J Neurophysiol 92: 2261-2273, 2004;
doi:10.1152/jn.00085.2004
0022-3077/04 $5.00
Suppression of Visually and Memory-Guided Saccades Induced by Electrical Stimulation of the Monkey Frontal Eye Field. II. Suppression of Bilateral Saccades
Yoshiko Izawa,
Hisao Suzuki and
Yoshikazu Shinoda
Department of Systems Neurophysiology, Graduate School of Medicine, Tokyo Medical and Dental University, Yushima, Tokyo 113-8519, Japan
Submitted 28 January 2004;
accepted in final form 10 May 2004
 |
ABSTRACT
|
|---|
To understand the neural mechanism of fixation, we investigated effects of electrical stimulation of the frontal eye field (FEF) and its vicinity on visually guided (Vsacs) and memory-guided saccades (Msacs) in trained monkeys and found that there were two types of suppression induced by the electrical stimulation: suppression of ipsilateral saccades and suppression of bilateral saccades. In this report, we characterized the properties of the suppression of bilateral Vsacs and Msacs. Stimulation of the bilateral suppression sites suppressed the initiation of both Vsacs and Msacs in all directions during and 
50 ms after stimulation but did not affect the vector of these saccades. The suppression was stronger for ipsiversive larger saccades and contraversive smaller saccades, and saccades with initial eye positions shifted more in the saccadic direction. The most effective stimulation timing for the suppression of ipsilateral and contralateral Vsacs was 40–50 ms before saccade onset, indicating that the suppression occurred most likely in the superior colliculus and/or the paramedian pontine reticular formation. Suppression sites of bilateral saccades were located in the prearcuate gyrus facing the inferior arcuate sulcus where stimulation induced suppression at 
40 µA but usually did not evoke any saccades at 80 µA and were different from those of ipsilateral saccades where stimulation evoked saccades at 50 µA. The bilateral suppression sites contained fixation neurons. The results suggest that fixation neurons in the bilateral suppression area of the FEF may play roles in maintaining fixation by suppressing saccades in all directions.
|
INTRODUCTION
|
|---|
Visual fixation of a stationary object has been likened to smooth pursuit of a target moving at zero velocity (Yabus 1967
). However, recent findings suggest that a unique mechanism for visual fixation is involved if the object is not moving. Fixation neurons, which show continuous activity during fixation, have been found in the parietal cortex (Mountcastle et al. 1975), frontal eye field (FEF) (Bizzi 1968; Bruce and Goldberg 1985; Suzuki and Azuma 1977), superior colliculus (SC) (Munoz and Guitton 1991; Munoz and Wurtz 1993a), and omnipause neuron (OPN) region in the brain stem (Cohen and Henn 1972; Keller 1974; Luschei and Fuchs 1972). During steady fixation, the threshold for electrical stimulation to elicit saccades in the FEF (Goldberg et al. 1986) or SC (Sparks and Mays 1983) is elevated, suggesting that a visual fixation system may have suppressive effects on the generation of saccades. To investigate the function of these areas containing fixation neurons, microstimulation was performed in the FEF (Azuma et al. 1986; Burman and Bruce 1997), SC (Munoz and Wurtz 1993b; Paré and Guitton 1994) and OPN region (Keller 1977; King and Fuchs 1977). Stimulation of the OPN region interrupted saccades in all directions (Keller et al. 1996) and suppressed both visually guided (Vsacs) (Keller 1977; King and Fuchs 1977) and memory-guided (Msacs) saccades (Gandhi and Keller 1999). Stimulation of the most rostral part of the SC suppressed saccades in bilateral directions, with predominant suppression of ipsilateral saccades (Munoz and Wurtz 1993b; Paré and Guitton 1994), and suppressed both Vsacs and Msacs (Munoz and Wurtz 1993b). Burman and Bruce (1997) reported that stimulation of the FEF mainly suppressed contraversive saccades and effectively suppressed Msacs but not Vsacs, whereas our previous study showed that FEF stimulation suppressed the initiation of ipsiversive saccades of both Vsacs and Msacs at low thresholds (Izawa et al. 2004). Although the preceding discrepancies have not yet been resolved, these suppressive effects of FEF stimulation on the generation of saccades may contribute to maintaining fixation on a target of interest by preventing saccades to other targets in the visual field.
To understand the nature of FEF suppression, we systematically examined the suppressive effects of electrical stimulation of the FEF on saccades in monkeys and reported the unilateral suppression of ipsiversive saccades on FEF stimulation in the previous report (Izawa et al. 2004). Such suppression sites for ipsiversive saccades were distributed widely in the FEF where electrical stimulation could induce saccades at 50 µA. During this mapping of suppression sites, we also found that stimulation of a localized area in the FEF, where electrical stimulation of 80 µA usually could not evoke saccades, suppressed both Vsacs and Msacs in all directions at low intensities. In this report, we describe the characteristic features of this suppression of saccades in all directions, compare the properties of the unilateral and bilateral suppression, and discuss the functional significance of this bilateral FEF suppression area for fixation and saccadic generation.
These results have been briefly reported previously (Izawa et al. 2001).
|
METHODS
|
|---|
Experiments were performed in two male Japanese monkeys (Macaca fuscata) that were also used in the preceding article (Izawa et al. 2004). Monkeys were prepared for the experiments using the same procedures and the same series of fixation, visually guided, and memory-guided saccade tasks described in the preceding article (Izawa et al. 2004).
We studied sites in the prearcuate gyrus in the classical FEF, where electrical stimulation (50 µA) evoked saccadic eye movements, and its surrounding area, where electrical stimulation of 50 µA did not evoke saccades (see Figs. 1 and 3 for recording area), when monkeys were performing fixation. Constant-current stimulation trains consisted of 40–60 monopolar cathodal pulses of 80 µA, 1-ms duration at 5-ms intervals unless stated otherwise. In this experiment, we examined the suppressive effect of stimulation of the sites in the FEF on saccades at 80 µA even though stimulation of these sites was not effective for eliciting electrically evoked saccades (Esacs).
View larger version (23K):
[in this window]
[in a new window]
|
FIG. 1. Suppression of bilateral horizontal visually guided saccades (Vsacs) by frontal eye field (FEF) stimulation. Inset: stimulus location in the left FEF. Site 1 for A and B and site 2 for C. IAS, inferior arcuate sulcus; SAS, superior arcuate sulcus; PS, principal sulcus. A: electrically evoked saccades (Esacs) during fixation by stimulation of the FEF. · · · , onset of electrical stimulation. Top and bottom traces of individual pairs: horizontal and vertical components of eye position, respectively. Bottom-most trace: event marker indicating the train duration of stimulation. Note that saccades with small amplitudes were evoked in an all-or-none manner at 70 and 80 µA. B: delayed onsets of ipsi- and contraversive horizontal 10° Vsacs caused by FEF stimulation at 15 µA. Top and bottom pairs of traces in individual records indicate control Vsacs without stimulation and Vsacs during stimulation of the same FEF site as in A, respectively. · · · , visual target onset. C: effects of varying the stimulus intensity on ipsiversive (left) and contraversive 10° Vsacs (right). Stimulation (60 pulses at 5-ms interval) was applied at different stimulus intensities (indicated on the left) at visual target onset ( · · · ). D: relationship between thresholds for the suppression of ipsiversive (ipsi) and contraversive Vsacs (contra) in 2 monkeys (Sui and Bell). r, correlation coefficient.
