| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Laboratoire de Neurobiologie de la Cognition, Centre National de la Recherche Scientifique, Marseille, France; 2Laboratory of Neuropsychology, National Institute of Mental Health, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland; and 3Istituto di Fisiologia Umana, Università di Palermo, Palermo, Italy
Submitted 21 October 2005; accepted in final form 4 February 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; Apicella et al. 1991; Kimura et al. 1984) and to primary motivating events (Apicella et al. 1997; Ravel et al. 1999, 2003). However, TANs displaying spatially selective responses to movement-triggering stimuli have been reported in monkeys performing limb-movement (Kimura 1986, 1992) and oculomotor tasks (Shimo and Hikosaka 2001). The selectivity of these neuronal responses in behavioral conditions involving rewarding stimuli for which the motor reaction is different raises the possibility that the network of the TANs might also play a role in the processing of spatial information that is essential for attentional purposes or for the initiation of specific motor acts. The influence of the spatial requirements of the task is therefore an aspect of TAN functioning that would be of particular interest with regard to the role of the striatum in the control of action. However, in the studies reporting spatial selectivity of TAN responses, the selection and guidance of movement direction were confounded with the location of the stimulus that initiated the movement. This implies that the spatially selective responses of TANs might have been related to the motor aspects of the intended movement or the spatial features of the stimulus. It remains to be established whether selectivity for stimulus location reflects the position of the stimulus calling for a movement in a particular direction or the direction of the movement itself.
Until now, there is little evidence that TANs participate in the initiation of movements. These neurons have been implicated in processes relating to stimulus detection, particularly with respect to a stimulus whose detection depends on the behavioral context. In caudate nucleus, for example, TAN responses can be contingent on a particular spatial location of a visual stimulus even if no reward is associated with this stimulus (Shimo and Hikosaka 2001). On the basis of these findings it has been suggested that TANs may have a role in the detection of the context in which stimuli and actions occur. According to this view, responses of TANs act to integrate the characteristics of a given behavioral condition to the control of goal-directed behavior. In addition, it has been suggested that TAN responses may also depend on motivational (Yamada et al. 2004) and attentional aspects of task performance (Blazquez et al. 2002). Yamada et al. (2004) found that TANs in the caudate nucleus had contrasting properties in their responses to task stimuli compared with those in the putamen. In particular, the fraction of TANs responding to movement-triggering stimuli was higher in the putamen, possibly reflecting the preferential involvement of this nucleus in motor function. This indicates that differences in the localization of TANs exhibiting distinct response properties may exist between striatal regions.
In the present study, we recorded TAN activity from the caudate nucleus and putamen in monkeys performing visually triggered reaching movements to spatially distinct targets. We also used a variant of the task that elicited reaching toward or away from the spatial location of the trigger stimulus to dissociate the spatial attributes of the stimulus from those of the intended movement. We report that the responsiveness of TANs can be influenced by the spatial features of the task, not just detecting stimuli linked to reward.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Three male macaque monkeys (monkeys ENK, TIM, ART; Macaca fascicularis), weighing 5–6 kg, were trained to make visually guided arm-reaching movements to obtain a liquid reward. Animal housing and experimental procedures were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and the French laws on animal experimentation. The behavioral apparatus was similar to that previously described (Apicella et al. 1997; Sardo et al. 2000). Each monkey was seated in a specially designed restraining box, facing a panel 30 cm from its head (Fig. 1). The panel contained two metal knobs (10 x 10 mm) separated by 20 cm horizontally, as well as two light-emitting diodes (LEDs), one above each knob, at eye level of the animal. An unmovable metal bar was mounted at the center of the panel at waist level. Each trial began by keeping the hand on the bar. After a variable period of 0.5–2.0 s, one LED was illuminated with a red light that served as trigger stimulus. In response to this signal, the monkey released the bar and made a movement to contact the knob below the illuminated light to receive a liquid reward.
