| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Journal of Neurophysiology Vol. 79 No. 2 February 1998, pp. 704-715
Copyright ©1998 by the American Physiological Society
Department of Otolaryngology, University of Pittsburgh, Pittsburgh, Pennsylvania 15213
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Scudder, Charles A., Ekatherina Y. Batourina, and George S. Tunder. Comparison of two methods of producing adaptation of saccade size and implications for the site of plasticity. J. Neurophysiol. 79: 704-715, 1998. Saccade accuracy is known to be maintained by adaptive mechanisms that progressively reduce any visual error that consistently exists at the end of saccades. Experimentally, the visual error is induced using one of two paradigms. In the first, the horizontal and medial recti of trained monkeys are tenectomized and allowed to reattach so that both muscles are paretic. After patching the unoperated eye and forcing the monkey to use the "paretic eye," saccades initially undershoot the intended target, but gradually increase in size until they almost acquire the target in one step. In the second, the target of a saccade is displaced in midsaccade so that the saccade cannot land on target. Again saccade size adapts until the target can be acquired in one step. Because adaptation with the latter paradigm is very rapid but adaptation using the former is slow, it has frequently been questioned whether or not the two forms of adaptation depend on the same neural mechanisms. We show that the rate of adaptation in both paradigms depends on the number of possible visual targets, so that when this variable is equated, adaptation occurs at similar rates in both paradigms. To demonstrate further similarities between the result of the two paradigms, an experiment using intrasaccadic displacements was conducted to show that rapid adaptation possesses the capacity to produce gain changes that vary with orbital position. The relative size of intrasaccadic displacements were graded with orbital position so as to mimic the position-dependent dysmetria initially produced by a single paretic extraocular muscle. Induced changes in saccade size paralleled the size of the displacements, being largest for saccades into one hemifield and being negligible for saccades into the other hemifield or in the opposite direction. Collectively, the data remove the rational for asserting that adaptation produced by the two paradigms depends on separate neural mechanisms. We argue that adaptation produced by both paradigms depends on the cerebellum.
Because saccades have too short a duration for vision to directly control their final position, saccade accuracy must be maintained over the long term by mechanisms that assess saccade error at the end of each saccade and modify subsequent movements. That such adaptive mechanisms do exist was dramatically demonstrated in patients with paresis of the extraocular muscles in one eye (Abel et al. 1978 Five juvenile rhesus monkeys were used in these experiments. All Procedures were approved by the Institute Animal Care and Use Committee and conform to guidelines issued by the National Institutes of Health. Two to four monkeys were used in each of the eye-movement recording paradigms described below.
Surgery
Monkeys were prepared for eye movement recording during an aseptic surgery in which they were deeply anesthetized with halothane (1.5% in oxygen/nitrous oxide) after induction with ketamine. A 16-mm, 3-turn magnetic search coil (Cooner AS-632) was sutured to the sclera of the right eye under the conjunctivum by using a method modified from Judge et al. (1980) Behavioral apparatus and tasks
Experiments were conducted in a dimly illuminated soundproof booth with the monkeys seated in a primate chair and their heads restrained. Surrounding their heads were two perpendicular Helmholtz coils, which were used in conjunction with the implanted search coil for the measurement of eye movements (CNC Engineering). Eye movement signals were calibrated by having the monkey alternately fixate stationary targets at ±15° horizontal and vertical eccentricity.
Adaptation paradigms
Two methods were used for inducing adaptive modification of saccade size. In the intrasaccadic step (ISS) paradigm, the target was stepped to a location in one of the above patterns (the primary target) and then displaced collinearly forward by 20% or backward by 30% of the original target displacement. These fractions were chosen because they yield about the same size error given that the primary saccades normally fall about 5% short of the primary target. ISSs were initiated ~8-30 ms after primary saccade onset. Within each adaptation session, only one type of ISS was presented and ISSs were triggered for every trial. Two or more days typically intervened between sessions or if not, ISSs of opposite directions were used on successive days. There was no evidence of carry-over from day to day, with preadaptation saccade size as a measure.
