reciprocal pulvinar connections. kkp. Neural. 53: 780-800, 1976. OGREN, M. P. AND HENDRICKSON, A. E. The distribution of pulvinar terminals in visual areas 17 and 18 of the monkey. Brain Res. 137: 343-35OJ977. PARTLOW, G. D., COI,ONNIER, M., AND SZABO, J. Thalamic projections of the superior colliculus in the rhesus monkey, Mucaca mulatta. A light and electron microscopic study. J. Ci,n?p. Neural. 17 1: 285-3 18, 1977. PERRYMAN, K. M., LINDSLEY, D. F., AND LINDSLEY, D. B. Pulvinar neuron responses to spontaneous and trained eye movements and to light flashes in squirrel monkeys. Elc)i’trocnc’c’l,lluIogr. Clin. Neurophysiol. 49: 152-151, 1980. PETERSEN, S. E., ROBINSON, D. L., AND KEYS, W. Pulvinar nuclei of the behaving rhesus monkey: visual responses and their modulations. J. Neurophysiol. 54: 867-886, 1985. RACZKOWSKI, D. AND DIAMOND, I. T. Cells of origin of several efferent pathways from the superior colliculus in Gulugo senegalensis. Brain Res. 146: 351-357, 1978. 496 ROBINSON B. J., OPTICAN, L. M., PODELL, M., AND SPITZER, H. Temporal encoding of two-dimensional patterns by single units in primate inferior temporal cortex. I. Response characteristics. J. Neurophysiol. ET AL. R. W. Neural basis of spontaneous optokinetic response produced by visual inversion. J. Comp. Physiol. Psychol. 43: 482-489, RICHMOND, SPERRY, 57: 132-146, 1987. RICHMOND, B. J. AND WURTZ, H.-L. Perception. In: Handbook oj-Physiology. Neurophysiology, Washington, DC: Am. Physiol. Sot., 1960, sect. 1, vol. III, p. 1595-1668. U~GERLEIDER, L. G. AND CHRISTENSEN, C. A. Pulvinar lesions in monkeys produce abnormal scanning of a complex visual array. Neuropsychologia. 17: 493-501, 1979. UNGERLEIDER, L. G. AND CHRISTENSEN, C. A. Pulvinar lesions in monkeys produce abnormal scanning of a complex visual array. Neurophysiol. 17: 493-501, 1979. WILSON, J. S. AND GOLDBERG, S. J. Inputs of the pulvinar and lateral posterior nucleus into the abducens nucleus of the cat. Exp. Neural. 68: R. H. Vision during saccadic eye movements. II. A corollary discharge to monkey superior colliculus. J. Neurophysiol. 43: 1156-l 167, 1980. RIGGS, L. A., VOLKMANN, F. C., MOORE, R. IS., AND ELLICOTT, A. G. Perception of suprathreshold stimuli during saccadic eye movement. Vision Res. 22: 423-428, 1982. ROBINSON, D. L., MCCLURKIN, J. W., AND KERTZMAN, C. Orbital position and eye movement influences on visual responses in the pulvinar nuclei of the behaving macaque. Exp. Brain Res. 82: 235-246, 1990. ROBINSON, D. L. AND PETERSEN, S. E. Response of pulvinar neurons to real and self-induced stimulus movement. Brain Res. 338: 392-394, 1985. ROBINSON, D. L. AND WURTZ, R. H. Use of an extraretinal signal by monkey superior colliculus neurons to distinguish real from self-induced stimulus movements. J. Neurophysiol. 39: 852-870, 1976. SAKATA, H., SHIBUTANI, ICI., KAWANO, K., AND HARRINGTON, T. L. Neural mechanisms of space vision in the parietal association cortex of the monkey. Vision Res. 25: 453-463, 1985. 1950. TEUBER, 72-86,198O. R. H. Comparison of effects of eye movements and stimulus movements on striate cortex neurons of the monkey. J. Neurophysiol. WURTZ, 32:987-994, 1969. R. H. AND GOLDBERG, M. E. Activity of superior colliculus in behaving monkey. III. Cells discharging before eye movements. J. Neurophysiol. 35: 575-586, 1972. WURTZ,