|
There were two task conditions, termed the "choice" and "dissociated" conditions. In each condition, the signal to move was the illumination of the red LED either on the left or on the right side of the panel, the location of the stimulus varying pseudorandomly from trial to trial. In the choice condition, the arm movement was directed toward the target corresponding to the location of the trigger stimulus. In the dissociated condition, the monkey is required to contact the target located ipsilaterally to the moving arm whenever the trigger stimulus was presented, regardless of its location. In this latter condition, designed to dissociate the direction of movement from the location of the trigger stimulus, the ipsilateral target was continuously surrounded by a white light to indicate the constant movement direction. The light indicating the appropriate arm-movement target was turned on at the beginning of the block of dissociated trials and turned off at the end of the block of dissociated trials. This visual signal indicated the rule that defined the dissociated condition, i.e., the mapping between the stimulus location (left or right) and the appropriate motor response (contact the illuminated target). For 50% of the trials, the movement was directed toward the spatial location of the trigger stimulus. For the remaining 50% of the trials, the movement was directed away from the spatial location of the trigger stimulus. The choice and dissociated conditions were conducted in separate blocks of 40–60 trials, the presence or absence of the continuously illuminated target providing an external cue to the monkey as to which of the two task conditions was being administered. In both conditions, monkeys had to release the bar within 1 s of the appearance of the trigger stimulus and touch the target within 1 s of releasing the bar. Trials were presented with an intertrial interval of 3–6 s. Before the recording experiments started, all three monkeys were trained on the choice condition until they reached a steady level of performance, performing at >95% correct daily, and monkeys ART and TIM were also trained on the dissociated condition. Monkeys ENK and TIM performed the task with the right arm and monkey ART worked with the left arm.
Electrophysiological recordings
On completion of training, animals underwent surgery under sodium pentobarbital anesthesia (Nembutal, 35 mg/kg, administered intravenously) and aseptic conditions. A stainless steel recording chamber (25 mm OD) was implanted on the skull over the left hemisphere. The center of the chamber was aimed at the anterior commissure according to the atlas of Szabo and Cowan (1984). In the same surgery session, two pairs of periorbital silver–silver chloride electrodes were implanted into the brow ridges and temporal bones for recording eye movements, and two stainless steel cylinders were fixed to the skull with surgical screws and dental acrylic for mechanical stabilization of the head during recording sessions. Antibiotics (ampicillin, 17 mg/kg, administered intramuscularly every 12 h) were injected on the day of surgery and for 5 days after the surgery. The recording chamber was filled with an antibiotic solution (flumequil) and sealed with a removable cap.
Single-neuron recordings were performed using tungsten microelectrodes inserted into the striatum through a stainless steel guide tube (0.6 mm OD). A hydraulic microdrive (MO95, Narishige) was used to advance the microelectrode into the striatum. Single-neuron impulses were amplified, filtered with a band pass of 0.3 to 1.5 kHz, displayed on oscilloscopes, and isolated using a window discriminator. The resulting standard digital pulses were stored on a computer, together with markers of task events. The computer also presented visual stimuli, delivered liquid reward, and measured behavioral parameters. Data on spike forms were also stored for off-line analysis of spike duration. We used well-established electrophysiological criteria to identify TANs, such as spontaneous discharge rate and duration of spikes (Alexander and DeLong 1985; Aosaki et al. 1994; Apicella et al. 1997; Hikosaka et al. 1989; Kimura et al. 1984). The characteristics of the recorded neurons were analyzed using a personal computer software developed by E. Legallet. The relationships of neuronal activity were assessed on-line in the form of rasters aligned on the task events. The electrode was advanced until the spikes generated by a single neuron were well isolated. Once a neuron was isolated, we first examined its activity in the choice condition. If the isolation could be adequately maintained for a sufficient period of testing, we changed to the dissociated condition. Each neuron was recorded for a total of 40–60 trials in each condition. Neuronal activity was occasionally recorded outside the context of the reaching task, when the liquid was automatically delivered alone once every 5–8 s, the interval between deliveries being randomized so that the time of reward remained unpredictable. This condition, called the "free reward condition," was designed to test the sensitivity of TANs to unexpected primary rewards (Apicella et al. 1997). Horizontal electrooculograms (EOGs) were recorded together with signals from neuronal activity during the behavioral testing and stored to disk as analog signals (sampling frequency: 500 Hz) for off-line analysis.
Data analysis
We calculated reaction time (RT) as the interval between the onset of the trigger stimulus and release of the bar and movement time (MT) as the interval from the release of the bar until the hand contacted the target. Behavioral performance was assessed by calculating the mean of RT and MT of correct responses for each stimulus location and each session. Quantitative analysis of EOG data was made off-line by single-trial analysis, and involved latencies of saccadic eye movements directed toward trigger stimuli. A Wilcoxon signed-rank test was used to statistically analyze any changes in the neuronal activity (Apicella et al. 1997; Sardo et al. 2000). Activity of each neuron after the onset of the trigger stimulus was compared with baseline activity determined during a control period of 500 ms just before the trigger presentation. A test window with a duration of 100 ms was moved in 10-ms steps starting at the onset of the trigger stimulus. A neuron was considered as responsive if its mean discharge was significantly different (P < 0.05) from the control period during at least five consecutive steps. The temporal characteristics of the responses of TANs to external stimuli typically included a short-lasting depression in activity, as indicated by perivent time histograms and the slopes of cumulative frequency distributions of neuronal impulses. We carried out a number of preliminary tests to determine the most adequate number of successive steps for the detection of this pause response with our sliding-window procedure based on the Wilcoxon test. In particular, the procedure allowed us to detect the period of reduced activity independently of the presence of additional response components, i.e., a rebound activation after the pause response or an initial brief increase in firing. We defined the latency of the response as the beginning of the first of five consecutive steps showing a significant difference as against baseline firing rate calculated during the control period. Response offset was determined in the same way by searching for a loss of significant differences during five steps. The percentages of responses in relation to the total number of neurons tested were calculated for each condition and each stimulus location. Differences in proportions of responding neurons among task conditions and stimulus locations were statistically assessed with the 
2 test. Response latencies, durations, and magnitudes were compared between stimulus locations with one-way ANOVAs.
