Optokinetic-vestibular interaction in patients with increased gain of the vestibulo-ocular reflex

We studied optokinetic nystagmus (OKN), optokinetic afternystagmus (OKAN) and visual-vestibular interaction in five patients with markedly elevated vestibulo-ocular reflex (VOR) gain due to cerebellar atrophy. All had impaired smooth pursuit, decreased initial slow phase velocity of OKN, and impaired ability to suppress the VOR with real or imagined targets. OKN slow phase velocity gradually built up over 25–45 s, reaching normal values for low stimulus velocities (30 deg/s). Initial velocity of OKAN was increased, but the rate of decay of OKAN was normal. These findings can be explained by models that include separate velocity storage and variable gain elements shared by the vestibular and optokinetic systems.     

Optokinetic nystagmus and afternystagmus in human beings: relationship to nonlinear processing of information about retinal slip

Summary In four normal human subjects we measured eye movements during full-field optokinetic stimulation (10–220 deg/s) and determined the relationship among retinal-slip velocity (drum velocity minus slow-phase eye velocity), the slow-phase velocity of optokinetic nystagmus (OKN) and the initial value of the slow-phase velocity of optokinetic afternystagmus (OKAN) measured in darkness. OKN and OKAN were maximum (63–84 and 11–19 deg/s, respectively) when retinal slip ranged from 30–100 deg/s. For higher values of retinal slip, OKN and OKAN fell (in 3 subjects) or reached a plateau (in the fourth). The amplitude of OKAN in human beings was much less than that reported in monkeys. The shape, however, of the curve relating retinal slip to the amplitude of OKAN was similar to that of monkeys. Furthermore, in both cases the curve resembles that obtained by plotting the results of experimental recordings of neural discharge in the nucleus of the optic tract as a function of retinal slip. These results imply that the processing of visual information for generation of OKAN is similar in monkeys and human beings but that the gain of the system is much less in human beings. We also found that fixation of a small target during optokinetic stimulation nearly completely prevented the development of OKAN while fixation of a small target for short periods after optokinetic stimulation did not alter the pattern of decay of OKAN. Thus, fixation may actively prevent the coupling of visual information into the velocity-storage mechanism.     

Neuronal substrates of motor learning in the velocity storage generated during optokinetic stimulation in the squirrel monkey

Chronic motor learning in the vestibuloocular reflex (VOR) results in changes in the gain of this reflex and in other eye movements intimately associated with VOR behavior, e.g., the velocity storage generated by optokinetic stimulation (OKN velocity storage). The aim of the present study was to identify the plastic sites responsible for the change in OKN velocity storage after chronic VOR motor learning. We studied the neuronal responses of vertical eye movement flocculus target neurons (FTNs) during the optokinetic after-nystagmus (OKAN) phase of the optokinetic response (OKR) before and after VOR motor learning. Our findings can be summarized as follows. 1) Chronic VOR motor learning changes the horizontal OKN velocity storage in parallel with changes in VOR gain, whereas the vertical OKN velocity storage is more complex, increasing with VOR gain increases, but not changing following VOR gain decreases. 2) FTNs contain an OKAN signal having opposite directional preferences after chronic high versus low gain learning, suggesting a change in the OKN velocity storage representation of FTNs. 3) Changes in the eye-velocity sensitivity of FTNs during OKAN are correlated with changes in the brain stem head-velocity sensitivity of the same neurons. And 4) these changes in eye-velocity sensitivity of FTNs during OKAN support the new behavior after high gain but not low gain learning. Thus we hypothesize that the changes observed in the OKN velocity storage behavior after chronic learning result from changes in brain stem pathways carrying head velocity and OKN velocity storage information, and that a parallel pathway to vertical FTNs changes its OKN velocity storage representation following low, but not high, gain VOR motor learning.     

Contribution of eye muscle proprioception to velocity-response characteristics of eye movements: involvement of the cerebellar flocculus.