|
|
View larger version (35K):
[in this window]
[in a new window]
|
FIG. 3. Stimulation and recording sites in the FEF to show suppression sites for ipsilateral and bilateral saccades. A: distribution of bilateral suppression sites in the FEF. Three planes of the mapping are shown. Single circles show tracks in which suppression of only ipsilateral Vsacs was observed, whereas double circles show tracks in which suppression of ipsilateral and bilateral Vsacs was observed. Arrow indicates a track in which depth-threshold relationship for suppressing bilateral Vsacs is shown in B and C. B: background neuronal activity of each stimulation depth shown in C. C: depth-threshold curves for eliciting Esacs (dotted lines) and suppressing ipsiversive (thin solid line) and contraversive Vsacs (thick solid line). ipsi, suppression of ipsiversive Vsacs; contra, suppression of contraversive Vsacs. D: activity of a fixation neuron observed at a bilateral suppression site in the FEF. Neuronal discharges in successive fixation trials are represented by raster. Histogram is the result of summing the dots on the raster in 50-ms bins. Horizontal and vertical components of eye position are shown above the raster. The raster and histogram are aligned with respect to bar-press onset (vertical dotted line). The 1st–4th short vertical lines below each raster line mark the appearance of the central fixation point, the brightening of the fixation point, bar release (reward onset), and reward offset, respectively.
|
|
Data acquisition and analysis were made in the same way as described before (Izawa et al. 2004).
|
RESULTS
|
|---|
Suppressive effects of FEF stimulation on bilateral Vsacs
As reported in the previous article (Izawa et al. 2004), we systematically examined the effects of stimulation of the FEF and its vicinity in the prearcuate gyrus at a track interval of 500 µm and at depth intervals of 200 or 400 µm. At individual stimulation sites, we first determined thresholds for eliciting Esacs by microstimulation and identified a site as being within the classical FEF when stimulation elicited Esacs at 50 µA (Bruce et al. 1985). In this experiment, we examined the effects of stimulation at the same sites on Vsacs and Msacs at 80 µA even though electrical stimulation did not evoke Esacs. We usually found that stimulation of a wide area within the FEF suppressed the initiation of only ipsiversive Vsacs and Msacs as described in the previous article (Izawa et al. 2004). However, during this mapping, we found that stimulation of some sites in the FEF elicited the suppression of not only ipsiversive but also contraversive saccades. A typical example of stimulation at such a site is shown in Fig. 1. In contrast to the typical FEF sites, this stimulation site in the FEF did not evoke any Esacs at 60 µA but evoked small Esacs (2.1°) with long latencies at 70 µA and with slightly shorter latencies at 80 µA in an all-or-none manner (Fig. 1A), indicating that these stimulus intensities were not sufficiently suprathreshold for Esacs. The suppressive effects on Vsacs were examined by stimulating the same site at 80 µA. Stimulation at 15 µA completely suppressed the generation of ipsiversive Vsacs during stimulation and for 50 ms after stimulation offset (Fig. 1B, ipsi). This suppression of ipsiversive Vsacs is very similar to the unilateral suppression of Vsacs reported in the previous paper (Izawa et al. 2004). Stimulation at the same site with the same stimulus parameters also suppressed contraversive Vsacs, and suppression continued for 10–50 ms after stimulation offset (Fig. 1B, contra). Sites at which bilateral Vsacs were suppressed at thresholds of <50 µA were found in 106 of the 279 tracks examined (monkey Sui: 35/136, monkey Bell: 71/143). At such bilateral suppression sites, Esacs were not usually observed at 80 µA, but very small Esacs with the amplitude of <4° were evoked at 
70 µA at 19 stimulation sites.
Stimulus parameters for effective suppression of bilateral Vsacs.
Using a procedure identical to that which we used to examine the effects of suppression on ipsiversive Vsacs (Izawa et al. 2004), we examined the effects of varying the stimulus intensity (Fig. 1C) and the number of stimulus pulses (Fig. 2) on the suppression of bilateral Vsacs. In the example in Fig. 1C, FEF stimulation at 10 µA slightly increased the latency of ipsiversive Vsacs, and stimulation at 15 µA completely suppressed the generation of ipsiversive Vsacs for up to 50 ms after stimulation offset. At the same stimulation site, contraversive Vsacs were also suppressed in some trials at 10 µA. With stimulus intensities of 15 and 20 µA, the suppression of contraversive Vsacs became strong and persisted for 50 ms after the offset of stimulation. However, the latencies of contraversive Vsacs fluctuated slightly even when suppression occurred with stronger stimulus intensities. This phenomenon was not observed for the suppression of ipsiversive Vsacs but was commonly observed for the suppression of contraversive Vsacs. The threshold for suppression was defined as the least current required to produce an increase in the saccade latency that was >2 SDs of the control saccade latency. The mean ± SD of the latency of control Vsacs were 181 ± 24 ms in monkey Sui (n = 261) and 188 ± 19 ms in monkey Bell (n = 245). Based on this criterion, we determined that the threshold for the suppression of both ipsiversive and contraversive Vsacs was 10 µA in the example in Fig. 1C, The thresholds for suppression were compared between ipsi- and contraversive Vsacs at all 106 sites at which the latencies of bilateral Vsacs were significantly increased by stimulation at <50 µA (Fig. 1D). The results showed that there was no significant difference in the suppression threshold between ipsi- and contraversive Vsacs (Wilcoxon signed-ranks test, P = 0.92, n = 106). A comparison of control and stimulation (20 µA) trials showed that the amplitudes of ipsiversive (U test, P = 0.53) and contraversive Vsacs (U test, P = 0.18) and the directions of ipsiversive (U test, P = 0.11) and contraversive Vsacs (U test, P = 0.66) were not significantly different (Fig. 1C). Accordingly, in contrast to the suppressed initiation of Vsacs, neither the amplitudes nor the directions of ipsiversive and contraversive Vsacs were affected by stimulation, indicating that the accuracy of Vsacs was maintained.
View larger version (20K):
[in this window]
[in a new window]
|
FIG. 2. Effects of varying the stimulus number on the suppression of ipsiversive (left) and contraversive 10° Vsacs (right). A train of 0–150 pulses (indicated on the left) (30 µA, 5-ms interval) was applied at visual target onset ( · · · ). Same stimulation site as in Fig. 1C.
|
|
The effects of varying the number of stimulus pulses were examined with a stimulus intensity fixed at a suprathreshold value of 15 µA at the same stimulation site as in Fig. 1C (Fig. 2). An increase in the number of stimulus pulses increased the latencies of ipsi- and contraversive Vsacs in a similar manner (y = 0.99x + 20, r = 0.97, P < 0.001, n = 99). Similar results were obtained at all 10 bilateral suppression sites examined. However, the suppression of Vsacs did not always persist throughout the entire duration of a stimulus train. This tendency was also observed in some trials for the stimulation of unilateral suppression sites (Izawa et al. 2004). However, for the stimulation of bilateral suppression sites, this fluctuation in the suppressive effect was observed more often for contraversive Vsacs.