Histology
Near the end of neuronal data collection, we made electrolytic lesions along four to six tracks in each monkey by passing negative currents through the microelectrode (20 µA for 20–30 s). After the completion of all testing, the monkeys were killed with an overdose of sodium pentobarbital and perfused transcardially with 0.9% saline followed by a fixative (4% paraformaldehyde, pH 7.4 phosphate buffer). Frozen coronal sections (50 µm thick) were cut through the region of the recordings and stained with cresyl violet. Based on the topographic organization of cortical projections, the region of the striatum explored in the present study was divided into two territories: the "sensorimotor" striatum, which predominantly includes the posterior portion of the putamen, and the "associative" striatum, which includes mainly the head and body of the caudate nucleus and the anterior portion of the putamen (Parent and Hazrati 1995). Very few neurons were recorded in the so-called limbic striatum, which is composed of the nucleus accumbens and adjacent ventromedial caudate and ventral putamen (Haber and McFarland 1999).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
The monkeys performed the task correctly on 
95% of the trials in both the choice and dissociated conditions. Table 1 shows the mean RTs and MTs for the three animals. In the choice condition, RTs to stimuli presented contralaterally to the moving arm were longer than RTs to ipsilateral stimuli (Student's t-test, P < 0.01) in monkeys ART and ENK. Although a trend to longer RTs between contralateral and ipsilateral stimuli can be seen in monkey TIM, this difference was not significant (P > 0.05). In all three monkeys, movements made in response to stimuli whose location was contralateral to the moving arm were associated with longer MTs than those made in response to ipsilateral stimuli (P < 0.01). Monkeys ART and TIM were also tested on the dissociated condition, i.e., the condition in which the direction of the reaching movement remained the same across trials irrespective of whether the trigger stimulus was presented on the side contralateral or ipsilateral to the moving arm. Monkey ART had shorter RTs when the location of the trigger stimulus and the target of movement corresponded in space (compatible trials) than when they did not correspond (incompatible trials) (P < 0.01). On the other hand, in monkey TIM, RTs of the compatible trials were in the same range as those of the incompatible trials (P > 0.05). In both monkeys, MTs of the incompatible trials were not different from MTs of the compatible trials (P > 0.05), suggesting that the manipulation of spatial stimulus–response correspondence did not produce a significant effect on the response execution side.
|
|
Neuronal data
Based on specific electrophysiological characteristics that were described previously (Alexander and DeLong 1985; Hikosaka et al. 1989; Sardo et al. 2000), we identified 216 neurons as TANs. Baseline firing rates of these neurons ranged from three to 18 spikes/s (7.3 ± 2.4, mean ± SD, n = 216). We further confirmed the identification of TANs by observing the changes in neuronal activity while monkeys performed the task. In general, TANs showed a response pattern similar to that reported previously, consisting in a brief, short-latency depression of activity after appearance of the trigger stimulus, which was often followed by a rebound facilitation in firing (Aosaki et al. 1994; Apicella et al. 1997; Kimura et al. 1984). After a first stage of recording from monkeys ENK and ART, new versions of the reaching task were introduced for the purpose of complementary analyses, thus explaining that the number of neurons tested in the choice and/or dissociated conditions in these two animals (monkey ENK, n = 23; monkey ART, n = 38) was smaller compared with monkey TIM (n = 155).