The vestibulo-ocular reflex (VOR) and optokinetic nystagmus (OKN) were examined in alert pigmented rabbits following interruption of proprioceptive afferents from the extraocular muscles in one eye by surgical section of the ophthalmic branch of the trigeminal nerve (V1 nerve). Deficits were mainly produced in movement dynamics of the ipsilateral eye including reduction of (1) the VOR gain at a high frequency of head rotation, (2) the OKN gain and (3) the velocity of quick eye movements in the OKN. In some of the rabbits examined, the cerebellar flocculus was lesioned by local injection of kainic acid before severance of the V1 nerve. No significant additional reductions of VOR or OKN gains were produced by V1 nerve section in the flocculus-lesioned rabbits. These results suggest that proprioceptive signals from eye muscles act to improve VOR and OKN dynamics through the neuronal mechanisms involving the cerebellar flocculus.     

The horizontal optokinetic nystagmus in the cat

Summary 1)Horizontal optokinetic eye nystagmus (OKN) and afternystagmus (OKAN) were recorded in the alert cat (head restrained) in response to velocity steps and sinusoidal optokinetic stimuli. 2)A strong dependency of OKN performance on stimulus pattern was found: responses were most regular and gain was high over a large range of stimulus velocities when the stimulus consisted of a high-contrast random dot pattern. 3) Following velocity steps, OKN showed a small amplitude fast rise in slow phase velocity (SPV) which was followed by a slow build-up to steady state. The amplitude of the initial jump in SPV increased with stimulus amplitude up to 30°/s and saturated afterwards. The plateau level of initial SPV ranged from 5 to 15°/s. 4) The slow build-up of SPV showed non-linearities, i.e. the time to steady state increased with stimulus amplitude and the slow rise of SPV was irregular. In most animals steady state SPV showed no signs of response saturation for step amplitudes up to 60–80°/s or more. The open-loop gain (steady state SPV/ retinal slip velocity) dependend on retinal slip velocity and decreased from 46 at 0.5°/s to 0.4 at about 60°/s. 5) OKAN I and II were consistently observed and occasionally OKAN III was noted. OKAN I durations (mean 13.8 +- 5.1 s) and OKAN II amplitudes were independent of stimulus magnitude. Initial SPV of OKAN I was typically the same as that of OKN, i.e. no fast fall was observed. Cessation of pattern rotation in light, however, produced a fast initial decay of SPV. 6) A least square fitting of OKAN time course was performed with various time functions. The SPV of OKAN I and II was best fitted with a damped sine wave, indicating that cat optokinetic system behaves like a second order underdamped system. 7) Sinusoidal stimuli produced strong response non-linearities. At a given frequency gain decreased with increasing stimulus amplitudes. Gain correlated best with stimulus acceleration. In addition, strong stimuli produced characteristic response distortions. 8) In the visual-vestibular conflict situation vectorial summation of VOR and OKN was observed only with small stimuli.Supported by grants nos. 3.505.79 and 3.403.83 from the Swiss National Science Foundation and Dr. Erik Slack-Gyr Foundation     

Longterm impairment of cat optokinetic nystagmus following visual cortical lesions

Summary Binocular and monocular gain of optokinetic nystagmus (OKN), OKN dynamics, vestibulo-ocular reflex (VOR) and VOR adaptation were measured in 5 normal cats and in 5 cats which underwent bilateral visual cortical lesions involving the 17–18 complex at least 4 months before testing. We observed longterm deficits after bilateral lesions involving area 17 and variable parts of area 18 but failed to observe deficits after 18–19 lesions. These deficits were limited to the OKN gain and the build-up time constant of OKN; the VOR and the optokinetic after-nystagmus (OKAN) time constant were within normal limits. Our results suggest that areas 17–18 operate in parallel to control the encoding of retinal slip velocity at the level of the nucleus of the optic tract (NOT) and the accessory optic system (AOS), which are known to represent the initial stage of the optokinetic pathways.     

Velocity storage in the vestibulo-ocular reflex arc (VOR)

Summary Vestibular and optokinetic nystagmus (OKN) of monkeys were induced by platform and visual surround rotation. Vision prolonged per-rotatory nystagmus and cancelled or reduced post-rotatory nystagmus recorded in darkness. Presumably, activity stored during OKN summed with activity arising in the semicircular canals. The limit of summation was about 120 °/s, the level of saturation of optokinetic after-nystagmus (OKAN). OKN and vestibular nystagmus, induced in the same or in opposite directions diminished or enhanced post-rotatory nystagmus up to 120 °/s. We postulate that a common storage mechanism is used for producing vestibular nystagmus, OKN, and OKAN. Evidence for this is the similar time course of vestibular nystagmus and OKAN and their summation. In addition, stored activity is lost in a similar way by viewing a stationary surround during either OKAN or vestibular nystagmus (fixation suppression).These responses were modelled using direct pathways and a non-ideal integrator coupled to the visual and peripheral vestibular systems. The direct pathways are responsible for rapid changes in eye velocity while the integrator stores activity and mediates slower changes. The integrator stabilizes eye velocity during whole field rotation and extends the time over which the vestibulo-ocular reflex can compensate for head movement.     