LOCATION OF BILATERAL SUPPRESSION SITES DEFINED BY DEPTH-THRESHOLD CURVES FOR ELICITING ESACS AND SUPPRESSING VSACS AND NEURONAL ACTIVITY OF FIXATION NEURONS.
To identify the location of bilateral suppression sites in the FEF, we systematically examined depth-thresholds for suppressing bilateral Vsacs in individual tracks. Figure 3C shows a typical example of depth-threshold curves for eliciting Esacs (dotted lines), suppressing ipsiversive Vsacs (thin solid line) and contraversive Vsacs (thick solid line) at a cortical site indicated by an arrow in Fig. 3A. Near the surface of the FEF, Esacs were evoked at 40 µA, and their thresholds decreased to a minimum at depths of 1,200 and 1,600 µm. The thresholds then gradually increased as the electrode was advanced to 2,400 µm deep. Suppression of ipsiversive Vsacs also occurred at the same depths as Esacs, but the thresholds for suppression were lower by 3–17 µA. At 2,800–4,000 µm, Esacs could not be evoked even at 80 µA, but ipsiversive Vsacs were still suppressed at 24 µA. In addition, the suppression of contraversive Vsacs appeared at 36 µA at 2,800 µm, and as the depth increased, the threshold for this contralateral suppression decreased and reached the minimum of 24 µA at 4,000 µm. At 4,400–5,200 µm, Esacs again appeared. As in this example, the bilateral suppression sites were usually found deeper than the unilateral suppression sites at the superficial depths along each track and were adjacent to or sometimes within the marginal area for evoking Esacs at 50 µA. At the depths where the bilateral suppression occurred, background activity was very high in every track (Fig. 3B), indicating that such bilateral suppression sites were located in the cerebral gray matter. In the bilateral suppression sites, we recorded single units and examined their neural activity in relation to visual stimuli, visual fixation, and saccadic eye movements. This study is still going on, but so far we have encountered many fixation neurons. Figure 3D shows a typical example of activity of a neuron observed at such a bilateral suppression site. This neuron discharged during fixation, so that it was classified as a fixation neuron (Suzuki and Azuma 1977). Even when the fixation spot disappeared for 400 ms during steady fixation, this neuron did not decrease its activity, indicating that this activity is not foveal visual-related. The distribution of bilateral suppression sites (double circles) and ipsilateral suppression sites (single circles and double circles) in three planes of the mapping was shown in Fig. 3A. The bilateral suppression sites were distributed at the caudal part of the arcuate gyrus facing the inferior arcuate sulcus and were adjacent to or involved in the classical FEF where Esacs were evoked at low intensities 50 µA.
SUPPRESSIVE EFFECTS OF FEF STIMULATION ON VSACS WITH DIFFERENT DIRECTIONS.
To determine the effective direction of suppression, we examined the suppressive effects of FEF stimulation on Vsacs by changing the direction of Vsacs to one of eight directions with their amplitude (10°) kept constant. The strong suppression of both ipsiversive (Fig. 4A, a–c) and contraversive (Fig. 4A, e–g) Vsacs in any direction was observed in all trials. Purely vertical upward (Fig. 4Ah) and downward Vsacs (Fig. 4Ad) were also sufficiently suppressed by stimulation. Latencies of delayed Vsacs in eight directions indicated that stimulation of this FEF site significantly suppressed Vsacs (Friedman ANOVA, P < 0.0001; Fig. 4B). When the latency difference between Vsacs in control and stimulation trials was used as an index of the strength of the suppression in each direction (dotted line in Fig. 4B), the latency differences were not significantly different in eight directions (Kruskal-Wallis ANOVA, P = 0.43), indicating that this suppression was uniform in all directions. Similar results were obtained at all 20 bilateral suppression sites examined. Stimulation of this bilateral suppression site delayed both the horizontal and vertical components of individual Vsacs in parallel (Fig. 4A), and the correlation between the latencies of horizontal and vertical components of individual oblique Vsacs was significant (y = 1.05x – 17.46, r = 0.97, P < 0.001, n = 24). The slope and intercept of the regression line was not significantly different from 1 and 0 (t-test, P = 0.34 and 0.36, respectively), indicating that both components of Vsacs were synchronously suppressed by stimulation.
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 4. FEF suppression of Vsacs with different directions. A: Vsacs to 10° visual targets in 8 directions. a–h: Vsacs to target directions (a–h) in B, which correspond to directions of 135, 180, 225, 270, 315, 0, 45, and 90°, respectively. Top 2 and bottom 2 sets of traces in individual records (a–h) indicate horizontal and vertical components of eye position in the control and during stimulation (40 µA, 60 pulses), respectively. At this stimulation site, Esacs were not evoked even at 90 µA. B: polar representation of median latencies of Vsacs in control (thin line) and stimulation trials (thick line) shown in A. Dotted line shows the latency difference between Vsacs in control and stimulation trials in each direction.
|
|
FEF SUPPRESSION OF BILATERAL VSACS WITH DIFFERENT AMPLITUDES.
Stimulation of a bilateral suppression site had differential effects on Vsacs with different amplitudes. While stimulus parameters were kept constant, the amplitudes of bilateral Vsacs were varied from 5 to 20° at 5° intervals (Fig. 5A). In control trials with no stimulation, the latency of Vsacs decreased as the amplitude increased (Kruskal-Wallis ANOVA, P < 0.05). In stimulation trials for ipsiversive Vsacs, the latencies of Vsacs increased as the stimulus intensities were increased (Friedman ANOVA, P < 0.0001; Steel method, P < 0.05 for 18 µA). Stimulation at 24 µA suppressed 10–20° Vsacs in some trials for up to 50 ms after stimulation offset, and larger saccadic amplitudes were associated with a greater delay in saccadic onset (Fig. 5B, · · · ). At 30 µA, ipsiversive Vsacs of all amplitudes were completely suppressed for up to 50 ms after stimulation offset. In stimulation trials for contraversive Vsacs, the latencies of Vsacs increased as the stimulus intensities were increased (Friedman ANOVA, P < 0.0001; Steel method, P < 0.05 for 24 µA). Stimulation at 24 µA strongly suppressed 5° Vsacs and weakly 10° and 15° Vsacs, so that smaller saccadic amplitudes were associated with a greater delay in saccadic onset (Fig. 5B, · · · ). Stimulation at 30 µA suppressed contraversive Vsacs of all amplitudes but less constantly those with larger amplitudes (Fig. 5B). Similar results were obtained at the other nine bilateral suppression sites tested: larger ipsiversive and smaller contraversive Vsacs were suppressed at weaker stimulus intensities, although both ipsiversive and contraversive Vsacs were strongly suppressed at sufficiently strong stimulus intensities.
View larger version (18K):
[in this window]
[in a new window]
|
FIG. 5. Suppressive effects of FEF stimulation on bilateral Vsacs with different amplitudes. A: horizontal 5°–20° Vsacs from the center in rightward (R) and leftward (L) directions. Stimulation of the left FEF (40 pulses) was applied at 70 ms after visual target onset ( · · · ) at different intensities as indicated at the top. B: latencies of horizontal 5°–20° Vsacs during FEF stimulation at different stimulus intensities including those shown in A. Abscissa, target position of Vsacs; Ordinate, medians and quartiles of Vsac latencies.
|
|
FEF SUPPRESSION OF BILATERAL VSACS WITH DIFFERENT INITIAL EYE POSITIONS.
