Exp Brain Res (1992) 90:526-538
BrainR 9 Springer-Verlag 1992
Habituation and adaptation of the vestibuloocular reflex: a model of differential control by the vestibu|ocerebellum Helen Cohen 1' *, Bernard Cohen 1' z, Theodore Raphan 3, and Walter Waespe 4 1 Department z Department 3 Department 4 Department
of Neurology, The Mount Sinai School of Medicine, New York, USA of Physiology and Biophysics, The Mount Sinai School of Medicine, New York, USA of Computer and Information Science, Brooklyn College, City University of New York, USA of Neurology, University of Zfirich, Switzerland
Received December 11, 1991/Accepted April 9, 1992
Summary. We habituated the dominant time constant of the horizontal vestibuloocular reflex (VOR) of rhesus and cynomolgus monkeys by repeated testing with steps of velocity about a vertical axis and adapted the gain of the VOR by altering visual input with magnifying and reducing lenses. After baseline values were established, the nodulus and ventral uvula of the vestibulocerebellum were ablated in two monkeys, and the effects of nodulouvulectomy and flocculectomy on VOR gain adaptation and habituation were compared. The VOR time constant decreased with repeated testing, rapidly at first and more slowly thereafter. The gain of the VOR was unaffected. Massed trials were more effective than distributed trials in producing habituation. Regardless of the schedule of testing, the VOR time constant never fell below the time constant of the semicircular canals ( ~ 5 s). This finding indicates that only the slow component of the vestibular response, the component produced by velocity storage, was habituated. In agreement with this, the time constant of optokinetic after-nystagmus (OKAN) was habituated concurrently with the VOR. Average values for VOR habituation were obtained on a per session basis for six animals. The VOR gain was adapted by natural head movements in partially habituated monkeys while they wore x 2.2 magnifying or x 0.5 reducing lenses. Adaptation occurred rapidly and reached about _ 30%, similar to values obtained using forced rotation. VOR gain adaptation did not cause additional habituation of the time constant. When the VOR gain was reduced in animals with a long VOR time constant, there were overshoots in eye velocity that peaked at about 6-8 s after the onset or end of constant-velocity rotation. These overshoots occurred at times when the velocity storage integrator would have been maximally activated by semicircular canal input. Since the activity generated in the canals is not * Present Address: Department of Otolaryngology, Baylor College of Medicine, 1 Baylor Plaza, Houston, TX 77030, USA Offprint requests to. Bernard Cohen, Box 1135, Annenberg 21-74, Mount Sinai School of Medicine, 1 East 100th Street, New York, NY 10029, USA. Correspondence to: B. Cohen
altered by visual adaptation, this finding indicates that the gain element that controls rapid changes in eye velocity in the VOR is separate from that which couples afferent input to velocity storage. Nodulouvulectomy caused a prompt and permanent loss of habituation, returning VOR time constants to initial values. VOR gain adaptation, which is lost after flocculectomy, was unaffected by nodulouvulectomy. Flocculectomy did not alter habituation of the VOR or of OKAN. Using a simplified model of the VOR, the decrease in the duration of vestibular nystagmus due to habituation was related to a decrement in the dominant time constant of the velocity storage integrator (1/ho). Nodulouvulectomy, which reversed habituation, would be effected by decreasing ho, thereby increasing the VOR time constant. Small values of h o would cause velocity storage to approach an ideal integrative process, leading the system to become unstable. By controlling the VOR time constant through habituation, the nodulus and uvula can stabilize the slow component of the VOR. VOR gain adaptation was related to a modification of the direct vestibular path gain gl, without altering the coupling to velocity storage go or its time constant (1/ho). The mismatched direct- and indirect- pathway gains simulated the overshoots in the dynamic response to a step in velocity, that were observed experimentally. We conclude that independent distributed elements in the VOR modify its dynamic response, under control of separate parts of the vestibulocerebellum.