Responses in the choice condition
Among 216 neurons tested in the choice condition, 181 responded to the visual stimulus that elicited arm movements. Of these 181 neurons, 127 (70%) responded to the trigger stimulus regardless of its location and 54 (30%) responded to only one location of the trigger stimulus. Figure 3A is an example of a neuron with a response to the trigger stimulus when it was presented on the side either contralateral or ipsilateral to the moving arm. This type of response was classified as spatially nonselective. In contrast, the neuron whose activity is shown in Fig. 3B had a response occurring only for a stimulus presented contralaterally to the moving arm and not for an ipsilateral stimulus. This type of response was classified as spatially selective. Approximately two thirds of the responses that were spatially selective (11/13, 85% in monkey ART; 6/8, 75% in monkey ENK; 17/33, 52% in monkey TIM) were found to occur only when the stimulus was contralateral to the moving arm, the remaining responses (2/13, 15% in monkey ART; 2/8, 25% in monkey ENK; 16/33, 48% in monkey TIM) occurring when the stimulus was ipsilateral.
|
2 = 9.53; df = 1; P < 0.01) and ENK (2 = 3.89; df = 1; P < 0.05), compared with monkey TIM, whereas the proportion of selective responses to the ipsilateral stimulus was similar for all three monkeys (P > 0.05). There were significantly more nonselective neurons in monkey TIM than in monkey ART (2 = 3.93; df = 1; P < 0.05). In monkeys ART and ENK, the spatial preference in terms of TAN responsiveness was obvious when the stimulus was presented contralaterally to the moving arm, the percentage of responses being higher to the contralateral stimulus than to the ipsilateral stimulus. In contrast, in monkey TIM, the proportion of spatially selective responses was not different between the two stimulus locations. It therefore appears that the spatial response bias that was observed behaviorally was paralleled by a spatial preference of TAN responses, a contralateral preference in TAN responses being associated with decreased movement speed for initiating movements in the contralateral direction.
|
|
). In the present study, responses to the delivery of liquid were seen in all 22 of the neurons tested in the free reward condition, these neurons also being responsive to the trigger stimulus in the reaching task. For example, the neuron for which activity is shown in Fig. 5 exhibited a response to the trigger stimulus that was specific for the contralateral location, but did not respond to reward in the reaching task. However, the same neuron gave a clear response to reward given alone outside the task. This indicates that TANs can respond to primary motivating events and to stimuli that predict reward and, in this latter case, may incorporate details concerning the spatial features of the task. It therefore appears that the spatial information was not absolutely required for producing TAN responses but demonstrates an interaction between motivational and spatial aspects. Figure 5 also illustrates eye movements recorded during neuronal data collection. As already mentioned, the direction of saccade covaried with the location of the trigger stimulus in the reaching task. On the other hand, in the free reward condition, the monkey moved his eyes spontaneously and randomly during the interreward interval and did not look at any particular location at the time of reward delivery. It therefore appears that TAN responses were not specifically related to eye movements.
|
To examine whether responses of TANs to the trigger stimulus are related to the location of the stimulus or to the direction of the associated movement, we used, in monkeys ART and TIM, the dissociated condition that involved constant movements toward the ipsilateral target, regardless of the location of the stimulus. We tested 21 neurons displaying spatially selective responses, all occurring when the stimulus was contralateral to the moving arm (six in monkey ART, 15 in monkey TIM), and 23 neurons with spatially nonselective responses (nine in monkey ART, 14 in monkey TIM).
When neurons showing spatially selective responses were tested in the dissociated condition, eight neurons maintained their spatial preference (i.e., they continued to respond to the contralateral stimulus), eight completely lost their responsiveness (i.e., they became unresponsive to both stimulus locations), two reversed their spatial preference (i.e., they became responsive to the ipsilateral stimulus), and three lost their spatial preference (i.e., they became responsive to both stimulus locations). Thus 62% of the spatially selective responses (13 of 21 neurons) were modified when the stimulus location was dissociated from the direction of the movement. An example of spatially selective response is illustrated in Fig. 6A. In both the choice and the dissociated conditions, this neuron showed a response to the trigger stimulus presented contralaterally to the moving arm, suggesting that the neuron's spatial preference depended on the location of the stimulus rather than the direction of the movement. Another example of a TAN with a selective response is shown in Fig. 6B. However, unlike the neuron described before, the spatial selectivity was not maintained when the condition was changed, suggesting that the response was specific for the spatial location of the target for reaching rather than being related to the location of the stimulus. In other neurons, changes in spatial selectivity could not be attributed simply to the location of the stimulus or to the direction of the movement. The neuron in Fig. 7A, for example, showed a response to the contralateral stimulus in the choice condition and this spatial preference was reversed in the dissociated condition, the neuron becoming responsive to the ipsilateral stimulus. Figure 7B shows an example of a neuron that responded to both stimulus locations when the condition was changed. It therefore appeared that spatial selectivity in the TAN response depended not only on stimulus location or the direction of movement but also on the particular condition in which the trigger stimulus elicited a movement.