Shifting the initial eye position strongly affected the latencies of Vsacs in bilateral suppression. The latencies of Vsacs in control trials increased as the initial eye position shifted from the center in the same direction as the Vsac direction (Fig. 6, A and B), and those with the initial eye positions of 15° were significantly longer than those of Vsacs with the central initial eye position (Steel method, P < 0.05). In stimulation trials for ipsiversive Vsacs (Fig. 6, A and C), Vsacs were more delayed with the initial eye position shifted toward the Vsac direction, and an increase of Vsac latencies was saturated as the stimulus intensity was increased (Friedman ANOVA, P < 0.0001; Steel method, P < 0.05 for 35 and 42 µA). For contraversive Vsacs, stimulation of the same site completely suppressed contraversive Vsacs with the initial eye position shifted from the center to the contralateral side (R), but the delay of contraversive Vsacs with the initial eye position shifted to the ipsilateral side (L) tended to decrease (Fig. 6, B and D). Because a similar result was obtained at the other bilateral suppression site examined, it was concluded that the FEF suppression of both ipsiversive and contraversive Vsacs increased as the initial eye position shifted toward the direction of Vsacs.
View larger version (28K):
[in this window]
[in a new window]
|
FIG. 6. Suppressive effects of FEF stimulation on bilateral Vsacs with different initial eye positions. A and B: horizontal ipsiversive (A) and contraversive 10° Vsacs (B) whose initial eye positions were varied at 5° intervals from the center to 20° ipsilaterally (L) or contralaterally (R). Top and bottom traces in individual pair records indicate horizontal eye position in control and stimulation trials, respectively. A train of 40 pulses was applied at 35 µA after 70-ms delay from visual target onset. C and D: latencies of ipsiversive 10° Vsacs (including those shown in A; C) and contraversive 10° Vsacs (including those in B; D) as a function of the initial eye position in the control and during FEF stimulation. Abscissa, initial eye position; ordinate, medians and quartiles of Vsac latencies.
|
|
EFFECTIVE STIMULUS TIMING FOR SUPPRESSING BILATERAL VSACS.
To determine the latest effective timing of FEF stimulation on Vsac suppression, we decreased the interval between visual target onset and stimulation onset from 160 to 90 ms (Fig. 7, Aa and Ba). The train duration was 200 ms, and the stimulus intensity was fixed at 40 µA. The median latency of control ipsiversive Vsacs with no stimulation was 176 ms (Fig. 7Aa). The suppressive effect was not observed at intervals of 160 ms. For intervals from 150 to 145 ms, the number of trials with suppressed ipsiversive Vsacs increased. At intervals of 140 ms, the initiation of ipsiversive Vsacs was delayed in almost all trials for up to 50 ms after stimulation offset. This time course of suppression was very similar to that observed for unilateral suppression sites in the FEF (Izawa et al. 2004). The suppression of contraversive Vsacs was induced by stimulation of the same site with the same stimulus parameters (Fig. 7Ba). The median latency of control contraversive Vsacs without stimulation was 171 ms. For contraversive Vsacs, stimulation at 150 ms increased the latencies of some Vsacs relative to those in control trials by 70 ms, and complete suppression, which continued for up to 60 ms after stimulus offset, first appeared at 145 ms (Fig. 7Ba). A further decrease in interval increased the frequency of such trials with complete suppression, and complete suppression was observed in most trials at intervals of 135 ms.
View larger version (39K):
[in this window]
[in a new window]
|
FIG. 7. Effects of varying the timing and pulse number of FEF stimulation on the suppression of bilateral Vsacs. Aa and Ba: effects of suppression on horizontal 10° Vsacs by varying the interval between target onset (dotted line) and stimulus train onset (thick bar; 40 µA, 40 pulses at 5-ms intervals). Ab and Bb: effects of increasing the number of stimulus pulses (thick bar; 40 µA, 5 ms intervals) on the suppression of horizontal 10° Vsacs. A train of stimuli started at target onset (dotted line). Timing and number of stimulus pulses are indicated on the left. Insets: time courses of the suppression of ipsiversive (A, a and b) and contraversive Vsacs (B, a and b) during FEF stimulation. Ordinate, median latencies and quartiles of Vsacs; abscissa, interval from target onset to stimulation onset for Aa and Ba, and train duration [(number of stimulus pulses – 1) multiplied by the 5-ms interval] for Ab and Bb. Thin horizontal line, median latency of control Vsacs.
|
|
To further understand where this suppression of Vsacs occurs in the pathway for generating Vsacs, we determined the earliest timing for the effective suppression of Vsacs in the second method. In the second method (Fig. 7, Ab and Bb), stimulus train onset was fixed at target onset and the train duration was increased by 5 or 10 ms by increasing the number of stimulus pulses so that the earliest effective timing for suppression could be determined. The initiation of ipsiversive Vsacs was gradually delayed with an increase in the pulse number and clearly delayed at a train duration of 215 ms (44 pulses). The latency of the suppressed Vsacs was significantly longer at 135 ms (28 pulses) or more than in the control (Kruskal-Wallis ANOVA, P < 0.0001; Steel method, P < 0.05; plot in Fig. 7Ab) and was slightly shorter than the timing (140–145 ms) determined by the first method. An increase in the number of stimulus pulses gradually delayed the initiation of contraversive Vsacs in the same manner as for ipsiversive Vsacs (Fig. 7Bb), and the delayed latency was significantly longer at 130 ms (27 pulses) or more than in the control (Kruskal-Wallis ANOVA, P < 0.0001; Steel method, P < 0.05). The time courses of FEF suppression on ipsiversive and contraversive Vsacs showed that the latest effective timing for suppression was 140 ms for ipsiversive Vsacs and 135 ms for contraversive Vsacs, and the earliest effective timing for suppression was 135 ms for ipsiversive Vsacs and 130 ms for contraversive Vsacs. Similar time courses were observed at the other three bilateral suppression sites examined.