Key words: Habituation Adaptation Vestibuloocular reflex (VOR) - Velocity storage - Nystagmus - Semicircular canals Visual system - Monkeys - Optokinetic nystagmus (OKN) - After-nystagmus (OKAN) - Nodulus - Flocculus - Vestibulocerebellum
Introduction The dominant behavior of the vestibuloocular reflex (VOR) can be characterized by two properties, its gain and
527 time constant. These describe the dynamics of ocular compensation due to head rotation. The gain of the VOR, represented by the ratio of eye velocity to head velocity, is a measure of the goodness of compensation. The dominant time constant determines the frequency range of compensation, a long time constant corresponding to good low-frequency compensation during rotation (Raphan et al. 1979) and circular locomotion (Solomon and Cohen 1992b). Under normal circumstances, the VOR gain and time constant are matched so that compensation is maintained close to unity in darkness by a plateau in slow-phase velocity for several seconds at the onset or end of rotation (Raphan et al. 1979). The mechanism for achieving this match and the structures that are responsible for it are not known. Transient changes in VOR gain occur when close targets are viewed during angular head movement (Viirre et al. 1986) and during head rotations that include linear acceleration and translation (Bronstein and Gresty 1991; Solomon and Cohen 1992a). The dominant time constant of the VOR and of optokinetic after-nystagmus (OKAN) can also be altered transiently. The time constant of the VOR becomes shorter for trials in which the velocity of rotation is increased (Raphan et al. 1979). Likewise, the time constant of O K A N can be reduced in single trials by increasing the duration of optokinetic stimulation (Biittner et al. 1976). It is also possible to induce semipermanent changes in VOR gain, defined here as VOR gain adaptation. Retinal slip (Miles and Eighmy 1980; Miles and Lisberger 1981; Melvill Jones 1985) or an imagined mismatch between visual and vestibular signals (Melvill Jones et al. 1984) induces gain changes in the VOR to reduce gaze error. The gain changes are "plastic", in that they considerably outlast the trial in which they were induced. Most commonly, they have been produced by wearing reversingoptics lenses (Stratton 1897; Gonshor and Melvill Jones 1976a, b) or telescopic lenses (Miles and Eighmy 1980; Miles and Lisberger 1981) for prolonged periods (see Melvitl Jones 1985 for a review), but there can also be cross-axis adaptation (Schultheis and Robinson 1981; see Harrison et al. 1986 for a review). Rotation in a subjectstationary or moving surround is also a powerful stimulus for VOR gain adaptation (Ito 1982; Miles and Eighmy 1980). The ability to adapt the VOR through vision is lost after flocculus lesions (Lisberger et al. 1984). Repeated presentation of the same stimulus produces semipermanent reductions in the VOR time constant (J/iger and Henn 1981; Jeannerod et al. 1981; Schmid and Jeannerod 1985), which we define as VOR habituation. Once established, habituation can be retained for long periods of time without reinforcement (Thorpe 1956). Vestibular habituation is best induced in alert subjects (Crampton 1964). It is direction-specific and spatially oriented (Crampton 1962; Guedry 1965; Young and Henn 1976; Clement et al. 1981; Jeannerod et al. 1981; Schmid and Jeannerod 1985). It has been induced by caloric stimulation (Collins 1974; H o o d and Pfaltz 1954), and by steps (Dodge 1923; Jeannerod et al. 1976) and sinusoids of angular velocity (Dodge 1923; Clement et al. 1981; Jeannerod et al. 1981; J/iger and Henn 1981). Neither the
functional significance nor the signals that cause habituation are known, but, presumably, habituation decreases a maladaptive aspect of responses evoked by motion stimuli (Schmid and Jeannerod 1985). The nodulus and ventral uvula are known to be important for mediating VOR habituation (Halstead et al. 1937; Singleton 1967; Waespe et al. 1985). Whether these structures also play a role in adaptive modification of the gain of the VOR is not known. They are not required for cross-axis adaptive plasticity (Mason and Baker 1989). Studies of adaptation have generally focussed on identifying the location of modifiable gain elements that govern visual vestibular interaction, but there are still a number of unresolved questions and controversies. VOR adaptation has alternately been modelled as a change in Purkinje cell output (Ito 1982) or as a variable gain element after velocity storage has combined with the direct pathway, close to the output velocity command of the VOR (Lisberger et al. 1984; Lisberger and Pavelko 1988; Lisberger 1988). It has been suggested that both VOR gain and time constant may be modified concurrently (Jeannerod et al. 1976), which would indicate that common gain elements control the slow and rapid components of the VOR. Lisberger et al. (1981) demonstrated that VOR and O K A N gains were adapted together, in apparent agreement with this. Many authors have proposed that the slow component of the VOR is also responsible for O K A N (Raphan et al. 1979; Demer 1981; Waespe et al. 1983; Cohen et al. 1987; Buizza et al. 1988; Katz et al. 1991), but this has also been disputed (Clement et al. 1981; Skavenski et al. 1981). The aims of this study were to characterize vestibular habituation induced by steps of angular velocity, to determine if adaptation and habituation were separate processes and to model and predict this behavior. We also wished to determine how lesions of the nodulus and ventral uvula would affect these processes, and compare them with the effects of flocculectomy and paraflocculectomy (Waespe et al. 1983; Lisberger et al. 1984).