|
|
|
Locations of neurons with trigger responses
Figure 9 illustrates the locations of all TANs studied in the choice condition in all three monkeys. It can be seen that neurons were sampled from the full extent of the explored striatum, with the exception of its ventral part. Although neurons with trigger responses of the different types were distributed over the entire striatal areas from which we recorded, we found more nonselective responses (2 = 5.19; df = 1; P < 0.05) and selective responses to the ipsilateral stimulus (2 = 4.55; df = 1; P < 0.05) in the associative striatum than in the sensorimotor striatum. On the other hand, incidences of selective responses to the contralateral stimulus, although higher in the sensorimotor striatum than in the associative striatum, did not vary significantly between these two striatal territories (2 = 3.75; df = 1; P > 0.05). The reconstructed recording positions for the 44 neurons studied in both the choice and dissociated conditions in the two monkeys are shown in Fig. 10. Incidences of the three types of neurons, with stimulus-dependent responses, movement-dependent responses, and context-dependent responses, did not vary significantly between the two striatal territories (P > 0.05).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Contralateral preference of the spatially selective responses
Our results show that approximately two thirds of the TANs displaying spatially selective responses preferred the contralateral stimulus location. Kimura (1986, 1992) described TANs in the putamen with selective responses to spatially distinct visual stimuli eliciting arm movements in different directions, but the relationship between the spatial selectivity of these responses and the laterality of stimulus presentation was not reported. In the caudate nucleus, the majority of TANs recorded during an oculomotor task were responsive to visual stimuli contralateral to the hemisphere from which recordings were made (Shimo and Hikosaka 2001). In our experiments, the recordings were always carried out in the striatum of the left hemisphere and TANs showed a spatial preference in their responses when the trigger stimulus was ipsilateral (monkey ENK) or contralateral (monkey ART) to the recorded striatum, indicating that there is no consistent relationship between the spatial selectivity of neuronal responses and the side of recording. These two monkeys performed the task with the left arm (monkey ART) or the right arm (monkey ENK), suggesting that the effects of stimulus laterality on TAN responses might be related to the arm to be used for the reaching movement. It is possible that the effects of stimulus laterality on TAN responses varied as a function of the requirements of the task, particularly the information type used to initiate the behavioral reaction. To perform the memory-guided saccade task that was used by Shimo and Hikosaka (2001), the monkey had to remember the spatial location of stimuli that indicated the future saccade direction and the increased responsiveness of TANs to stimuli presented contralaterally to the recorded striatum may have been related to spatial memory for the location of the stimulus. In the task that we used, stimuli triggered immediate movements directed toward targets and the predominance of TAN responses to stimuli presented contralaterally to the moving arm was accompanied by a variation in the speed of the motor reaction elicited by the stimulus: arm movements usually were faster when target side and responding arm corresponded than when they did not correspond. This is consistent with behavioral studies in humans showing that ipsilateral reaches are initiated more quickly and were completed more rapidly than were reaches directed at contralateral targets (Fisk and Goodale 1985). These results raise the possibility that, in the context of a reaching task, the preference of TANs for contralateral stimuli reflects a degree of sensitivity to the correspondence between the effector and the side toward which the motor response was directed. Whether the requirements of the task being performed account for the apparent differences between our results and those of Shimo and Hikosaka (2001) remains an issue for further investigation.
Nature of the responses to the stimulus triggering the reaching movement
Although previous studies have shown that the spatial location of the stimulus that elicited or instructed the direction of arm or eye movements may be reflected in the response of TANs (Kimura 1986, 1992; Shimo and Hikosaka 2001), none was designed to dissociate the spatial features of the stimulus from the direction of the movement associated with that stimulus. We have addressed this issue in the present study by using the dissociated version of the task in which the monkey reached toward or away from the spatial location of the trigger stimulus. We found that a substantial number of TANs changed their response properties when the task was switched from the choice to the dissociated condition, permitting identification of three different types of responses to the trigger stimulus that are predominantly influenced by the sensory, motor, or contextual aspects of the reaching task. These results suggest the existence of subgroups of TANs that may play a preferential role in distinct processes related to spatial information processing. Recently, Yamada et al. (2004) reported that different subsets of TANs modulate their responsiveness according to the stimulus used for triggering actions, the motivational context of actions, or combinations of action and context. Together with this previous finding, our data provide further evidence that TANs are less homogeneous in their response properties than currently appreciated.