Torsional and horizontal vestibular ocular reflex adaptation: three-dimensional eye movement analysis

This study used visual-vestibular conflict to effect short-term torsional and horizontal adaptation of the vestibulo-ocular reflex (VOR). Seven normal subjects underwent sinusoidal whole-body rotation about the earth-vertical axis for 40 min (±37°/s, 0.3 Hz) while viewing a stationary radial pattern fixed to the chair (×0 viewing). During adaptation and testing in darkness, the head was pitched either up or down 35° to excite both the horizontal and torsional VOR. The eyes were kept close to zero orbital elevation. Eye movements were recorded with a dual search coil in a three-field magnetic system. VOR gain was determined by averaging peak eye velocity from ten cycles of chair oscillation in complete darkness. The gain of the angular horizontal VOR (response to rotation about the head rostral-caudal axis) was significantly reduced after training in both head orientations. Angular torsional VOR gain (head rotation about the naso-occipital axis) was reduced in both head orientations, but this reached statistical significance only in the head down position. These results suggest that torsional and horizontal VOR gain adaptation, even when elicited together, may be subject to different influences depending upon head orientation. Differences between head up and down could be due to the relatively greater contribution of the horizontal semicircular canals with nose-down pitch. Alternatively, different VOR-adaptation processes could depend on the usual association of the head down posture to near viewing, in which case the torsional VOR is relatively suppressed.     

Torsional eye movements during psychophysical testing with rotating patterns

Torsional eye movements were measured while subjects viewed a large, high contrast windmill pattern rotating at 53°/s or a small (5° diameter) dot pattern rotating at 115°/s. Both stimuli generated rotational eye movements consisting of torsional optokinetic nystagmus (tOKN) superimposed on a slow torsional drift in the direction of pattern rotation. With the wide-field windmill stimulus, torsional drifts of up to 7° over 20 s were found. The dot pattern produced drifts of up to 2° over 5–20 s. In both cases, the slow-phase speeds during tOKN were low (0.5–1°/s). We conclude that reductions in slip speed are minimal with rotating stimuli, so torsional eye speeds will have a minimal effect on investigations of rotational motion aftereffect strength and perceived speed. While the slow-phase tOKN gain is low, the slow drift in torsional eye position will have significant effects on psychophysical results when the tests rely on keeping selected regions of the stimulus confined to specific areas of the retina, as is the case for phantom or remote motion aftereffects.     

Difference between horizontal and vertical optokinetic nystagmus in cats at upright position

The slow-phase velocity (SPV) of optokinetic nystagmus (OKN) and optokinetic after nystagmus (OKAN) in response to a velocity step of surround rotation in the horizontal direction is composed of the rapid and slow components in the cat: a rapid rise, a slow rise to a steady state, a rapid fall, and a slow decline to 0 deg/s. The rapid and slow components are attributed to the direct pathway and velocity storage neuronal mechanisms, respectively. The difference between horizontal and vertical OKN has been reported in the monkey at the upright position, but the slow and rapid components have not been distinguished. The present study compared horizontal OKN-OKAN with vertical OKN-OKAN in the cat at the upright position, distinguishing the rapid and slow components. Constant velocity rotation of a random dot pattern at a velocity of 5 to 160 deg/s was used for optokinetic stimulation. The results: First, the amplitude of the rapid rise was relatively small in all SPV directions and all stimulus velocities investigated, with a slight upward-SPV preference to the downward-SPV (maximum 6.4, 6.0, and 3.4 deg/s in horizontal, upward, and downward SPV directions, respectively). Second, the steady state velocity was large during horizontal OKN (maximum 69.0 deg/s), small during upward-SPV OKN (12.9 deg/s), and missing (SPV is negligibly small and irregular) during downward-SPV OKN, indicating a large directional difference of OKN. Third, the acceleration of the slow rise decreased with the stimulus velocity at higher stimulus velocities >20 deg/s during both horizontal and upward-SPV OKN, suggesting strong nonlinearity in the velocity charge system. Fourth, the decay time course of the OKAN was described by the time constant of the exponential function, and the time constant was longer during horizontal (mean, 8.3 s at a stimulus velocity of 20 deg/s) than during upward-SPV (5.4 s) OKAN, suggesting that the velocity discharge system is relatively linear compared with the velocity charge system. It is concluded that horizontal OKN-OKAN is much larger than vertical OKN-OKAN in the cat at the upright position, and this directional difference is caused mainly by the directional difference in the velocity storage mechanism, but not in the direct pathway mechanism.     