Materials and methods Data from four rhesus monkeys (Macaca mulatta, M1170, M1171, M 1173, M 1175) and five cynomolgus monkeys (Macacafascicularis, M 1178, M 1186, M 1190, M 1195, M1197) were used in these studies. We also utilized relevant data from rhesus monkeys used in a previous work on flocculectomy (Waespe et al. 1983) and from an accompanying study on nodulouvulectomy (Cohen, Waespe and Raphan, unpublished data). Magnetic scleral search coils were attached to one eye under anesthesia (6 mg/kg ketamine, 1.6 mg/kg xylazine intravenously; maintenance doses as required) and sterile surgical conditions to record eye movements. Stainless steel screws were fixed to the skull with dental acrylic cement. Stainless steel bolts, embedded in acrylic cement on the skull, were used to restrain the monkeys' heads painlessly during testing and to secure magnifying and reducing lenses. Monkeys received analgesics and antibiotics for several days after surgery to alleviate pain and counter infection (morphine sulfate 2 mg intramuscularly x 2; cephazolin 0.1 gin, intramuscularly twice daily x 5). During testing, animals received steps of angular velocity about a vertical axis from 30~ to 150~ in a three-axis vestibular and optokinetic stimulator (see Raphan et al. 1981 and Dai et al. 1991 for
528 a description) or on a one-axis rate table (Raphan et al. 1979). In both stimulators the monkey was surrounded by an optokinetic drum with 10~ stripes. Chair accelerations and decelerations were approximately 1000~ 2 at the beginningand end of rotation, approximating a step in velocity. We define the nystagmus induced by rotation at a constant velocity as the step response. Techniques for recording and differentiating eye movements have been described in detail in previous papers (Cohen et al. 1977, 1987; Raphan et al. 1979; Waespe et al. 1983; Katz et al. 1991; Dai et al. 1991). In brief, eye position was recorded by amplifiers with a bandwidth of DC to 35 Hz and was electronically differentiated by a filtered system with a 38 ms time constant to obtain the slow-phase eye velocity. Eye velocity was calibrated by rotating the animals around a vertical axis at 30~ in light. It was assumed that the gain of compensatory eye movements (eye velocity/stimulus velocity) was close to unity in this condition (Raphan et al. 1979). Voltages representing horizontal and vertical eye position and chair and drum velocity were displayed on a chart recorder and stored on FM magnetic tape. Rhesus monkeys received up to 0.2 mg/kg of amphetamine sulfate 45 min before testing, to maintain alertness. The cynomolgus monkeys remained alert without medication. Gains were calculated by comparing the initial jump in the slowphase eye velocity at the onset or end of rotation in darkness with the stimulus velocity, normalized with reference to the calibration trial. We used the dominant time constant as the parameter of habituation because it accounts for both magnitude and duration of response and is unaffected by small amounts of residual nystagmus. According to the model of the VOR proposed by Raphan et al. (1979), the step response of the VOR is composed of two time constants, one representing activity from the cupula and the second from the velocity storage integrator. The time constant of a composite single exponential fit to the data was used to approximate this response, and we refer to it as the dominant time constant. This was obtained by measuring the area under the slow-phase velocity envelope with a digital planimeter and dividing it by the peak slow-phase eye velocity at the onset or end of rotation (Cohen et al. 1977; Raphan et al. 1979). Measurement errors of the area were not larger than about 5%. Because of technical limitations of the planimeter, time constants of less than 2.5 s were not measured reliably. Therefore, this value served as a minimal level. To produce habituation, monkeys were given 20 steps of velocity about a vertical axis in darkness (VOR trials), as well as 8 steps of surround velocity in light at 30~ and 60~ (OKN trials). Two naive monkeys were also studied over five consecutive days, with trials in randomized order, to test for an order effect of velocity. The gain of the VOR was adapted by having the animals wear magnifying or reducing spectacles attached to the implanted head bolts. The lenses, obtained from Designs for Vision, Ronkonkoma, NY, were x 2.2 telescopic (magnifying) and x 0.5 reverse telescopic (reducing) lenses with 16~ and 72~ monocular fields of view, respectively. A third set of x 1.0 spectacles, with a 50~ field of view, controlled the potential effects of constriction of the visual field of the magnifying lenses. There were none. The lenses were housed in black aluminum frames that eliminated peripheral vision. The edges of the apparatus were padded with black rubber to reduce light leakage and enhance wearing comfort. A line drawing of the magnifying lenses is shown in the inset of Fig. 5. Monkeys naive to vestibular testing have large changes in time constant at the onset of testing. To hold this variable constant in our initial studies of adaptation, we first habituated VOR time constants to about 50% of their original value. In each adaptation session the monkey was pretested with steps of velocity and fitted with a pair of lenses. It was then returned to its cage and encouraged to manipulate small objects such as food pellets and to move about normally. At appropriate times, the animal was placed in the vestibular stimulator and its head was fixed. The lenses were removed in darkness, and steps of velocity were repeated. Each monkey wore the magnifying lenses and reducing lenses for periods from 20min to 120 h. Pilot studies showed no order effect of duration of lens exposure, provided the monkey had readapted back to normal before the next exposure period.