Data collection and analysis
Eye movement and target data were collected before adaptation, at the start of adaptation (the initiation of ISSs or patch-switching), at the end of adaptation, and after adaptation (after the termination of ISSs or returning the patch to the original eye). These will be referred to as the pre-, start-, end-, and postadaptation records. Data were also collected at 0.5-, 1-, 2-, 3-h, and sometimes at the intervening 0.5-h intervals. Typically data from 100 trials were collected. One minute usually intervened between start of the adaptation paradigm and the collection of the "start-adaptation" data and 100-130 trials intervened between the start of adaptation and the midpoint of the start-adaptation data. Horizontal and vertical eye and target position were digitized on-line and encoded on VCR tape (Vetter 4000A) for backup. Data were sampled every 5 ms. A record of the number of primary target steps and the monkeys performance was also kept. Performance was measured as time-on-target, which was the fraction of the total running time that the monkey was within the reward window. Monkeys were counted as being on target after a target step if they acquired the target within 0.5 s during a normal trial (no ISS) or 0.8 s during any adaptation trial (ISS or paretic-muscle). Because the monkeys performance had a measurable effect on the results of ISS adaptation, no data were used after the cumulative performance fell below 90% or after performance fell below 80% in a particular 30-min interval. In the paretic-muscle paradigm, unlike the ISS-paradigm, recovery does not ensue when the monkey makes spontaneous saccades instead of following the target. Hence the 90% cumulative performance criterion was dropped, but the 80% interval criterion was maintained.
Basic results from the ISS paradigm
Adaptation using intrasaccadic steps produced reliable changes in saccade gain in all cases, albeit at disparate rates that depended on the target pattern. Figure 1 illustrates typical trials obtained from the pre-, start-, end-, and post-adaptation records (A-D, respectively) during a session with backward ISSs with the 2 × 1 target pattern. With the onset of ISSs (B), saccades consistently overshot the primary target position. After about 2,000 trials (C) however, saccades were reduced sufficiently in size that many of them accurately acquired the final target position. Cessation of the ISSs (D) caused the saccades to severely undershoot the target.
Rapid and very rapid adaptation
As illustrated in Fig. 2, there were sizable differences between the gain measurements obtained during the preadaptation test trials and the trials at the start of adaptation. Compared to the preadaptation gains, the start-adaptation gains were consistently higher during forward-ISS trials and lower during backward-ISS trials. The average size of these differences depended on both the direction of ISS steps and the target pattern in a manner very similar to the rates of change reported above (Table 1). Changes were smallest for the forward-ISS trials using the 5 × 3 pattern (0.0141; 1.5%) and largest for the backward-ISS trials with the 2×1 pattern (
Position-dependent ISS paradigm
Figure 3 illustrates the results of adapting a monkey using the position-dependent adaptation paradigm. In this paradigm, only leftward saccades directed at targets in the left visual field were associated with backward ISSs, while all other saccades were associated with no ISSs. Figure 3 shows that saccade gain adapted faithfully according to the pattern of ISSs. The gains of end- and postadaptation leftward saccades (
Paretic-muscle paradigm
In the paretic-muscle paradigm, monkeys viewed their environment for several days while wearing a patch over either the normal or the paretic eye. Switching the patch to the other eye and requiring the monkey to follow targets landing on the 2 × 1 or 5 × 3 grid produced significant within-session changes in saccade gain in all cases (Table 3.). As seen in Fig. 5, when monkeys were forced to use their normal eye after previous adjustment to using their paretic eye, the gains of horizontal saccades were initially very high (top, Paretic
The results of this study support the hypothesis that adaptation of saccade gain by the ISS-paradigm and the paretic-eye paradigm share the same adaptive mechanisms. The major objection to this hypothesis was that the rates of adaptation produced by each paradigm seemed to be extremely different (Optican 1985 Dependence of adaptation rate on the complexity of the target array
The first experiment showed that backward ISSs produced gain decreases more rapidly than forward ISSs produced gain increases, as was shown by others in monkeys and humans (Deubel 1987 Rate of adaptation; from very rapid to very slow
A consistent finding in these experiments was that the gain measured during preadaptation trials (no ISSs) differed from that measured during start-adaptation trials (with ISSs). The direction and magnitude of these changes correlated with the direction and magnitude of the longer-term within-session changes and so were quite sizable with the 2 × 1 target pattern. Experiments showed that this was a very rapid form of adaptation, which took place within the space of 20 or so trials.
Short- and long-term mechanisms
Previously, questions about the mechanisms of saccadic adaptation were framed in terms of the paradigms used to produce adaptation. "Short-term" adaptation was the product of the ISS paradigm and "long-term" adaptation was the product of the paretic-muscle paradigm. This study shows that the paradigm is irrelevant when considering the existence of long- and short-term mechanisms of adaptation. Indeed, it is difficult to see why the nervous system would care whether dysmetria were produced by displacing a target or by muscle tenectomy. Both interventions result in a visual error and require a corrective saccade to reach the target [but see Miller et al. (1981)
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
; Kommerell et al. 1976; Optican et al. 1985). Whereas the normal eye was able to saccade to visible targets accurately, the paresis in the other eye necessarily rendered its eye movements much smaller. Hence, when the patient was forced to use the "paretic eye" by patching the normal eye, saccades initially undershot the target (hypometria). Over days of forced use of the paretic eye however, saccades increased in size until they were nearly accurate. When the patch was switched and the normal eye was used again, saccades in this eye overshot the target initially (hypermetria), but adaptive mechanisms restored normal accuracy over succeeding days.