We found that 50% of the TAN responses to the trigger stimulus were not dependent on whether the monkey moved the arm toward or away from the spatial location of the stimulus, suggesting that these responses reflect the location of the stimulus itself. This is consistent with the prominent role of TANs in detecting stimuli linked to rewards. Previous single-neuron recording studies in behaving monkeys have demonstrated that TAN responses are time-locked to the trigger stimulus but not to the associated movement (Aosaki et al. 1994; Kimura 1992; Yamada et al. 2004). The activity changes reported here further indicate that TAN responses may also reflect the location of the stimulus in space. Because monkeys' reaching movements were usually preceded by a saccade directed toward the stimulus presented at either of the two locations, both in the choice and dissociated conditions, there is also a possibility that TAN responses, instead of being involved in the detection process, were associated with saccadic eye movements. However, examination of eye movement data under our task conditions has shown that TAN responses were not directly linked with the direction of gaze or with the initiation of eye movements. On the other hand, it is conceivable that these responses might reflect the processes of spatial attention because it has been suggested that TANs may be involved in some attentional functions (Blazquez et al. 2002). The absence of control over the spatial distribution of attention in our experiments does not allow us to determine to what degree TAN responses are dependent on attentional demands of the task being performed.
The present study has also demonstrated that 25% of the TAN responses to the trigger stimulus were related to the direction of reach, irrespective of the side where the stimulus was presented. Thus even if the responsiveness of TANs is traditionally considered to be more sensory than motor, our findings indicate that at least some TANs do participate in the processing of motor-dependent spatial information. This suggests that TANs may have a role in specific aspects of motor control, such as intention to move the arm in a particular direction. Finally, it was of special interest in our study to find a third type of TAN responses that did not simply reflect the location of the stimulus or the direction of the movement, but the interaction of these two aspects in a condition demanding a particular use of sensory information for the selection of an action. Thus 25% of the TANs displayed specific response properties only under certain circumstances. For example, some neurons responded with a spatial preference for one stimulus location in the choice condition, but reversed or lost their preference when tested in the dissociated condition. Such responses, characterized as context dependent, were interpreted to reflect the way in which sensory information is used to elicit reaching movements. In the choice condition, monkeys were required to detect the spatial information contained in the stimulus to make movements directed toward spatially distinct targets. In the dissociated condition, the stimulus provided no information other than a temporal reference for reaching toward a constant spatial target, regardless of the location of the stimulus. It is conceivable that TANs have a role in switching between conditions involving a different use of the information contained in the stimulus. Our findings are consistent with a role of TANs in signaling changes in context when stimuli are relevant as trigger signals for action. Such an interpretation is in accordance with the report of Shimo and Hikosaka (2001) showing that TANs respond selectively to visual stimuli in a context-dependent manner, being influenced by changes in reward probability. The response of TANs, according to these authors, may be seen as signaling the act of switching between different contexts involving specific reward schedules. Other evidence, recently obtained by Yamada et al. (2004), point to the fact that differences in the context in which task stimuli are presented might contribute to the differential responsiveness of TANs. In addition, these authors reported that TANs in the caudate nucleus may participate in the process of context encoding, whereas those in the putamen may carry information about movement initiation. This suggests that TAN responses associated with the processing of contextual and motor types of information may show some differences in their localization.
Possible origins of the sensitivity of TANs to spatial features of a task
It is known that different sensory inputs converge on TANs, from the midbrain dopaminergic neurons, the thalamus, and the cerebral cortex. Although the response of TANs to stimuli has been reported to be dependent on dopaminergic afferents (Aosaki et al. 1994; Raz et al. 1996; Watanabe and Kimura 1998), a differential responsiveness similar to the spatial selectivity of TAN responses reported here has not been observed in dopaminergic neurons (Schultz 1998), which suggests that TANs are more concerned than dopaminergic neurons with information-processing mechanisms necessary for the organization of spatially directed behavior. On the other hand, many of the neurons recorded in the centromedian/parafascicular complex of the thalamus respond to movement-triggering stimuli and these responses have been associated with visuospatial processes, including spatial attention (Minamimoto and Kimura 2002). Because this thalamic area has been reported to be essential for the correct functioning of TANs (Matsumoto et al. 2001), it is possible that such thalamic neuronal responses provide spatial information to the striatum.