We also adapted the VOR gain in an animal naive to vestibular testing, by rotating it and the visual surround at the same velocity at 0.25 Hz for several hours. This "forced rotation" caused the VOR gain to drop about 25-30% over a 4-h period (Miles and Eighmy 1980). The effects of gain adaptation on the VOR time constant were studied in this paradigm using constant-velocity rotation at 60~ in darkness (steps of velocity) before and at hourly intervals during the adaptation procedure. The results were confirmed in other animals. Following completion of testing, the nodulus and uvula were ablated in two rhesus monkeys (Ml173 and Ml175) under anesthesia using sterile surgical conditions through a posterior fossa approach. This procedure is described elsewhere (Waespe et al. 1985). Starting several weeks later the animals were retested for their ability to habituate and modify the VOR time constant and gain. Data after nodulouvulectomy were compared to data from a rhesus monkey that had had its flocculus and paraflocculus removed bilaterally. The surgical technique and the extent of lesion in this monkey (M 1) have been reported previously (Waespe et al. 1983). At the conclusion of lesion experiments, the animals were deeply anesthetized and perfused through the heart with saline and 10% formalin. The brains were removed, sectioned and stained with eresyl violet, and the lesions were reconstructed.
Results
Effects of habituation on the VOR I n n o r m a l a n i m a l s the initial j u m p in velocity at the onset of p e r r o t a t o r y n y s t a g m u s is followed by a plateau that m a i n t a i n s eye velocity close to the initial velocity for a b o u t 5 s (Fig. 1A). The response then decays to zero with a d o m i n a n t time c o n s t a n t of a b o u t 15-30 s ( R a p h a n et al. 1979). The p o s t r o t a t o r y response is a m i r r o r image, generally h a v i n g the. same gain, time c o n s t a n t a n d d y n a m i c behavior. W i t h repeated testing, p e r r o t a t o r y a n d postrotatory n y s t a g m u s habituates, a n d the character of the response is altered. I n one m o n k e y the time c o n s t a n t was reduced from 29 s in the first test (Fig. 1A) to 6 s, 91 sessions later (Fig. 1B), a n d the plateau was drastically reduced. The gain was unaffected. C h a n g e s in time c o n s t a n t as a function of time (Fig. 2A, C) a n d session (Fig. 2B, D) after the onset of testing were generally similar for the two rhesus m o n k e y s most extensively tested (Ml173: Figs. 2A, B; M l 1 7 5 : Fig. 2C, D). There were differences, however. I n M l 1 7 3 the time c o n s t a n t declined rapidly initially a n d then fell over a time course of a b o u t 550 days (Fig. 2A) in 91 trials (Fig. 2B). This trend was m a i n t a i n e d despite relatively long periods of inactivity. F o r example, the V O R time c o n s t a n t held at a b o u t 15 s from the 75th to the 175th day d u r i n g an i n t e r r u p t i o n in testing due to e q u i p m e n t p r o b lems (indicated by the filled arrow o n the left side of Fig. 2A). Testing resumed in a c o n c e n t r a t e d fashion at a b o u t 175 days, a n d the time c o n s t a n t fell from a b o u t 15 to 7 s over a 25-day period. It r e m a i n e d there d u r i n g a second hiatus of 80 days (open arrow), until intensive testing resumed at 325 days, which drove the time c o n s t a n t to 6 s over a b o u t 575 days, when a n o d u l o u v u l e c t o m y was performed. Repeated testing caused less change in the time course of the V O R in M l 1 7 5 . The time c o n s t a n t fell from a n initial value of 19-20 to 11 s in 950 days (Fig. 2C) over 58 trials (Fig. 2D). H a b i t u a t i o n was n o t m a i n t a i n e d as well,
529
Pre-Op
A
1ST SESSION
0
0
t
BrainR 9 Springer-Verlag 1992
Habituation and adaptation of the vestibuloocular reflex: a model of differential control by the vestibu|ocerebellum Helen Cohen 1' *, Bernard Cohen 1' z, Theodore Raphan 3, and Walter Waespe 4 1 Department z Department 3 Department 4 Department