). Switching the patch and forcing the monkey to use the normal eye produced saccadic hypermetria that also was gradually eliminated. The changes in saccade size had an approximately exponential time course with time constants of <1 day and 0.5 days respectively.
; McLaughlin 1967; Miller et al. 1981). Saccade size gradually decreases or increases, respectively, so that saccades eventually land on target. The rate of change varies widely. In monkeys undergoing steps of multiple sizes, the modification of saccade size was fit with an exponential having a rate constant of 200-600 trials (Deubel 1987; Fuchs et al. 1996; Straube et al. 1997), but in humans generating saccades between just two target locations, adaptation of saccade size may be completed in as few as 6 trials (McLaughlin 1967; Miller et al. 1981; Weisfeld 1972). Adaptation produced by the ISS-paradigm has sometimes been called "rapid" saccadic adaptation to distinguish it from that produced by the paretic-muscle paradigm, which seems to be an order of magnitude slower.
; Wolf et al. 1984). Similarly, adaptation of saccades in one direction generalizes little or not at all to perpendicular saccades (Deubel 1987; Frens and van Opstal 1994). Adaptation to target steps of one amplitude only partially generalizes to target steps of a different amplitude (Albano 1996; Frens and van Opstal 1994; Straube et al. 1997), whereas adaptation produced at one eye position generalizes well, albeit incompletely, to saccades at different eye positions (Albano 1996).
; Optican 1985). However differences may be more apparent than real because the two types of adaptation have not been investigated under similar conditions. In the paretic-muscle paradigm, monkeys and humans make spontaneous saccades to self selected targets at widely varying displacements, directions, and positions in space, so that generalization of adaptation is minimal. In the human experiments where typically only one muscle is paretic (Abel et al. 1978; Kommerell et al. 1976), the innervation required to make saccades of the same size varies enormously with eye position in the orbit (Optican et al. 1985). It could be the case that saccades of a certain size, direction, and position adapt almost independently and do not benefit from adaptation of saccades of other sizes, directions, and positions. In contrast, saccades produced in the ISS-paradigm are of restricted size, direction, and position in space. Under these latter conditions, generalization insures that adaptation of saccades with mildly different parameters will be mutually reinforcing and the overall rate of adaptation will increase. A second difference is that subjects in the ISS-paradigm are required to acquire the target, whereas subjects with a paretic eye may tolerate considerable error. If, for instance, corrective saccades convey an error signal for adaptation as some have postulated (Albano and King 1989; Schweighofer et al. 1996), mandatory acquisition of the target would be expected to increase the rate of adaptation.
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
. Lead wires were passed under the skin to a connector on the top of the skull. Three lugs for the stabilization of the head were built from dental acrylic layered about stainless screws attached to the skull. One lug was positioned at the top front center of the skull and two at the back sides behind both ears. These lugs mated with bars rigidly attached to the primate chair in which monkeys sit during the recording sessions.
). For saccades in one horizontal direction (which will always be denoted as right), there were no ISSs. In the opposite (left) direction, there were no ISSs if the target was located in the right visual field, but targets located in the left visual field were followed by ISSs whose relative size varied linearly with the distance of the primary target from the vertical meridian at rate of 25% per 10° eccentricity (a target landing at 10° left would be followed by a 25% ISS). The stimulus was designed to produce adaptation in only one direction and one half of the visual field. This paradigm will be called the "position-dependent ISS" paradigm.
). The user manually scrolled the data and placed a cursor near desired saccades. The computer (PDP 11/73 or 80486-based) found peak velocity and independently searched backward and forward in time for the first point where eye position reached steady-state initial or final eye positions. The computer marked these points and the user accepted them or manually marked those that were in error. Relevant derived values include reaction times and the amplitudes of the horizontal and vertical saccadic components. Saccade gain, defined as saccade amplitude divided by target displacement was computed for polar values as well as component values (e.g., horizontal component size/horizontal target displacement). Most of the data reported below refer to the gain of the horizontal component. In the paretic-muscle paradigm, gain was always computed for the viewing (unpatched) eye.