Optokinetic eye movements elicited by an apparently moving visual pattern in guinea pigs

Based on known anatomical and physiological properties, guinea pigs could be expected to show a lack of cortical impact on the basic properties of optokinetic nystagmus (OKN) such as eye movements elicited by a rotating striped drum. The present study aimed to clarify the question of whether the guinea pig's brainstem-mediated circuit is capable of generating horizontal OKN with the animal binocularly viewing a stationary stroboscopically illuminated striped pattern (apparent-motion OKN). A striped drum with a pattern periodicity of 2.37 degrees was rotated around the animal. The OKN buildup under real stimulation conditions was found to be largely devoid of a rapid initial rise in slow-phase eye velocity (direct OKN component). Only in a few recordings was a slight initially faster acceleration seen eliciting slow-phase eye velocities of 2 degrees/s at the most. The lack of a rapid initial rise proves that the optokinetic reflex loop does not possess a functionally significant cortically mediated component. To achieve apparent-motion OKN, initially in each recording real-motion OKN was elicited by the constantly illuminated rotating pattern. Then, 60 or 90 s later the drum was stopped and the illumination switched over from constant to flashing light. Following Grüsser and Behrens (1979), we calculated an apparent stimulus velocity (= flash frequency x 2.37 degrees) and an apparent OKN gain. We found that all animals continued in nystagmus under the flashing light condition. The slow-phase eye velocity was the same as under real pattern motion, with gain values of between 0.36 and 0.7. In two animals we produced an apparent gain close to unity by suddenly doubling the flash interval, i.e., halving the apparent drum velocity. This gain, however, did not remain at a high level: The slow-phase eye velocity decreased until the previous lower gain was reached again. The data indicate that cortical mechanisms are not necessary to produce nystagmus under apparent-motion stimulation and neither phi nor sigma phenomena of motion perception satisfactorily explain the apparent-motion OKN in the guinea pig. Our data provide evidence that the brainstem-located velocity storage mechanism plays a decisive role in keeping the eye in motion under apparent-motion stimulation.     

Optokinetic nystagmus in the rabbit and its modulation by bilateral microinjection of carbachol in the cerebellar flocculus

Summary 1. In the alert, pigmented rabbit, eye movements were recorded during optokinetic nystagmus (OKN) and during optokinetic afternystagmus (OKAN). These responses were elicited by steps in surround-velocity ranging from 5–110°/s during binocular as well as monocular viewing. 2. In the baseline condition, OKN showed an approximately linear build-up of eye velocity to a steady-state, followed by a linear decay of eye velocity during OKAN after the lights were turned off. Build-up during binocular viewing was characterized by a constant, maximum eye-acceleration (about 1°/s²) for stimulus velocities up to 60°/s. OKAN, instead, was characterized by a fixed duration (about 10 s) for stimulus velocities up to 20°/s. Steady-state eye velocity saturated at about 50°/s. 3. Monocular stimulation in the preferred (nasal) direction elicited a build-up that was on average twice as slow as during binocular stimulation. Steady-state velocity during monocular stimulation saturated at about 20°/s. OKAN was of equal duration as during binocular stimulation. In the non-preferred direction, a very irregular nystagmus was elicited without velocity build-up. The stronger response to binocular stimulation, compared to the responses under monocular viewing condition in either nasal and temporal direction suggests potentiation of the signals of either eye during binocular viewing. 4. OKN and OKAN were re-assessed after intra-floccular microinjection of the nonselective cholinergic agonist carbachol. In the binocular viewing condition, eye-acceleration during build-up was strongly enhanced from 1°/s² before to 2.5°/s² after injection. The saturation level of steady-state eye velocity was also increased, from 50°/s before to more than 60°/s after carbachol. The duration of OKAN, however, was shortened from 10 s before to 6 s after injection. The response to monocular stimulation in the preferred direction revealed similar changes. 5. The flocculus appears to be involved in the control of the dynamics of OKN in the rabbit. Cholinergic mechanisms affect the floccular control of the rate at which slow-phase velocity can be built up and the rate of decay of eye velocity during OKAN. Cholinergic stimulation of the flocculus enhances the dynamics of OKN, while velocity storage is shortened.     