of Neurology, The Mount Sinai School of Medicine, New York, USA of Physiology and Biophysics, The Mount Sinai School of Medicine, New York, USA of Computer and Information Science, Brooklyn College, City University of New York, USA of Neurology, University of Zfirich, Switzerland
Received December 11, 1991/Accepted April 9, 1992
Summary. We habituated the dominant time constant of the horizontal vestibuloocular reflex (VOR) of rhesus and cynomolgus monkeys by repeated testing with steps of velocity about a vertical axis and adapted the gain of the VOR by altering visual input with magnifying and reducing lenses. After baseline values were established, the nodulus and ventral uvula of the vestibulocerebellum were ablated in two monkeys, and the effects of nodulouvulectomy and flocculectomy on VOR gain adaptation and habituation were compared. The VOR time constant decreased with repeated testing, rapidly at first and more slowly thereafter. The gain of the VOR was unaffected. Massed trials were more effective than distributed trials in producing habituation. Regardless of the schedule of testing, the VOR time constant never fell below the time constant of the semicircular canals ( ~ 5 s). This finding indicates that only the slow component of the vestibular response, the component produced by velocity storage, was habituated. In agreement with this, the time constant of optokinetic after-nystagmus (OKAN) was habituated concurrently with the VOR. Average values for VOR habituation were obtained on a per session basis for six animals. The VOR gain was adapted by natural head movements in partially habituated monkeys while they wore x 2.2 magnifying or x 0.5 reducing lenses. Adaptation occurred rapidly and reached about _ 30%, similar to values obtained using forced rotation. VOR gain adaptation did not cause additional habituation of the time constant. When the VOR gain was reduced in animals with a long VOR time constant, there were overshoots in eye velocity that peaked at about 6-8 s after the onset or end of constant-velocity rotation. These overshoots occurred at times when the velocity storage integrator would have been maximally activated by semicircular canal input. Since the activity generated in the canals is not * Present Address: Department of Otolaryngology, Baylor College of Medicine, 1 Baylor Plaza, Houston, TX 77030, USA Offprint requests to. Bernard Cohen, Box 1135, Annenberg 21-74, Mount Sinai School of Medicine, 1 East 100th Street, New York, NY 10029, USA. Correspondence to: B. Cohen
altered by visual adaptation, this finding indicates that the gain element that controls rapid changes in eye velocity in the VOR is separate from that which couples afferent input to velocity storage. Nodulouvulectomy caused a prompt and permanent loss of habituation, returning VOR time constants to initial values. VOR gain adaptation, which is lost after flocculectomy, was unaffected by nodulouvulectomy. Flocculectomy did not alter habituation of the VOR or of OKAN. Using a simplified model of the VOR, the decrease in the duration of vestibular nystagmus due to habituation was related to a decrement in the dominant time constant of the velocity storage integrator (1/ho). Nodulouvulectomy, which reversed habituation, would be effected by decreasing ho, thereby increasing the VOR time constant. Small values of h o would cause velocity storage to approach an ideal integrative process, leading the system to become unstable. By controlling the VOR time constant through habituation, the nodulus and uvula can stabilize the slow component of the VOR. VOR gain adaptation was related to a modification of the direct vestibular path gain gl, without altering the coupling to velocity storage go or its time constant (1/ho). The mismatched direct- and indirect- pathway gains simulated the overshoots in the dynamic response to a step in velocity, that were observed experimentally. We conclude that independent distributed elements in the VOR modify its dynamic response, under control of separate parts of the vestibulocerebellum.