15° and 15°25° steps were always similar and not significantly different, despite the higher proportion of 10° target steps. Consequently, saccades of all sizes were pooled. Similarly, leftward and rightward saccade data were pooled except for the position-dependent ISS paradigm. In the paretic-muscle paradigm, leftward and rightward saccade gains were frequently dissimilar, but were averaged because their difference was not currently of interest. The pooled variance was computed from the separate left- and right-saccade variances, and so did not include the difference between the means. Because data were collected at fixed times and not fixed trial numbers, small gain corrections were sometimes made so gains could be compared at the fixed trial numbers of 950 and 1,900 (the mean number of trials at 30 and 60 min). The exponential best fit of gain versus trials (see below) was displaced so as to pass through the actual gain, and the gain at 950 or 1,900 trials was computed from the displaced exponential. In the ISS paradigm, these adjustments amounted to a few percent of the original value, but the performance rate in the paretic-muscle paradigm differed systematically, so adjustments were up to 10%. Gains were statistically compared using Student's t-test with one-tailed probabilities because there was always a clear prediction about the direction of the changes.
; Henson 1978). For the paretic-muscle paradigm, the theoretical limit was taken as 0.95 when viewing with the normal eye, and 0.95, 0.85, or 0.65 for monkeys G, M, or W when viewing with the paretic eye. These numbers represent the average gain measured after long-term (days) viewing with the paretic eye. The normalized gain data were plotted against number of trials and were fit with exponentials having variable asymptotes and rate constants using the method of least squares.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
View larger version (15K):
[in a new window]
FIG. 1.
Target steps and saccadic responses from individual trials recorded during intrasaccadic step (ISS) paradigm. Preadaptation data are plotted in A, start-adaptation data in B, end-adaptation data are plotted in C, and post-adaptation in D. With initiation of 30% backward ISSs in B, saccades initially overshot target, but eventually declined in size so that they essentially acquired target without a corrective saccade (C). When ISSs were discontinued (D), reduced saccade size persisted and saccades grossly undershot target.
View larger version (20K):
[in a new window]
FIG. 2.
Changes in horizontal component saccade produced by adaptation with +20% forward ISSs (A, circles), and 30% backward ISSs (B, triangles) is plotted against number of adaptation trials. Changes are computed relative to gains of start-adaptation trials. Preadaptation data are plotted at 130 trials. Each mean represents data from 60-120 saccades. Error bars correspond to standard deviation about mean. 2 × 1 × 20° refers to 2 × 1 target pattern with 20° spacing; 5 × 3 × 10° refers to 5 × 3 target pattern with 10° spacing.
View this table:
TABLE 1.
Gains and gain changes produced by ISS-adaptation
95% for both patterns. Exponential fits were of much less utility for the gain-increasing forward-ISS sessions. For the 5 × 3 pattern, gain sometimes increased so slowly that the data were almost linear within the duration of an experimental session. This meant that the rate constants and asymptotes of the fit were dependent on small amounts of curvature and were highly susceptible to experimental error. For three sessions, fit asymptotes considerably exceeded the theoretical limit and were set at 100%. Despite this, rate constants for the 2 × 1 and 5 × 3 patterns (1,661 and 4,955 trials) differed significantly with and without these three sessions. Asymptotic gain, according to the fits, occurred at about 75% of the maximum possible.
View this table:
TABLE 2.
Normalized gain changes produced by the adaptation paradigms
0.078; 8.2%). All differences were significantly different from zero (P < 0.025). After adaptation, opposite differences were seen between the end-adaptation and postadaptation gains (Table 1). For the 5 × 3 target pattern, the prestart and postend differences had nearly identical magnitudes, so that the prepost and start-end measures of adaptation were also nearly identical. However for the 2 × 1 pattern, the postend gain differences significantly exceeded the prestart gain differences. In other words, it appeared as if the changes in saccade size produced during adaptation incompletely transferred to saccades in the absence of ISSs.
0.0023/trial (0.244%/trial) for backward ISSs. If sustained for 30 trials, adaptation at this rate could account for all the difference between the gains measured in the pre- and start-adaptation trials.
View larger version (18K):
[in a new window]
FIG. 3.
Gain of horizontal component of saccades plotted as a function of horizontal ending position of primary target step. 
and 
: gains of leftward saccades (adapted direction). 
, gains of rightward saccades. Open symbols depict gains before adaptation and closed symbols depict gains after adaptation. Adaptation produced a gain decrement that varied with leftward target eccentricity (compare and 
), but had no effect on rightward saccades (
vs. ). Error bars represent standard deviations about mean.