Suppressive effects of FEF stimulation on bilateral Msacs
We compared the suppressive effects of stimulation of the bilateral suppression area on Vsacs and Msacs because different neural pathways might be responsible for the generation of Vsacs and Msacs. We first identified a bilateral suppression site in the FEF for Vsacs. In the example in Fig. 8A, stimulation at 20 µA delayed the generation of bilateral Vsacs, and stimulation at 40 µA completely suppressed this generation. When the same stimulation was applied at the offset of the central fixation point, which was a cue for the initiation of Msacs, bilateral Msacs were delayed at 20 µA and completely suppressed during stimulation at 40 µA (Fig. 8B). The thresholds for the suppression of bilateral Vsacs and Msacs were 20 µA at this stimulation site. As in this example, the thresholds for the suppression of both ipsi- and contraversive saccades were usually very similar for Msacs (y = 0.87x + 2.27, r = 0.84, P < 0.01, n = 8; Fig. 8D) as well as for Vsacs (y = 0.90x +3.04, r = 0.87, P < 0.01, n = 8; Fig. 8C), and the thresholds for the suppression of both Vsacs and Msacs were also very similar for ipsiversive (y = 0.84x + 4.72, r = 0.83, P < 0.01, n = 8; Fig. 8E) and contraversive saccades (y = 0.88x +1.56, r = 0.85, P < 0.01, n = 8; Fig. 8F) at identical stimulation sites. In addition, the depth-threshold curves for the suppression of Msacs and Vsacs were usually similar along individual penetration tracks (not shown). However, at some stimulation sites, these thresholds were slightly different from each other. Among eight stimulation sites tested, the threshold difference for the suppression of Vsacs and Msacs was 10 µA for ipsiversive saccades (n = 7; identical, 6; lower for Vsacs, 1) and contraversive saccades (n = 7; identical, 5; lower for Vsacs, 1; lower for Msacs, 1). The threshold difference was between 10 and 20 µA for ipsiversive saccades (n = 1) and contraversive saccades (n = 1). In these two cases with a large threshold difference, the thresholds for the suppression of Msacs were much lower than those for the suppression of Vsacs.
View larger version (21K):
[in this window]
[in a new window]
|
FIG. 8. Suppressive effects of FEF stimulation on bilateral Vsacs and memory-guided saccades (Msacs). A: suppression of ipsiversive and contraversive horizontal 10° Vsacs by stimulation (60 pulses) at increasing stimulus intensities (indicated on the left). B: suppression of ipsiversive and contraversive horizontal 10° Msacs induced by stimulation of the same site in the FEF as in A. C and D: relationship between thresholds for the suppression of ipsiversive and contraversive Vsacs (C) and Msacs (D) at individual stimulation sites. E and F: relationship between thresholds for the suppression of ipsiversive (E) and contraversive (F) Vsacs and Msacs at individual stimulation sites.
|
|
To examine the preferred directions of suppression, we compared the suppressive effects of FEF stimulation at identical stimulation sites on Vsacs and Msacs in eight different directions. The amplitudes of Vsacs and Msacs were fixed at 10°, and the stimulus intensity and duration were fixed at 35 µA and 300 ms, respectively. We first identified a bilateral suppression site for Vsacs and then examined the effects of stimulation at the same site on Vsacs (Fig. 9A) and Msacs in eight directions (Fig. 9B). Both Vsacs and Msacs were completely suppressed in all eight directions. Plots of the median latencies of delayed Vsacs (Fig. 9C) and Msacs (Fig. 9D) showed that stimulation at this FEF site clearly suppressed both Vsacs and Msacs in all directions (Friedman ANOVA, P < 0.0001). At all five bilateral suppression sites for Vsacs examined, Msacs were also suppressed in all eight directions. To investigate the effects of FEF stimulation on vectors of Msacs, we compared the endpoints of saccades in control and stimulation trials in Fig. 9. The endpoints of Vsacs to 10° visual targets constantly converged to targets in all eight directions in the control. Even when stimulation strongly suppressed generation of Vsacs, the endpoints of Vsacs in stimulation trials almost overlapped those in control trials. The differences in the amplitude and direction of Vsacs between control and stimulation trials were not significant in any direction (U test, P > 0.05 and P > 0.05, respectively, for each target location). In contrast, Msacs showed greater fluctuation and elevation in their endpoints of eye position in the control than Vsacs, but the endpoints of Msacs in stimulation trials almost overlapped those in control trials, indicating that the amplitude and direction of Msacs were maintained in all eight directions (U test, P > 0.05 and P > 0.05, respectively, for each target location). As in this example, the vectors of Msacs were not affected by stimulation at most stimulation sites (U test, P > 0.05, n = 6), except for two sites where the amplitudes of Msacs tended to be slightly hypometric (U test, P < 0.05, n = 2), with little or no change in their directions.
View larger version (27K):
[in this window]
[in a new window]
|
FIG. 9. Suppression caused by FEF stimulation of Vsacs and Msacs in different directions. A and B: Vsacs (A) and Msacs (B) to 10° visual targets in 8 directions in the control (top 2 traces in a–h) and during stimulation of the identical site in the left FEF (35 µA, 60 pulses; bottom 2 traces in a–h). Same arrangement as in Fig. 4. C and D: polar representation of median latencies in control (thin line) and stimulation (thick line) trials of Vsacs (C) and Msacs (D) shown in A and B, respectively.
|
|
To determine the timing for the suppression of Msacs, we changed the interval between stimulation onset and disappearance of the central fixation point (cue to start an Msac) from 170 to –1,000 ms at one stimulation site (Fig. 10). Negative values indicate that stimulation onset preceded the disappearance of the central fixation point. The interval of –1,000 ms corresponds to the disappearance of an instruction target. The stimulus intensity and the number of stimulus pulses were fixed at 60 µA and 60 pulses (295-ms duration), respectively. With a decrease in the interval from 170 ms, the suppressive effect on ipsiversive Msacs became prominent in almost all trials at intervals of 130 ms, and the latencies of Msacs recovered to the level in the control with a time course that depended on the train duration (Fig. 10B). This time course of the suppression of Msacs was very similar to that for the suppression of Vsacs. Because the median latency of ipsiversive Msacs was 170 ms, the latest effective timing corresponds to 40 ms before Msac onset. The period from –200 to –1,000 ms in the Msac task corresponded to the memory period of the target position, during which stimulation applied did not affect the suppression of the generation of Msacs. However, during the period from –150 ms or later, the latencies of Msacs gradually increased significantly (Kruskal-Wallis ANOVA, P < 0.0001; Steel method, P < 0.05), which means that the earliest effective train onset of 295-ms duration should be between –200 and –150 ms. Therefore the earliest effective timing was between 95 (–200 ms) and 145 ms (–150 ms) after the disappearance of the fixation point (see Fig. 7, Ab and Bb).
View larger version (22K):
[in this window]
[in a new window]
|
FIG. 10. Effects of the timing of FEF stimulation on the suppression of Msacs. A: suppression of ipsiversive 10° Msacs caused by changing the interval (indicated on the left) between fixation target offset (right dotted line) and the onset of a train of stimuli (thick bars; 60 µA, 60 pulses). Each trace started at the onset of an instruction target flash, and the left dotted line indicates its offset. A positive interval indicates that the stimulation onset followed fixation target offset as a cue to start Msacs (right dotted line). B: time course of the suppression of ipsiversive Msacs during FEF stimulation. Median latencies and quartiles of Msacs (ordinate) were plotted as a function of stimulation onset relative to fixation light offset (abscissa). Long arrow and thin horizontal line, median latency of control Msacs (170 ms).