Key words: Habituation Adaptation Vestibuloocular reflex (VOR) - Velocity storage - Nystagmus - Semicircular canals Visual system - Monkeys - Optokinetic nystagmus (OKN) - After-nystagmus (OKAN) - Nodulus - Flocculus - Vestibulocerebellum
Introduction The dominant behavior of the vestibuloocular reflex (VOR) can be characterized by two properties, its gain and
527 time constant. These describe the dynamics of ocular compensation due to head rotation. The gain of the VOR, represented by the ratio of eye velocity to head velocity, is a measure of the goodness of compensation. The dominant time constant determines the frequency range of compensation, a long time constant corresponding to good low-frequency compensation during rotation (Raphan et al. 1979) and circular locomotion (Solomon and Cohen 1992b). Under normal circumstances, the VOR gain and time constant are matched so that compensation is maintained close to unity in darkness by a plateau in slow-phase velocity for several seconds at the onset or end of rotation (Raphan et al. 1979). The mechanism for achieving this match and the structures that are responsible for it are not known. Transient changes in VOR gain occur when close targets are viewed during angular head movement (Viirre et al. 1986) and during head rotations that include linear acceleration and translation (Bronstein and Gresty 1991; Solomon and Cohen 1992a). The dominant time constant of the VOR and of optokinetic after-nystagmus (OKAN) can also be altered transiently. The time constant of the VOR becomes shorter for trials in which the velocity of rotation is increased (Raphan et al. 1979). Likewise, the time constant of O K A N can be reduced in single trials by increasing the duration of optokinetic stimulation (Biittner et al. 1976). It is also possible to induce semipermanent changes in VOR gain, defined here as VOR gain adaptation. Retinal slip (Miles and Eighmy 1980; Miles and Lisberger 1981; Melvill Jones 1985) or an imagined mismatch between visual and vestibular signals (Melvill Jones et al. 1984) induces gain changes in the VOR to reduce gaze error. The gain changes are "plastic", in that they considerably outlast the trial in which they were induced. Most commonly, they have been produced by wearing reversingoptics lenses (Stratton 1897; Gonshor and Melvill Jones 1976a, b) or telescopic lenses (Miles and Eighmy 1980; Miles and Lisberger 1981) for prolonged periods (see Melvitl Jones 1985 for a review), but there can also be cross-axis adaptation (Schultheis and Robinson 1981; see Harrison et al. 1986 for a review). Rotation in a subjectstationary or moving surround is also a powerful stimulus for VOR gain adaptation (Ito 1982; Miles and Eighmy 1980). The ability to adapt the VOR through vision is lost after flocculus lesions (Lisberger et al. 1984). Repeated presentation of the same stimulus produces semipermanent reductions in the VOR time constant (J/iger and Henn 1981; Jeannerod et al. 1981; Schmid and Jeannerod 1985), which we define as VOR habituation. Once established, habituation can be retained for long periods of time without reinforcement (Thorpe 1956). Vestibular habituation is best induced in alert subjects (Crampton 1964). It is direction-specific and spatially oriented (Crampton 1962; Guedry 1965; Young and Henn 1976; Clement et al. 1981; Jeannerod et al. 1981; Schmid and Jeannerod 1985). It has been induced by caloric stimulation (Collins 1974; H o o d and Pfaltz 1954), and by steps (Dodge 1923; Jeannerod et al. 1976) and sinusoids of angular velocity (Dodge 1923; Clement et al. 1981; Jeannerod et al. 1981; J/iger and Henn 1981). Neither the
functional significance nor the signals that cause habituation are known, but, presumably, habituation decreases a maladaptive aspect of responses evoked by motion stimuli (Schmid and Jeannerod 1985). The nodulus and ventral uvula are known to be important for mediating VOR habituation (Halstead et al. 1937; Singleton 1967; Waespe et al. 1985). Whether these structures also play a role in adaptive modification of the gain of the VOR is not known. They are not required for cross-axis adaptive plasticity (Mason and Baker 1989). Studies of adaptation have generally focussed on identifying the location of modifiable gain elements that govern visual vestibular interaction, but there are still a number of unresolved questions and controversies. VOR adaptation has alternately been modelled as a change in Purkinje cell output (Ito 1982) or as a variable gain element after velocity storage has combined with the direct pathway, close to the output velocity command of the VOR (Lisberger et al. 1984; Lisberger and Pavelko 1988; Lisberger 1988). It has been suggested that both VOR gain and time constant may be modified concurrently (Jeannerod et al. 1976), which would indicate that common gain elements control the slow and rapid components of the VOR. Lisberger et al. (1981) demonstrated that VOR and O K A N gains were adapted together, in apparent agreement with this. Many authors have proposed that the slow component of the VOR is also responsible for O K A N (Raphan et al. 1979; Demer 1981; Waespe et al. 1983; Cohen et al. 1987; Buizza et al. 1988; Katz et al. 1991), but this has also been disputed (Clement et al. 1981; Skavenski et al. 1981). The aims of this study were to characterize vestibular habituation induced by steps of angular velocity, to determine if adaptation and habituation were separate processes and to model and predict this behavior. We also wished to determine how lesions of the nodulus and ventral uvula would affect these processes, and compare them with the effects of flocculectomy and paraflocculectomy (Waespe et al. 1983; Lisberger et al. 1984).