30%-ISS, and 0%-ISS (normal) trials in a pseudorandom order using the 5×3 pattern. This paradigm produced no consistent differences between the gains associated with the +20% and 30%-ISS trials.
and ) were significantly smaller than the gains of the pre- and start-adaptation control saccades ( and 
). Overall, after adaptation, there was a significant trend for saccades landing further to the left to have lower gains than those landing to the right. At the leftmost (20°) targets, saccade gain was about 20% lower than before adaptation. In contrast, the gains of rightward preadaptation and rightward postadaptation saccades were substantially the same ( and ), as were the gains of rightward start- and end-adaptation saccades (not illustrated).
20° and 10°; cf. Becker 1989). The larger slopes seen at the start of adaptation (cf, monkey S) presumably are the product of the very rapid adaptation that was reported earlier for ordinary ISS adaptation. The existence of a positive slope indicates that the gain change was graded with the size of the ISS at each eccentricity rather than being uniformly depressed.
View larger version (38K):
[in a new window]
FIG. 4.
Comparison of slopes of regression lines fit to gain of horizontal component of saccades versus primary target step horizontal ending position computed for saccades in adapted (leftward) direction. Pre- and postadaptation slopes are plotted in same bar, and start- and end-adaptation slopes are plotted in adjacent bar. Slopes have units of %-change in gain per degree of target eccentricity. Height of each bar represents magnitude of post- or end-adaptation slopes, so that size of white or hatched area represents post-minus-pre and end-minus-start slopes; that is, increment in slope produced by adaptation. Error bars represent standard errors of slope. Data were obtained from nine experiments using 3 monkeys (M, S, G). Postadaptation data could not be collected for one session (G1).
Normal; Table 3). Over the course of a few thousand trials, saccade gain declined and became almost equal to that obtained after long-term viewing with the normal eye. Exponentials fit to the normalized data from individual sessions with the 2 × 1 target pattern all reached an asymptote of about 100% (nonnormalized gain = 0.95). With the 5 × 3 target pattern, saccade gain declined more slowly and usually did not reach the 100% theoretical limit. In fact, exponential fits only twice had asymptotes near 100% and the fit asymptotes for all sessions averaged 80% (Table 2; nonnormalized gain 
1.16). At the end of the experimental session, the patch was switched back to the paretic eye to compare pre- and postadaptation gain changes in that eye. The raw gain changes were always smaller in the paretic eye than in the (just adapted) normal eye, as would be expected because eye movements are smaller in general in the paretic eye. In normalized values, the gain of saccades measured in paretic eye changed 60%, which compares favorably with the average 80% change measured in the normal eye.
View this table:
TABLE 3.
Gains and gain changes (×100) produced by paretic eye adaptation
View larger version (19K):
[in a new window]
FIG. 5.
Gains of horizontal component of saccades produced by adaptation with paretic-muscle paradigm is plotted against number of adaptation trials. Top: (Paretic Normal), monkeys had been using their paretic eye and patch was switched so that they used their normal eye during experiment. Bottom: (Normal Paretic), monkeys used their paretic eye during experiment. Error bars represent standard deviations about mean. - - -, at a gain of 0.85, is average saccade gain that this monkey (M) achieved after long-term use of paretic eye and therefore depicts maximum gain that monkey could achieve during experiment.
Paretic). Saccade gain never reached an asymptotic value within the experimental session. Exponential fits to the normalized data revealed long time constants of over 3,000 trials and asymptotes slightly over 70%. When the patch was switched back to the normal eye at the end of the experimental session, the raw gain changes were always larger than in the (just adapted) paretic eye. Again, this would be expected from the relative mobility of the two eyes. In normalized values, the gain measured in the normal eye changed 47%, which is just slightly larger than the average 44% change measured in the paretic eye. The small differences found between the normalized gains in the normal and paretic eye are probably the product of nonlinearities in the paretic eye and not monocular adaptation (Snow et al. 1985).
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
; Optican and Robinson 1980). The current experiments provide an explanation for this apparent difference, and show that when examined under similar circumstances, adaptation produced by the two paradigms have similar features. The first experiment showed that the rate of adaptation became slower as the complexity of the target array increased and, in fact, was extremely slow when produced by gain-increasing forward ISSs using the most complex (5 × 3) target pattern. This situation is most comparable to previous studies with the paretic-muscle paradigm in which gain increases occurred as monkeys or humans scanned the multiplicity of targets provided by the real world (Able et al. 1978; Kommerell et al. 1976; Optican et al. 1985; Optican and Robinson 1980). The third experiment showed that the rate of adaptation produced with the paretic-muscle paradigm had a similar dependence on target complexity. Moreover, a direct comparison of the rates of adaptation produced by each paradigm, using targets of the same complexity, showed that they are similar and, in fact, adaptation using the paretic-muscle paradigm could sometimes be slightly faster. Hence the point is no longer tenable that differing rates produced by the two adaptation paradigms must reflect the existence of separate adaptation mechanisms. Finally, the results of the position-dependent ISS paradigm show that adaptation produced by the ISS paradigm possesses a capacity required for adaptation to naturally occurring extraocular muscle paresis, namely the capacity to adjust saccade size according the location of the target in head-referenced space (Optican et al. 1985).