|
|
Relationship between FEF suppression and the refractoriness of saccades
It is known that a saccade is followed by a refractory period for a saccade in the same direction as the preceding saccade (Robinson and Fuchs 1969). Therefore it is possible that stimulation of a bilateral suppression site evokes unnoticeable Esacs whose refractory period might cause the suppression of subsequent contraversive Vsacs. The amplitude of Esacs depends on the initial eye position even when they are evoked by stimulation with identical stimulus parameters. Figure 11A shows an example of the effect of changing the initial eye position on the amplitude of contraversive Esacs. Stimulation of a classical left FEF site produced 19.8° Esacs with only a horizontal component from the center initial eye position. As the initial eye position was shifted toward the direction opposite the Esacs (leftward), the amplitude of Esacs gradually increased and became 30.2° at an initial eye position of left 20°. As the initial eye position was shifted toward the direction of the Esacs (rightward), the amplitude of Esacs gradually decreased and became 6.0° at an initial eye position of right 20° (Fig. 11A). As in this example, the amplitude of Esacs greatly increased as the initial eye position changed toward the direction opposite the Esacs. Taking advantage of this property, we examined the possibility that stimulation of a bilateral suppression site might produce tiny Esacs with an amplitude of <0.3°, which was below the sensitivity of our eye measuring system. Stimulation of a bilateral suppression site suppressed bilateral Vsacs at a threshold of 24 µA but did not evoke any identifiable Esacs at the central fixation position at 42 µA. Even though the initial eye position was shifted to the direction opposite the Esacs, Esacs could not be identified with our recording system (Fig. 11B). This result supports the interpretation that the suppression of contraversive Vsacs and Msacs cannot be attributed to the refractory period of tiny contraversive Esacs.
View larger version (12K):
[in this window]
[in a new window]
|
FIG. 11. Effects of varying the initial eye position on Esacs elicited by stimulation of the classical FEF. A: Esacs evoked from different initial eye positions by stimulation of the classical FEF (40 µA, 40 pulses) during the fixation period. Initial eye positions were varied at 5° intervals from the center to 20° ipsilaterally (L) or contralaterally (R). Top and bottom traces in individual pair records indicate horizontal and vertical eye positions, respectively. Dotted horizontal lines, primary eye position. Dotted vertical lines, onset of stimulation. Note that the Esac amplitude increased as the initial eye position was shifted toward the direction opposite the Esacs. B: stimulation (42 µA, 40 pulses) of a bilateral suppression site did not evoke any Esacs even by varying the initial eye position. The threshold for the bilateral suppression of Vsacs at this stimulation site was 24 µA. Same arrangement as in A. Bottom traces, train duration for A and B.
|
|
The interaction between Esacs and Vsacs was examined at 12 classical FEF sites to further exclude the possibility that a refractory period after a small contraversive Esac might cause the suppression of contraversive Vsacs. It was not easy to examine the effects of the refractoriness of Esacs on the generation of Vsacs because Esacs evoked during the central fixation period return to the initial eye position by return saccades (Goldberg et al. 1986). Accordingly, we examined the effects of preceding Vsacs on Esacs. Changes in the latency and amplitude of Esacs were analyzed by varying the onset of Vsacs relative to the stimulation onset for Esacs (Fig. 12A). The shortest period for evoking Esacs after the onset of Vsacs, namely the refractory period for Vsacs, was 170 ms. The duration of the refractory period of Vsacs depends on the amplitude and direction of the Vsacs. The effects of the amplitude (Fig. 12B) and direction (Fig. 12C) of preceding Vsacs on the latency of Esacs were examined by varying the Vsac amplitude with a fixed direction and the Vsac direction with a fixed amplitude, respectively. Because the latency of Esacs varied depending on the initial eye position, we measured, as a control latency for the different eye positions, the latency of Esacs evoked by stimulation at 600 ms after Vsac onset. We then used the latency difference between Esacs evoked by stimulation at the onset of Vsacs and at 600 ms after Vsacs as a measure of the latency delay caused by the refractoriness of Vsacs with different amplitudes or directions. Generally, Vsacs greater than Esacs (15.7° in this case) had a stronger suppressive effect on the latency of Esacs (Fig. 12B). As to the effect of the Vsac direction on Esac latency, the suppressive effect of preceding Vsacs on the latency of Esacs was strongest when the Vsacs were in the same direction as the Esacs (Fig. 12Ce) and was weaker as the Vsac direction differed from the direction of the Esacs. The influence of preceding Vsacs on the vector of Esacs was different between horizontal and vertical components of saccades. As shown in Fig. 12, B and C, the vector of Esacs turned toward the vicinity of the endpoint of control Esacs with a central initial eye position. This phenomenon was similar to the influence of spontaneous saccades on Esacs reported by Schlag and Schlag-Rey (1987). In addition, the results shown in Fig. 12B, including reversal in direction of the saccades, were similar to the effects of preceding Vsacs on the amplitude of SC-evoked Esacs (Nichols and Sparks 1995). However, the present interaction between Vsacs and Esacs appeared to occur during and after the resettable integrator's discharge period, which was estimated by Nichols and Sparks (1995). These previous studies reported the effects of preceding Vsacs on the amplitude and direction of Esacs but did not systematically analyze the influence of preceding Vsacs on Esac latencies. The present results showed that the effects of refractoriness on saccades were different from those of FEF suppression on Vsacs in several ways. First, refractoriness following Vsacs exerted a stronger suppressive effect on the latency of Esacs that were smaller than the preceding Vsacs. Accordingly, if stimulation of a bilateral suppression site evoked tiny Esacs, the suppression caused by the refractoriness of such tiny Esacs should produce a stronger suppression of smaller Vsacs. However, this is contrary to the present finding that contraversive Vsacs with an amplitude larger than unidentifiable tiny Esacs were strongly suppressed. Second, refractoriness following Vsacs exerted a stronger suppressive effect on the latency of Esacs with a direction similar to that of the preceding Vsacs. In contrast, the suppression of contraversive Vsacs by stimulation of a bilateral suppression site was typically independent of the direction of Vsacs. Third, refractoriness following Vsacs strongly affected the vector of Esacs, whereas the vector of contraversive Vsacs was little influenced by stimulation of a bilateral suppression site. Therefore these findings exclude the possibility that the suppression of contraversive Vsacs might be caused by refractoriness following unidentifiable tiny Esacs.
View larger version (39K):
[in this window]
[in a new window]
|
FIG. 12. Interaction of Vsacs and Esacs to examine the refractoriness of saccades. A: effects of varying the interval between Vsac onset and FEF stimulation (20 µA, 40 pulses) on the Esac latency. The onset of contraversive horizontal 15° Vsac was varied relative to stimulation onset (dotted line) for eliciting Esacs. A positive interval indicates that Vsacs preceded the stimulation onset. Stimuli were triggered, with varying delays, by the computer-detected onset (100°/s) of Vsacs for positive intervals and by visual target onset for negative intervals, and the intervals between Vsac onset and stimulation onset were measured later. Thick line, median latencies of Esacs (ordinate) were plotted as a function of stimulation onset relative to the Vsac onset (abscissa). Thin horizontal line, median latency of control Esacs. B: effects of the amplitude of Vsacs on the latency of Esacs. Stimulation (30 µA, 40 pulses) was triggered by the onset (delay 0 ms) of contraversive 8–28° Vsacs. The same stimulation was applied 600 ms after the onset of Vsacs (delay 600 ms) to evoke control Esacs at corresponding eye positions. Thick line, latency differences of Esacs (ordinate) between delay 0 ms and delay 600 ms were plotted as a function of the amplitude of Vsacs (abscissa). Thin vertical line, median amplitude of control Esacs. C: effects of the direction of Vsacs on the latency of Esacs. Stimulation (30 µA, 40 pulses) was triggered by the onset (delay 0 ms) of 16° Vsacs with 8 directions at 18° intervals around the direction of Esacs. The same stimulation was triggered at 600 ms after Vsac onset (delay 600 ms) to evoke Esacs at corresponding eye positions. Latency differences between Esacs at delay 0 ms and delay 600 ms were used as a measure of the strength of suppression and plotted in a polar representation of the Vsac direction. Arrow indicates the direction of control Esacs. Same stimulation site as in B.