Materials and methods Data from four rhesus monkeys (Macaca mulatta, M1170, M1171, M 1173, M 1175) and five cynomolgus monkeys (Macacafascicularis, M 1178, M 1186, M 1190, M 1195, M1197) were used in these studies. We also utilized relevant data from rhesus monkeys used in a previous work on flocculectomy (Waespe et al. 1983) and from an accompanying study on nodulouvulectomy (Cohen, Waespe and Raphan, unpublished data). Magnetic scleral search coils were attached to one eye under anesthesia (6 mg/kg ketamine, 1.6 mg/kg xylazine intravenously; maintenance doses as required) and sterile surgical conditions to record eye movements. Stainless steel screws were fixed to the skull with dental acrylic cement. Stainless steel bolts, embedded in acrylic cement on the skull, were used to restrain the monkeys' heads painlessly during testing and to secure magnifying and reducing lenses. Monkeys received analgesics and antibiotics for several days after surgery to alleviate pain and counter infection (morphine sulfate 2 mg intramuscularly x 2; cephazolin 0.1 gin, intramuscularly twice daily x 5). During testing, animals received steps of angular velocity about a vertical axis from 30~ to 150~ in a three-axis vestibular and optokinetic stimulator (see Raphan et al. 1981 and Dai et al. 1991 for
528 a description) or on a one-axis rate table (Raphan et al. 1979). In both stimulators the monkey was surrounded by an optokinetic drum with 10~ stripes. Chair accelerations and decelerations were approximately 1000~ 2 at the beginningand end of rotation, approximating a step in velocity. We define the nystagmus induced by rotation at a constant velocity as the step response. Techniques for recording and differentiating eye movements have been described in detail in previous papers (Cohen et al. 1977, 1987; Raphan et al. 1979; Waespe et al. 1983; Katz et al. 1991; Dai et al. 1991). In brief, eye position was recorded by amplifiers with a bandwidth of DC to 35 Hz and was electronically differentiated by a filtered system with a 38 ms time constant to obtain the slow-phase eye velocity. Eye velocity was calibrated by rotating the animals around a vertical axis at 30~ in light. It was assumed that the gain of compensatory eye movements (eye velocity/stimulus velocity) was close to unity in this condition (Raphan et al. 1979). Voltages representing horizontal and vertical eye position and chair and drum velocity were displayed on a chart recorder and stored on FM magnetic tape. Rhesus monkeys received up to 0.2 mg/kg of amphetamine sulfate 45 min before testing, to maintain alertness. The cynomolgus monkeys remained alert without medication. Gains were calculated by comparing the initial jump in the slowphase eye velocity at the onset or end of rotation in darkness with the stimulus velocity, normalized with reference to the calibration trial. We used the dominant time constant as the parameter of habituation because it accounts for both magnitude and duration of response and is unaffected by small amounts of residual nystagmus. According to the model of the VOR proposed by Raphan et al. (1979), the step response of the VOR is composed of two time constants, one representing activity from the cupula and the second from the velocity storage integrator. The time constant of a composite single exponential fit to the data was used to approximate this response, and we refer to it as the dominant time constant. This was obtained by measuring the area under the slow-phase velocity envelope with a digital planimeter and dividing it by the peak slow-phase eye velocity at the onset or end of rotation (Cohen et al. 1977; Raphan et al. 1979). Measurement errors of the area were not larger than about 5%. Because of technical limitations of the planimeter, time constants of less than 2.5 s were not measured reliably. Therefore, this value served as a minimal level. To produce habituation, monkeys were given 20 steps of velocity about a vertical axis in darkness (VOR trials), as well as 8 steps of surround velocity in light at 30~ and 60~ (OKN trials). Two naive monkeys were also studied over five consecutive days, with trials in randomized order, to test for an order effect of velocity. The gain of the VOR was adapted by having the animals wear magnifying or reducing spectacles attached to the implanted head bolts. The lenses, obtained from Designs for Vision, Ronkonkoma, NY, were x 2.2 telescopic (magnifying) and x 0.5 reverse telescopic (reducing) lenses with 16~ and 72~ monocular fields of view, respectively. A third set of x 1.0 spectacles, with a 50~ field of view, controlled the potential effects of constriction of the visual field of the magnifying lenses. There were none. The lenses were housed in black aluminum frames that eliminated peripheral vision. The edges of the apparatus were padded with black rubber to reduce light leakage and enhance wearing comfort. A line drawing of the magnifying lenses is shown in the inset of Fig. 5. Monkeys naive to vestibular testing have large changes in time constant at the onset of testing. To hold this variable constant in our initial studies of adaptation, we first habituated VOR time constants to about 50% of their original value. In each adaptation session the monkey was pretested with steps of velocity and fitted with a pair of lenses. It was then returned to its cage and encouraged to manipulate small objects such as food pellets and to move about normally. At appropriate times, the animal was placed in the vestibular stimulator and its head was fixed. The lenses were removed in darkness, and steps of velocity were repeated. Each monkey wore the magnifying lenses and reducing lenses for periods from 20min to 120 h. Pilot studies showed no order effect of duration of lens exposure, provided the monkey had readapted back to normal before the next exposure period.