; Straube et al. 1997; Straube and Deubel 1995). More importantly, the experiment showed that both forward and backward ISSs produced gain changes more slowly with the 5 × 3 target pattern than with the 2 × 1 pattern. Quantitatively, gain increases were about 47% slower and gain decreases were 36% slower. A similar trend was reported for human subjects (Miller et al. 1981). The rate dependence on the number of targets is a direct consequence of the finding reported by others that adaptation is relatively specific to the size, direction, and position of the adapting target steps (Albano 1996; Deubel 1987; Frens and van Opstal 1994, 1997; Miller 1981; Straube et al. 1997). Adaptation produced by exposure to target steps of a single size and direction transfers (generalizes) only to saccades directed at targets having similar sizes and directions. Frens and van Opstal (1994) described this restricted region as the "adaptation field." In contrast, if adaptation were mediated by parametric gain control, all trials would produce the same incremental gain change for all saccades regardless of the size, direction, or position of the target step. Because this is not the case, the total number of trials is divided among the different adaptation fields. This causes each adaptation field to receive less exposure to the adapting stimuli and saccades directed at that field will change gain proportionally less.
; Frens and van Opstal 1994; Straube et al. 1997). A possible reduction in the transfer of gain changes to saccades that differ in size from the adapted size has been more controversial (Deubel et al. 1986; Wolf et al. 1984), but the majority of the more recent studies using monkeys and humans have clearly shown such transfer is limited (Albano 1996; Frens and van Opstal 1994, 1997; Straube et al. 1997). A reduction in the transfer of adaptation to saccades outside the trained position in space has been the hardest to demonstrate (Albano 1996; Semmlow et al. 1989). Probably, the nervous system is predisposed to transfer adaptation to a wide region of space, but will restrict adaptation to a specific region when such transfer is contraindicated. For instance, in the position-dependent ISS paradigm, the large ISSs that accompany target steps to 20° left produced adaptation at that location, which might have transferred to other target positions in the absence differential training. The smaller ISSs associated with target steps to these other positions, however, indicate to the nervous system that the transfer of such a large gain change is inappropriate. In this circumstance, the nervous system clearly manages to restrict the transfer and to grade the amount of adaptation on the basis of the landing position of the primary target step.
; Straube et al. 1997) but have not been systematically examined. A valid question is why such changes, which could be quite large, have not been reported more often. A methodological answer may lie in the common practice of analyzing the data by generating a scatter plot of saccade gain against trial number and fitting the entire data set with an exponential. The initial rapid drop would be fit as part of the exponential and the discontinuity in rate obscured by the large scatter of saccade gains. A more intriguing answer is that it has been seen and corresponds to the very rapid changes seen in some human experiments (Erkelens and Hulleman 1993; McLaughlin 1967; Miller et al. 1981; Semmlow et al. 1989; Wiesfeld 1972). As in the current study, the very rapid changes were most prominent when there were only two target locations and the subjects simply made saccades back and forth. The normalized gain changes were smaller in monkeys (7.77/30% ISS = 26%, Table 1) than in humans (40% to 70%) (Abrams et al. 1992; Albano 1996; Albano and King 1989; Erklens and Hullman 1993; Miller et al. 1981; Semmlow et al. 1989), but not so much as to preclude the possibility that rapid gain changes in both species were produced by similar mechanisms. The proposal that monkeys and humans share this form of very rapid adaptation is consistent with the idea that other differences between monkey and human saccade adaptation (cf, Deubel 1987) are simply quantitative and not qualitative.
; Fuchs et al. 1996; Straube et al. 1997). Similarly, the 830-trial rate-constant observed for gain increases of leftward and rightward saccades (1,660/2) is also comparable to rate constants seen in these same studies. Rate constants for the 5 × 3 target pattern cannot be compared with other studies because similar experiments have not been done.
for a discussion of the theoretical importance of proprioception]. But, having dissociated the paradigm from the issue of short- and long-term adaptation, the question still remains whether or not there are separate short- and long-term mechanisms. The current experiments may not address this issue, because it is likely that adaptation in the paretic-muscle paradigm, as currently conducted, invoked the same short-term mechanisms as did the ISS-paradigm. On the other hand, as argued above, there is reason to believe that both paradigms invoke a spectrum of mechanisms from very rapid to very slow.