|
|
|
DISCUSSION
|
|---|
The present study showed that electrical stimulation of a localized area in the FEF caused the suppression of Vsacs and Msacs in all directions. Stimulation of this area usually did not evoke any Esacs at 80 µA but suppressed saccades in all directions at a low threshold (40 µA). The thresholds for suppression were almost the same for both Vsacs and Msacs and also for ipsi- and contraversive saccades. We reported previously the suppression of only ipsiversive Vsacs and Msacs by stimulation of the FEF (Izawa et al. 2004). Electrical stimulation of this unilateral suppression area easily evoked saccades at 50 µA. Therefore the bilateral suppression area was different from this unilateral suppression area in that electrical stimulation did not evoke Esacs at 50 µA. Unilateral suppression sites were distributed widely throughout the classical FEF (Izawa et al. 2004), whereas bilateral suppression sites were localized in the caudal part of the arcuate gyrus facing the inferior arcuate sulcus and deeper than sites for the unilateral suppression. This bilateral suppression area did not correspond to the central area of the classical FEF but may belong to a subregion of the classical FEF where electrical stimulation produces small Esacs and visual neurons there have small receptive fields near the fovea (Azuma et al. 1988) or may be adjacent to it. The exact distribution of bilateral suppression sites in and around the FEF must be determined by the histological reconstruction of stimulation tracks. If this suppression area is related to ocular fixation, this area may be regarded as the subregion of the FEF that should be defined in a broader sense as an area in the arcuate gyrus related to various kinds of eye movements including fixation.
Burman and Bruce (1997) reported that electrical stimulation of the FEF most effectively suppressed Msacs directed contraversive to the stimulated hemisphere. There are several differences between the present suppression and that of Burman and Bruce (1997): the suppression occurred on the initiation of both Vsacs and Msacs and thresholds for suppressing them were almost equal, both ipsi- and contraversive saccades were almost equally suppressed and thresholds for suppression were almost equal for saccades in both directions, and the vectors of Vsacs and Msacs were not usually affected. At some sites in the bilateral suppression area, the suppression thresholds for ipsiversive saccades were higher than those for contraversive saccades. Accordingly, stimulation of such sites appeared to suppress only contraversive saccades at subthreshold for the suppression of ipsiversive saccades. At some of such sites, stimulation evoked ipsiversive slow eye movements, most likely smooth pursuit eye movements (Bruce et al. 1985). If these sites are stimulated, reciprocal inhibition for horizontal smooth pursuits may result in the suppression of contraversive saccades and the facilitation of ipsiversive saccades at the motoneuronal level. Actually, the suppressive effects of FEF stimulation reported by Burman and Bruce (1997) were mainly on contraversive Msacs, and ipsiversive Msacs were facilitated in a reciprocal fashion. Furthermore, the preceding interpretation is consistent with their report that their stimulation sites were mainly located near the spur of the arcuate sulcus (Burman and Bruce 1997) similar to the smooth pursuit subregion of the FEF (MacAvoy et al. 1991). The present bilateral suppression area was located more laterally and was clearly different from their suppression area.
The accuracy of bilateral Vsacs and Msacs was not affected by stimulation of the bilateral suppression sites, although some Msacs tended to be hypometric at some sites. Therefore the results suggest that stimulation of the bilateral suppression sites may suppress the saccadic trigger system for ipsi- and contraversive saccades but does not affect the metric controlling system (Fuchs et al. 1985; Scudder et al. 2002; Sparks et al. 1987). The latest and earliest effective stimulus timings for the suppression were 140 and 135 ms for ipsiversive Vsacs (40 and 45 ms before the saccade onset, respectively), and 135 and 130 ms for contraversive Vsacs (45 and 50 ms before the saccade onset, respectively) from visual target onset, respectively, and the time course of this suppression was similar to the time course after stimulation of the unilateral suppression area. Therefore the neural mechanism responsible for suppression of ipsiversive saccades may be common to both the bi- and unilateral suppression sites in the FEF. Based on the stimulus timings for the bilateral suppression, FEF stimulation are most likely to suppress the SC and/or paramedian pontine reticular formation (PPRF) as discussed for the unilateral suppression in the previous paper (Izawa et al. 2004).
The effects of stimulating the bilateral suppression sites were usually similar for both Vsacs and Msacs in terms of thresholds and directionality of suppression. These findings suggest that the neural mechanism for the present bilateral suppression is different from that for the previously reported unilateral suppression (Izawa et al. 2004), and a single neural mechanism may be responsible for the suppression of ipsi- and contraversive saccades. Stimulation of the OPN area interrupts saccades in any direction (Keller 1977; Keller et al. 1996; King and Fuchs 1977). Therefore the present omnidirectional suppression of Vsacs and Msacs may be conveyed via OPNs if neurons in the FEF project to OPNs directly (Stanton et al. 1988) or indirectly via the SC (Gandhi and Keller 1997; Langer and Kaneko 1990; Paré and Guitton 1994; Raybourn and Keller 1977). The preceding finding of the omnidirectional interruption of saccades by stimulation of the OPN area (Keller et al. 1996) does not necessarily mean that single OPNs project bilaterally. OPNs projecting ipsi- or contralaterally might be intermingled on one side. If OPNs project to contralateral excitatory burst neurons (EBNs), stimulation of the midline area will activate bilateral OPNs and result in either the equal or asymmetric suppression of bilateral saccades. In fact, the present suppressive effect was different in some respects for contra- and ipsiversive Vsacs. First, ipsiversive Vsacs were almost totally suppressed during FEF stimulation, whereas contraversive Vsacs were not suppressed in all trials with the same stimulation. Second, the suppression threshold was lower for ipsiversive Vsacs with larger amplitudes and contraversive Vsacs with smaller amplitudes. These findings suggest that the neural mechanism for the suppression of contraversive Vsacs at least partly differs from that for the suppression of ipsiversive Vsacs. In this regard, the bilateral suppression area might be a part of the unilateral suppression area and contain an additional neural mechanism for the suppression of contraversive saccades. The ipsilateral suppression of the bilateral suppression area should not be ascribed to reciprocal inhibition of Esacs because stimulation did not evoke Esacs at 80 µA but induced suppression at 40 µA. Regarding the suppression of contraversive saccades, we were able to exclude the possibility of interaction between Vsacs and refractoriness of tiny Esacs (Fig. 11). Using an optical imaging method, Seidemann et al. (2002) reported that electrical stimulation of the FEF evoked hyperpolarization after depolarization in the stimulated FEF. Therefore this inhibition in the stimulated FEF might be partially responsible for the suppression of contraversive saccades. The rostral SC receives input from the FEF (Segraves and Goldberg 1987), and the SC has intracollicular inhibitory connections (Munoz and Istvan 1998). FEF stimulation inhibits saccade-related neurons in the ipsilateral SC (Schlag-Rey et al. 1992) or may activate neurons in the substantia nigra reticulata via the subthalamus (Carpenter et al. 1968; Kitano et al. 1998; Künzle and Akert 1977) and thereby inhibit SC output neurons. Stimulation of the rostral pole of the SC suppresses bilateral saccades (Gandhi and Keller 1999; Munoz and Wurtz 1993b; Paré and Guitton 1994), although the functional contribution of collicular projection to OPNs for fixation is controversial (Gandhi and Keller 1999; Kaneko 1996). If the rostral SC receives input from the present bilateral suppression area, this pathway will produce suppression of bilateral saccades.