Another possible source of afferents that underlies TAN responses related to spatial features of the task is the cerebral cortex. Recently, TANs in the putamen have been found to receive inputs from both the primary motor cortex and supplementary motor area (Nambu et al. 2002) in which neuronal responses that reflect either the stimulus location or the direction of the movement have been recorded (Alexander and Crutcher 1990b; Crutcher et al. 2004; Shen and Alexander 1997). Spatial information may also reach the striatum through projections from the dorsolateral prefrontal cortex and posterior parietal areas that contain neurons that are strongly related to visuospatial functions (Colby and Goldberg 1999; Funahashi et al. 1990; Steinmetz and Constantinidis 1995). Thus inputs from functionally distinct regions of the frontal and parietal visuospatial processing areas could mediate the spatial information used by TANs for reaching behavior. However, cortical inputs impose on the striatum a functional organization so that the structure can be divided into a sensorimotor territory and an associative territory (Parent and Hazrati 1995). In the present study, TANs from these two striatal territories were compared to assess regional differences in response properties and no striking differences in the anatomical distribution of TANs with spatial sensitivity appear in our data. In particular, the finding that TAN responses that were dependent on movement direction did not tend to be clustered in the posterior putamen is not consistent with the existence of a segregated population of TANs that would be predominantly "motor." As mentioned earlier, a differential distribution in the response properties of TANs was reported by Yamada et al. (2004), neurons responding to movement-triggering stimuli being more frequent in the putamen than in the caudate nucleus. Further studies are therefore necessary to establish whether contrasting properties in TAN responses exist among striatal regions and to clarify the relationship between topographic mapping of corticostriatal projections and TAN responses associated with the processing of different types of information.
Influence of the spatial features of a task on the activity of striatal projection neurons
Striatal neurons that discharge spontaneously at low rates and exhibit distinct activity changes in various phases of a learned task are referred to as phasically active neurons (PANs), which correspond to GABAergic medium spiny projection neurons in the striatum (Alexander and DeLong 1985; Wilson and Groves 1981). There is evidence that certain types of task-related PANs are more represented in particular striatal territories according to the convergence of cortical inputs on segregated regions of the striatum. In particular, PANs that are involved in movement execution are largely found in the posterior putamen, whereas neurons associated with more cognitive aspects of motor control, such as preparation for movements, display an increasing gradient from the posterior to anterior regions of the putamen (Alexander and Crutcher 1990a). Previous studies showed the spatial selectivity of PANs in that part of putamen associated with movement execution (Crutcher and Alexander 1990; Crutcher and DeLong 1984; Kimura 1990; Ueda and Kimura 2003). Only one study of PAN activity has attempted to dissociate the spatial features of the movement-triggering stimulus from those of the associated movement (Alexander and Crutcher 1990b) and it was found that spatially selective preparatory and/or movement-related activations can be related to the direction of movement, independent of stimulus location or to the spatial location of the stimulus, independent of the movement. These neurons thus share some features in common with the response properties of TANs we observed. Although the role that PANs play with respect to TAN activity has not been clearly defined, it is possible, as suggested by Shimo and Hikosaka (2001), that the sensitivity of TANs to the spatial aspects of the task may be related to certain specificities in the relation of PAN activity to the spatial location of the stimulus or the intended movement direction. However, it is not clear to what extent the response properties of TANs depend on cortical inputs or to what extent they are elaborated by the properties of surrounding PANs.
Functional implications
Our study provides new information about response properties of TANs in the primate striatum. These neurons, presumed to be cholinergic interneurons, occupy a central position within the striatal circuitry, being primarily devoted to the detection of motivationally relevant stimuli (Apicella 2002). The present findings that distinct groups of TANs are sensitive to the location of the stimulus, the direction of movement, or the context in which stimulus–movement combination occurs suggest that these neurons are not limited to motivational functions, but may play a role in the processing of information about the spatial features of task performance as well. Theoretically, the motivational relevance of a stimulus appears to be the primary factor in determining the response of TANs to it. The sensitivity of these neurons to spatial features may reflect a role in integrating motivational information with specific aspects of goal-directed actions toward targets, including the detection of stimulus location, the selection of the required movement, and the detection of the context of action. The response properties of TANs could be very important functionally regarding the monkey's ability to adapt to the changed requirements of a reaching task. We suggest that the striatal network of the TANs may serve to integrate reward- and spatial-related information, which in turn can affect the motor output system of the striatum to achieve appropriate spatially oriented behavior.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: P. Apicella, Laboratoire de Neurobiologie de la Cognition, CNRS, Centre Saint-Charles, 3 place Victor Hugo, 13331 Marseille cedex 3, France (E-mail: paul.apicella{at}up.univ-mrs.fr)
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Alexander GE and Crutcher MD. Neural representations of the target (goal) of visually guided arm movements in three motor areas of the monkey. J Neurophysiol 64: 164–178, 1990b.
Alexander GE and DeLong MR. Microstimulation of the primate neostriatum. II. Somatotopic organization of striatal microexcitable zones and their relation to neuronal response properties. J Neurophysiol 53: 1417–1430, 1985.