Unilateral cholinergic stimulation of the rabbit's cerebellar flocculus: asymmetric effects on optokinetic responses

Summary In previous work, we have demonstrated an acceleration of the buildup of slow-phase velocity of optokinetic nystagmus (OKN) after bilateral floccular injection of the aselective cholinergic agonist carbachol (Tan and Collewijn 1991; Tan et al. 1992a). In the present study we investigated the effects of unilateral floccular injections of carbachol. Such unilateral injections specifically enhanced the buildup of OKN slow-phase velocity in the direction toward the injected flocculus (ipsiversive). During binocular optokinetic stimulation, this enhancement was expressed in the motion of both eyes. Acceleration of the eye contralateral to the injected flocculus increased from 1 to about 2°/s², while the acceleration of the ipsilateral eye increased from 1 to about 1.5°/s². In contrast, buildup of contraversive OKN was unchanged. No changes were found in the steady-state OKN and optokinetic afternystagmus (OKAN). Monocular optokinetic stimulation was only effective in the nasal direction, and the effects of unilateral injection of carbachol were disconjugate. Ipsiversive OKN was enhanced only in the contralateral, seeing eye, while the response of the ipsilateral, covered eye was unchanged. We hypothesize that the directionally specific effect of unilateral cholinergic floccular stimulation on OKN is due to enhancement of predominantly the excitatory phase of modulation of the Purkinje cell's simple-spike activity by carbachol, without a marked effect of carbachol on the inhibitory phase of simple-spike modulation.     

Stimulation of the nodulus and uvula discharges velocity storage in the vestibulo-ocular reflex

The nodulus and sublobule d of the uvula of rhesus and cynomolgus monkeys were electrically stimulated with short trains of pulses to study changes in horizontal slow-phase eye velocity. Nodulus and uvula stimulation produced a rapid decline in horizontal slow phase velocity, one aspect of the spatial reorientation of the axis of eye rotation that occurs when the head is tilted with regard to gravity during per- and post-rotatory nystagmus and optokinetic after-nystagmus (OKAN). Nodulus and uvula stimulation also reproduced the reduction of the horizontal time constant of post-rotatory nystagmus and OKAN that occurs during visual suppression. The brief electric stimuli (4–5 s) induced little slow-phase velocity and had no effect on the initial jump in eye velocity at the onset or the end of angular rotation. Effects of stimulation were unilateral, suggesting specificity of the output pathways. Activation of more caudal sites in the uvula produced nystagmus with a rapid rise in eye velocity, but the effects did not outlast the stimulus and did not affect VOR or OKAN time constants. Thus, stimulation of caudal parts of the uvula did not affect eye velocity produced by velocity storage. We postulate that the nodulus and sublobule d of the uvula control the time constant of the yaw axis (horizontal) component of slow-phase eye velocity produced by velocity storage.     

Effects of midline medullary lesions on velocity storage and the vestibulo-ocular reflex