). The locus of short-term adaptation, as induced by the ISS-paradigm, is uncertain. Fitzgibbon et al. (1986) thought it might be upstream of the superior colliculus, because the size of saccades evoked by microstimulation of the superior colliculus were not modified after the size of visually evoked saccades was adaptively changed. Moreover, they argued that inflections in the velocity profiles of partially adapted saccades were the product of programming of two nearly simultaneous saccades. Similar velocity inflections were seen in our experiments during gain-decreasing adaptation with the paretic-muscle paradigm, but were mainly evident when the monkeys were not fully aroused. The evidence implicating the superior colliculus is not compelling however, because the same group also found that short-term adaptation was abolished after large lesions of the fastigial nucleus (Goldberg et al. 1993). This would seem to indicate that the cerebellum also mediates short-term adaptation. Moreover Frens and van Opstal (1997) found that neuronal discharges in the superior colliculus did not change during short-term adaptation using the ISS paradigm. Although it could be the case that both long- and short-term adaptation each reside at more than one site, we wish to consider the possibility that both do take place in the cerebellum.
; Miles and Eighmy 1980). This is analogous to having monkeys or patients naturally view the world while their nonparetic eye is patched. On the other hand, enduring VOR adaptation can also be rapidly produced by using the same optical devices and continuous forced vestibular stimulation, or by simulating the visual-vestibular conflict produced by these devices using optokinetic and smooth pursuit stimuli (Lisberger et al. 1984; Scudder and Fuchs 1992). This is, of course, similar to the rapid adaptation produced in this and similar experiments. A final similarity is that gain increases proceed more slowly than gain decreases (Miles and Eighmy 1980). Regarding the site of plasticity, VOR adaptation, whether short or long-term, was shown to require the cerebellum (Lisberger et al. 1984; Robinson 1976; Zee et al. 1981) and neurophysiological changes that might be the source of VOR gain changes were recorded in the vestibulocerebellum and its target structure, the vestibular nucleus (Lisberger et al. 1994a,b). Thus the changes that accompany adaptation would seem to be limited to the cerebellum and its deep nuclei, and the different rates in adaptation may reflect different cellular mechanisms acting at perhaps different synapses within the same system (Crepel and Jailard 1991; Ito 1989; Linden et al. 1991; Llano et al. 1991; Racine et al. 1986).
) indicates that short-term adaptation takes place in a structure receiving retinotopically place-coded information. Neurons in the brainstem saccadic burst generator are not place coded (cf, Fuchs et al. 1985) so it is very unlikely that adaptation occurs there. Neurons in the superior colliculus are place coded, but for adaptation to take place there, it might require a rapid and massive reorganization of the numerous inputs that converge from many cortical areas to produce a modified sensory-motor transformation (Sparks and Hartwich-Young 1989; Wurtz and Albano 1980). Moreover, this modified transformation would need to vary with eye position in the orbit to produce the results of the position-dependent ISS paradigm. The recent findings of Frens and van Opstal (1997) confirm the absence of any neural correlates of adaptation in the superior colliculus. The "oculomotor vermis," on the other hand, receives place-coded input from other structures (Crandall and Keller 1985; Ohtsuka and Noda 1992) as well as information about eye position in the orbit via mossy-fiber inputs (Kase et al. 1980). Although eye-position information is only weakly encoded at the output of the cerebellum in normal animals (Fuchs et al. 1993), it could be used when needed e.g., to compensate for severe orbital nonlinearities produced by paresis of one extraocular muscle. The inflections in saccade velocity that were sometimes observed here and elsewhere (Fitzgibbon et al. 1986), as well as the inconsistently observed alterations in saccade dynamics (Albano and King 1989; Straube and Duebel 1995; Straube et al. 1997) are probably the product of incompletely coordinated cerebellar input signals arriving from different sources (Melis and van Gisbergen 1996) and cerebellar output signals directed at different neuronal pools in the pontine saccadic burst generator (Noda et al. 1990). Finally, the cerebellum is an intricate structure possessing many levels, types of cells, and types of synapses, and is accordingly richly endowed with a variety plastic mechanisms (Crepel and Jailard 1991; Ito 1989; Linden et al. 1991; Llano et al. 1991; Racine et al. 1986). In summary, there is ample reason to consider that all adaptation in the saccadic system takes place in the cerebellum and that the spectrum of adaptation rates reflects a spectrum of plastic mechanisms.
| |
ACKNOWLEDGEMENTS |
|---|
This work was supported by National Eye Institute Grants EY-09210 and P30 EY-08098 and the Eye and Ear Institute Foundation of Pittsburgh.
| |
FOOTNOTES |
|---|
Address for reprint requests: C. A. Scudder, Dept. of Otolaryngology, University of Pittsburgh, 200 Lothrop St., Pittsburgh, PA 15213.