Recently the FEF has been shown to contain neurons that are related to smooth pursuit eye movement (Bruce et al. 1985; Tanaka and Fukushima 1998; Tanaka and Lisberger 2002) and convergence (Fukushima et al. 2002; Gamlin and Yoon 2000). In addition, the FEF also contains neurons that are related to fixation (Bizzi 1968; Bruce and Goldberg 1985; Suzuki and Azuma 1977). The act of fixation makes the brain stem saccadic system less susceptible to the effects of extraneous commands to make a saccade (Goldberg et al. 1986). Actually, the present bilateral suppression area of the FEF contained fixation neurons and therefore may be responsible for controlling fixation and modulating other eye-movement systems. In humans, it has been reported that patients with a unilateral frontal lobe lesion have difficulty in suppressing reflexive saccades to both sides in an anti-saccade task (Guitton et al. 1985). This finding may be related to dysfunction of the present bilateral suppression area. Conversely, abnormal function of the bilateral suppression area might explain the clinical disorder known as spasm of fixation, which causes an impaired initiation of saccades in the presence of a fixation target (Johnston et al. 1992).
Suppression of bilateral Vsacs increased when the initial eye position was shifted toward the direction of Vsacs. Because the initial position of the eyes in the orbits affects the ratio of the eye and head contribution to the large gaze shift (Freedman and Sparks 1997), this mechanism may also be involved in the present initial eye position effects that may contribute to eye and head coordination. Shifting the initial eye position toward the direction of Vsac induces motoneuronal depolarization, which might inactivate sodium currents and low-threshold calcium currents (Gueritaud 1994; Russier et al. 2003) and delay spike initiation. In addition, a certain class of pause neurons whose activity is correlated with eye position has higher discharge rates with a more ipsilateral eye position (Keller 1974). Therefore these pause neurons may cause the stronger suppression of centrifugal Vsacs if they project to ipsilateral EBNs. During fixation, this mechanism may contribute to eye centering (Bender 1955).
Larger ipsiversive Vsacs were suppressed at lower thresholds by FEF stimulation. This finding suggests that fixation neurons at the bilateral suppression sites may decrease their activity during smooth pursuits so that only small catch-up saccades are allowed to occur during ipsiversive smooth pursuits to follow a target accurately. This decrease in activity of the fixation neurons may account for decrease in the activity of OPNs during smooth pursuit (Missal and Keller 2002). On the other hand, smaller contraversive Vsacs were suppressed at lower thresholds by FEF stimulation. This property may assure the maintenance of fixation by suppressing the generation of small saccades to distracting objects near the fixation point, although it may help generate saccades to change a line of sight to targets that appear in the periphery of a contralateral visual field. To further understand the functional role of the bilateral suppression area in the FEF, further studies are required to analyze activity of neurons in the FEF and the SC in relation to fixation and various kinds of eye movements.
|
GRANTS
|
|---|
This research was supported by a grant from Core Research for Evolutional Science and Technology of Japan Science and Technology Corporation to Y. Shinoda, a grant of the 21st Century COE Program to Y. Shinoda and a Grant-in-Aid for Scientific Research from the Ministry for Education, Science and Culture of Japan to Y. Izawa.
|
ACKNOWLEDGMENTS
|
|---|
We thank Prof. K. Kubota and Dr. I. Sugihara for providing valuable comments on the initial manuscript and M. Takada for invaluable technical assistance.
|
FOOTNOTES
|
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Address for reprint requests and other correspondence: Y. Shinoda, Dept. of Systems Neurophysiology, Graduate School of Medicine, Tokyo Medical and Dental University, 1-5-45, Yushima, Bunkyo-ku, Tokyo 113-8519, Japan (E-mail: yshinoda.phy1{at}med.tmd.ac.jp).
|
REFERENCES
|
|---|
Azuma M, Nakayama H, Sasaki Y, and Suzuki H. Relation between visual input and motor outflow for eye movements in monkey frontal eye field. Behav Brain Res 27: 93–98, 1988.[CrossRef][ISI][Medline]
Azuma M, Nakayama H, and Suzuki H. Suppression of visually triggered saccades by electrical stimulation of the monkey frontal eye field (Abstract). J Physiol Soc Jpn 48: 266, 1986.
Bender MB. The eye-centering system. AMA Arch Neurol Psychiatry 73: 685–699, 1955.[Medline]
Bizzi E. Discharge of frontal eye field neurons during saccadic and following eye movements in unanesthetized monkeys. Exp Brain Res 6: 69–80, 1968.[ISI][Medline]
Bruce CJ and Goldberg ME. Primate frontal eye fields. I. Single neurons discharging before saccades. J Neurophysiol 53: 603–635, 1985.[Abstract/Free Full Text]
Bruce CJ, Goldberg ME, Bushnell MC, and Stanton GB. Primate frontal eye fields. II. Physiological and anatomical correlates of electrically evoked eye movements. J Neurophysiol 54: 714–734, 1985.[Abstract/Free Full Text]
Burman DD and Bruce CJ. Suppression of task-related saccades by electrical stimulation in the primate's frontal eye field. J Neurophysiol 77: 2252–2267, 1997.[Abstract/Free Full Text]
Carpenter MB, Fraser RAR, and Shriver JE. The organization of the pallidosubthalamic fibers in the monkey. Brain Res 11: 522–559, 1968.[CrossRef][Medline]
Cohen B and Henn V. Unit activity in the pontine reticular formation associated with eye movements. Brain Res 46: 403–410, 1972.[CrossRef][ISI][Medline]
Freedman EG and Sparks DL. Eye-head coordination during head-unrestrained gaze shifts in rhesus monkeys. J Neurophysiol 77: 2328–2348, 1997.[Abstract/Free Full Text]
Fuchs AF, Kaneko CRS, and Scudder CA. Brain stem control of saccadic eye movements. Annu Rev Neurosci 8: 307–337, 1985.[CrossRef][ISI][Medline]
Fukushima K, Yamanobe T, Shinmei Y, Fukushima J, Kurkin S, and Peterson BW. Coding of smooth eye movements in three-dimensional space by frontal cortex. Nature 419: 157–162, 2002.[CrossRef][Medline]
Gamlin PD and Yoon K. An area for vergence eye movement in primate frontal cortex. Nature 407: 1003–1007, 2000.
TOP
ABSTRACT
INTRODUCTION