Summary 1. Crossing fibers were sectioned at the midline of the medulla caudal to the abducens nucleus in four cynomolgus monkeys. In two animals the lesions caused the time constant of horizontal and vertical per- and post-rotatory nystagmus to fall to 5–8 s. The slow rise in optokinetic nystagmus (OKN), as well as optokinetic after-nystagmus (OKAN) and cross-coupling of horizontal to vertical OKN and OKAN were abolished. Steady state velocities could not be maintained during off-vertical axis rotation (OVAR). Pitch and yaw nystagmus were affected similarly. We conclude that the ability to store activity related to slow phase eye velocity, i.e., velocity storage, was lost in these monkeys for nystagmus about any axis. Velocity storage was partially affected by a small midline lesion in the same region in a third animal. There was no effect of a more superficial midline section in a fourth monkey, and it served as a control. 2. The gain (eye velocity/head velocity) of the vestibuloocular reflex (VOR) was unaffected by the midline lesions. Saccades were normal, as was the ability to hold the eyes in eccentric gaze positions. The gain of the fast component of OKN increased in one monkey to compensate for the loss of the slow component. 3. One animal was tested for its ability to adapt the gain of the VOR due to visual-vestibular mismatch after lesion. Average changes in gain in response to wearing magnifying (2.2 x) and reducing (0.5 x) lenses, were + 35% and — 30%, respectively. This is within the range of normal monkeys. Thus, a midline lesion that abolished velocity storage did not alter that animal's ability to adapt the gain of the VOR. 4. Lesions that reduced or abolished velocity storage interrupted crossing fibers in the rostral medulla, caudal to the abducens nuclei. Cells that contributed axons to this portion of the crossing fibers are most likely located in central portions of the medial vestibular nucleus (MVN) and/or in rostral portion of the descending vestibular nucleus (DVN). The implication is that velocity storage arises from neurons in MVN and DVN whose axons cross the midline.Supported by NS-00294, SFB 220-D8 and Core Center Grant EY-01867     

Vestibular and optokinetic eye movements evoked in the cat by rotation about a tilted axis

Summary Horizontal and vertical eye movements were recorded from cats in response to either a) off-vertical axis rotation (OVAR) at a range of velocities (5–72 deg/s) and a range of tilts (0–60 deg) or b) horizontal (with respect to the cat) optokinetic stimulation (10–80 deg/s), also around a range of tilted axes (0–60 deg). The responses to stopping either of these stimuli were also measured: post-rotatory nystagmus (PRN) following actual rotation, and optokinetic after nystagmus (OKAN) following optokinetic stimulation. The response found during OVAR was a nystagmus with a bias slow-phase velocity that was sinusoidally modulated. The bias was dependent on the tilt and reached 50% of its maximum velocity (maximum was 73±23% of the table velocity) at a tilt of 16 deg. The phase of modulation in horizontal eye velocity bore no consistent relation to the angular rotation. The amplitude of this modulation was roughly correlated with the bias with a slope of 0.13 (deg/s) modulation/(deg/s) bias velocity. There was also a low-velocity vertical bias with the slow-phases upwardly directed. The vertical bias was also modulated and the amplitude depended on the bias velocity (0.27 (deg/s) modulation/ (deg/s) bias velocity). When separated from the canal dependent response, the build up of the OVAR response had a time constant of 5.0±0.8 s. Following OVAR there was no decline in the time constant of PRN which remained at the value measured during earth-vertical axis rotation (EVAR) (6.3±2 s). The peak amplitude of PRN was reduced, dependent on the tilt, reaching only 20% of its EVAR value for a tilt of 20 deg. When a measurable PRN was found, it was accompanied by a slowly-emerging vertical component (time constant 5.4±2s) the effect of which was to vector the PRN accurately onto the earth horizontal. OKN measured about a tilted axis showed no differences in magnitude or direction from EVAR OKN even for tilts as large as 60 deg. OKAN following optokinetic stimulation around a tilted axis appeared normal in the horizontal plane (with respect to the animal) but was accompanied by a slowly emerging (time constant 4.1±2 s) vertical component, the effect of which was to vector the overall OKAN response onto the earth horizontal for tilts less than 20 deg. These results are compared with data from monkey and man and discussed in terms of the involvement of the velocity storage mechanism.     

Up-down asymmetry in human vertical optokinetic nystagmus and afternystagmus: contributions of the central and peripheral retinae