Received 15 May 1997; accepted in final form 14 October 1997.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  J. P. Herman, M. R. Harwood, and J. Wallman Saccade Adaptation Specific to Visual Context J Neurophysiol, April 1, 2009; 101(4): 1713 - 1721. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Cecala and E. G. Freedman Head-Unrestrained Gaze Adaptation in the Rhesus Macaque J Neurophysiol, January 1, 2009; 101(1): 164 - 183. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Cotti, M. Panouilleres, D. P. Munoz, J.-L. Vercher, D. Pelisson, and A. Guillaume Adaptation of reactive and voluntary saccades: different patterns of adaptation revealed in the antisaccade task J. Physiol., January 1, 2009; 587(1): 127 - 138. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Soetedjo, Y. Kojima, and A. F. Fuchs Complex Spike Activity in the Oculomotor Vermis of the Cerebellum: A Vectorial Error Signal for Saccade Motor Learning? J Neurophysiol, October 1, 2008; 100(4): 1949 - 1966. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Kojima, K. Yoshida, and Y. Iwamoto Microstimulation of the Midbrain Tegmentum Creates Learning Signals for Saccade Adaptation J. Neurosci., April 4, 2007; 27(14): 3759 - 3767. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Salman, J. A. Sharpe, M. Eizenman, L. Lillakas, T. To, C. Westall, M. Dennis, and M. J. Steinbach Saccadic Adaptation in Children J Child Neurol, December 1, 2006; 21(12): 1025 - 1031. [Abstract] [PDF]
|
|
|
|
|
|
N. Takeichi, C.R.S. Kaneko, and A. F. Fuchs Discharge of Monkey Nucleus Reticularis Tegmenti Pontis Neurons Changes During Saccade Adaptation J Neurophysiol, September 1, 2005; 94(3): 1938 - 1951. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Alahyane and D. Pelisson Long-lasting modifications of saccadic eye movements following adaptation induced in the double-step target paradigm Learn. Mem., July 1, 2005; 12(4): 433 - 443. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Kojima, Y. Iwamoto, and K. Yoshida Memory of Learning Facilitates Saccadic Adaptation in the Monkey J. Neurosci., August 25, 2004; 24(34): 7531 - 7539. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. E. Das, S. Ono, R. J. Tusa, and M. J. Mustari Conjugate Adaptation of Saccadic Gain in Non-Human Primates With Strabismus J Neurophysiol, February 1, 2004; 91(2): 1078 - 1084. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Scudder and D. M. McGee Adaptive Modification of Saccade Size Produces Correlated Changes in the Discharges of Fastigial Nucleus Neurons J Neurophysiol, August 1, 2003; 90(2): 1011 - 1026. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. MacAskill, T. J. Anderson, and R. D. Jones Adaptive modification of saccade amplitude in Parkinson's disease Brain, July 1, 2002; 125(7): 1570 - 1582. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Edelman and M. E. Goldberg Effect of Short-Term Saccadic Adaptation on Saccades Evoked by Electrical Stimulation in the Primate Superior Colliculus J Neurophysiol, April 1, 2002; 87(4): 1915 - 1923. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Maurer, T. Mergner, C. H. Lucking, and W. Becker Adaptive changes of saccadic eye-head coordination resulting from altered head posture in torticollis spasmodicus Brain, February 1, 2001; 124(2): 413 - 426. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Takagi, H. Abe, S. Hasegawa, T. Usui, H. Hasebe, A. Miki, and D. S. Zee Context-Specific Adaptation of Pursuit Initiation in Humans Invest. Ophthalmol. Vis. Sci., November 1, 2000; 41(12): 3763 - 3769. [Abstract] [Full Text]
|
|
|
|
|
|
S. Barash, A. Melikyan, A. Sivakov, M. Zhang, M. Glickstein, and P. Thier Saccadic Dysmetria and Adaptation after Lesions of the Cerebellar Cortex J. Neurosci., December 15, 1999; 19(24): 10931 - 10939. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