We also adapted the VOR gain in an animal naive to vestibular testing, by rotating it and the visual surround at the same velocity at 0.25 Hz for several hours. This "forced rotation" caused the VOR gain to drop about 25-30% over a 4-h period (Miles and Eighmy 1980). The effects of gain adaptation on the VOR time constant were studied in this paradigm using constant-velocity rotation at 60~ in darkness (steps of velocity) before and at hourly intervals during the adaptation procedure. The results were confirmed in other animals. Following completion of testing, the nodulus and uvula were ablated in two rhesus monkeys (Ml173 and Ml175) under anesthesia using sterile surgical conditions through a posterior fossa approach. This procedure is described elsewhere (Waespe et al. 1985). Starting several weeks later the animals were retested for their ability to habituate and modify the VOR time constant and gain. Data after nodulouvulectomy were compared to data from a rhesus monkey that had had its flocculus and paraflocculus removed bilaterally. The surgical technique and the extent of lesion in this monkey (M 1) have been reported previously (Waespe et al. 1983). At the conclusion of lesion experiments, the animals were deeply anesthetized and perfused through the heart with saline and 10% formalin. The brains were removed, sectioned and stained with eresyl violet, and the lesions were reconstructed.
Results
Effects of habituation on the VOR I n n o r m a l a n i m a l s the initial j u m p in velocity at the onset of p e r r o t a t o r y n y s t a g m u s is followed by a plateau that m a i n t a i n s eye velocity close to the initial velocity for a b o u t 5 s (Fig. 1A). The response then decays to zero with a d o m i n a n t time c o n s t a n t of a b o u t 15-30 s ( R a p h a n et al. 1979). The p o s t r o t a t o r y response is a m i r r o r image, generally h a v i n g the. same gain, time c o n s t a n t a n d d y n a m i c behavior. W i t h repeated testing, p e r r o t a t o r y a n d postrotatory n y s t a g m u s habituates, a n d the character of the response is altered. I n one m o n k e y the time c o n s t a n t was reduced from 29 s in the first test (Fig. 1A) to 6 s, 91 sessions later (Fig. 1B), a n d the plateau was drastically reduced. The gain was unaffected. C h a n g e s in time c o n s t a n t as a function of time (Fig. 2A, C) a n d session (Fig. 2B, D) after the onset of testing were generally similar for the two rhesus m o n k e y s most extensively tested (Ml173: Figs. 2A, B; M l 1 7 5 : Fig. 2C, D). There were differences, however. I n M l 1 7 3 the time c o n s t a n t declined rapidly initially a n d then fell over a time course of a b o u t 550 days (Fig. 2A) in 91 trials (Fig. 2B). This trend was m a i n t a i n e d despite relatively long periods of inactivity. F o r example, the V O R time c o n s t a n t held at a b o u t 15 s from the 75th to the 175th day d u r i n g an i n t e r r u p t i o n in testing due to e q u i p m e n t p r o b lems (indicated by the filled arrow o n the left side of Fig. 2A). Testing resumed in a c o n c e n t r a t e d fashion at a b o u t 175 days, a n d the time c o n s t a n t fell from a b o u t 15 to 7 s over a 25-day period. It r e m a i n e d there d u r i n g a second hiatus of 80 days (open arrow), until intensive testing resumed at 325 days, which drove the time c o n s t a n t to 6 s over a b o u t 575 days, when a n o d u l o u v u l e c t o m y was performed. Repeated testing caused less change in the time course of the V O R in M l 1 7 5 . The time c o n s t a n t fell from a n initial value of 19-20 to 11 s in 950 days (Fig. 2C) over 58 trials (Fig. 2D). H a b i t u a t i o n was n o t m a i n t a i n e d as well,
529
Pre-Op
A
1ST SESSION
0
0
t