Summary The vertical optokinetic nystagmus (OKN) of 10 normal subjects and the optokinetic afternystagmus (OKAN) of 3 subjects were measured with the magnetic search coil technique. In order to assess the relative contributions of various retinal areas to the up-down asymmetry in OKN the central and peripheral visual fields were selectively stimulated in four OKN conditions. In the full-field OKN condition the stimulus was a 61°×64° display of moving random-dots. Overall, full-field OKN gains elicited by upward motion were significantly higher than those elicited by downward motion at stimulus velocities between 30 and 70°/s. In the periphery-only OKN condition a 3° or 6°-wide vertical band occluded the center of the full-field display. Nine of the 10 subjects displayed OKN in this condition. For 6 subjects, the addition of the 6° band to the full field resulted in an increase in the up-down asymmetry at stimulus velocities above 30°/s. For the other three subjects there was a decline in the gains of both upward and downward OKN when the 3° or 6° band was present; the result was directionally symmetric OKN gains. In the central-strip OKN condition only a 6°-wide central vertical strip of moving dots was visible. The gains of central-strip OKN were not significantly different from the full-field responses. A servo controlled centrally-located 10°× 6° moving display was used in the center-only OKN condition. In this condition both upward and downward gains were attenuated and there was no up-down asymmetry. OKAN was measured following a 50-s exposure to either the full-field or center-only OKN display. The stimulus velocity was 30°/s. After viewing the full-field display the 3 subjects displayed OKAN with slow phases upward following upward OKN but there was no downward OKAN following downward OKN. In contrast, there was no consistent directional asymmetry following exposure to the center-only display. The disappearance of the upward preponderance in OKN and OKAN with occlusion of the peripheral retina suggests that the directional asymmetry in vertical OKN exists in the slow OKN system.     

Effects of prolonged weightlessness on horizontal and vertical optokinetic nystagmus and optokinetic after-nystagmus in humans

Horizontal and vertical optokinetic nystagmus (OKN) and optokinetic after-nystagmus (OKAN) provided by a partial-field, binocular optokinetic stimulator were recorded in one astronaut before, during, and after a 25-day space flight. A ground-based study was performed on six control subjects. During the flight experiment, performed on flight days 5, 18, 19, and 21, the subject either had their feet attached to the deck or was free-floating. Vertical OKN gain only slightly increased in weightlessness compared with ground data, but the center of interest (CI) during vertical OKN, evaluated by the eye position in the saggital plane at the end of the fast phases relative to the straight-ahead direction, was found to be significantly changed during long-term exposure to weightlessness. The horizontal CI showed very little change in-flight, but the gain was increased. The time constant for the astronaut was small for vertical OKAN, but there was an increase in slow-phase velocity (SPV) by the end of the flight, which returned to normal postflight. These results partly confirm the data obtained during head-tilt studies on the ground and are in accordance with the hypothesis of a gravity-dependent control of vertical gaze direction during orientation reflexes.     

Vergence-dependent adaptation of the vestibulo-ocular reflex

The gain of the vestibulo-ocular reflex (VOR) normally depends on the distance between the subject and the visual target, but it remains uncertain whether vergence angle can be linked to changes in VOR gain through a process of context-dependent adaptation. In this study, we examined this question with an adaptation paradigm that modified the normal relationship between vergence angle and retinal image motion. Subjects were rotated sinusoidally while they viewed an optokinetic (OKN) stimulus through either diverging or converging prisms. In three subjects the diverging prisms were worn while the OKN stimulus moved out of phase with the head, and the converging prisms were worn when the OKN stimulus moved in-phase with the head. The relationship between the vergence angle and OKN stimulus was reversed in the fourth subject. After 2 h of training, the VOR gain at the two vergence angles changed significantly in all of the subjects, evidenced by the two different VOR gains that could be immediately accessed by switching between the diverged and converged conditions. The results demonstrate that subjects can learn to use vergence angle as the contextual cue that retrieves adaptive changes in the angular VOR.     

Vertical and torsional optokinetic eye movements in the rabbit

Summary Rabbits were placed inside a striped drum, which was rotated at selected constant speeds around the animal's sagittal or bitemporal axis. Eye position was recorded by means of the scleral search coil system. A regular vertical or rotatory optokinetic nystagmus (OKN) was constantly obtained. The ratioslow phase eye velocity/drum velocity (=gain) amounted to 0.7–0.9 for stimulus velocities up to 1°/sec, and declined progressively for higher stimulus velocities. The overall input-output relations for torsional and vertical OKN were very similar to those found previously for horizontal OKN. Upward and downward motion were equally effective as a stimulus for each eye apart. The same was true for nasal and temporal rotation.In darkness, rotatory and vertical drift of the eye was seen, as described before for the horizontal plane. These findings support the hypothesis that the OKN system stabilizes the eyes on the (non-rotating) visual surroundings.It is proposed that vertical, torsional as well as horizontal OKN are mediated by sub-sets of similar retinal direction-selective cells as described in the literature.