Aosaki T, Graybiel AM, and Kimura M. Effect of the nigrostriatal dopamine system on acquired neural responses in the striatum of behaving monkeys. Science 265: 412–415, 1994a.
Aosaki T, Tsubokawa H, Ishida A, Watanabe K, Graybiel AM, and Kimura M. Responses of tonically active neurons in the primate's striatum undergo systematic changes during behavioral sensorimotor conditioning. Neuroscience 14: 3969–3984, 1994b.[Abstract]
Apicella P. Tonically active neurons in the primate striatum and their role in the processing of information about motivationally relevant events. Eur J Neurosci 16: 2017–2026, 2002.[CrossRef][ISI][Medline]
Apicella P, Legallet E, and Trouche E. Responses of tonically discharging neurons in the monkey striatum to primary rewards delivered during different behavioral states. Exp Brain Res 116: 456–466, 1997.[CrossRef][ISI][Medline]
Apicella P, Schultz W, and Scarnati E. Tonically discharging neurons of monkey striatum respond to preparatory and rewarding stimuli. Exp Brain Res 84: 672–675, 1991.[ISI][Medline]
Blazquez PM, Fujii N, Kojima J, and Graybiel AM. A network representation of response probability in the striatum. Neuron 33: 973–982, 2002.[CrossRef][ISI][Medline]
Colby CL and Goldberg ME. Space and attention in parietal cortex. Annu Rev Neurosci 22: 319–349, 1999.[CrossRef][ISI][Medline]
Crutcher MD and Alexander GM. Movement-related neuronal activity selectively coding either direction or muscle pattern in three motor areas of the monkey. J Neurophysiol 64: 151–163, 1990.
Crutcher MD and DeLong MR. Single cell studies of the primate putamen. II. Relations to direction of movement and pattern of muscular activity. Exp Brain Res 53: 244–258, 1984.[ISI][Medline]
Crutcher MD, Russo GS, Ye S, and Backus DA. Target-, limb-, and context-dependent neural activity in the cingulate and supplementary motor areas of the monkey. Exp Brain Res 158: 278–288, 2004.[ISI][Medline]
Fisk JD and Goodale MA. The organization of eye and limb movements during unrestricted reaching to targets in contralateral and ipsilateral visual space. Exp Brain Res 60: 159–178, 1985.[ISI][Medline]
Funahashi S, Bruce CJ, and Goldman-Rakic PS. Visuospatial coding in primate prefrontal neurons revealed by oculomotor paradigm. J Neurophysiol 63: 814–831, 1990.
Haber SN and McFarland NR. The concept of the ventral striatum in nonhuman primates. Ann NY Acad Sci 877: 33–48, 1999.
Hikosaka O, Sakamoto M, and Usui S. Functional properties of monkey caudate neurons. I. Activities related to saccadic eye movements. J Neurophysiol 61: 780–798, 1989.
Kimura M. The role of primate putamen neurons in the association of sensory stimuli with movement. Neurosci Res 3: 436–443, 1986.[CrossRef][Medline]
Kimura M. Behaviorally contingent property of movement-related activity of the primate putamen. J Neurophysiol 63: 1277–1296, 1990.
Kimura M. Behavioral modulation of sensory responses of primate putamen neurons. Brain Res 578: 204–214, 1992.[CrossRef][ISI][Medline]
Kimura M, Rajkowski J, and Evarts EV. Tonically discharging putamen neurons exhibit set dependent responses. Proc Natl Acad Sci USA 81: 4998–5001, 1984.
Matsumoto N, Minamimoto T, Graybiel AM, and Kimura M. Neurons in the thalamic CM-Pf complex supply striatal neurons with information about behaviorally significant sensory events. J Neurophysiol 85: 960–976, 2001.
Minamimoto T and Kimura M. Participation of the thalamic CM-Pf complex in attentional orienting. J Neurophysiol 87: 3090–3101, 2002.
Nambu A, Kaneda K, Tokuno H, and Takada M. Organization of corticostriatal motor inputs in monkey putamen. J Neurophysiol 88: 1830–1842, 2002.
Parent A and Hazrati LN. Functional anatomy of the basal ganglia. I. The cortico-basal ganglia-thalamo-cortical loop. Brain Res Rev 20: 91–127, 1995.[CrossRef][Medline]
Ravel S, Legallet E, and Apicella P. Tonically active neurons in the monkey striatum do not preferentially respond to appetitive stimuli. Exp Brain Res 128: 531–534, 1999.[CrossRef][ISI][Medline]
Ravel S, Legallet E, and Apicella P. Responses of tonically active neurons in the monkey striatum discriminate between motivationally opposing stimuli. Neuroscience 23: 8489–8497, 2003.
