- Email: [email protected]
Adaptability of the vestibulo-ocular reflex to vision reversal in strobe reared cats
Brain Research, 209 (1981) 35-45 © Elsevier/North-HollandBiomedicalPress
35
ADAPTABILITY OF THE VESTIBULO-OCULAR REFLEX TO VISION REVERSAL IN STROBE REARED CATS
G. MANDL,G. MELVILLJONES and M. CYNADER Department of Physiology, McGill University, Montreal and (M.C.) Department of Psychology, Dalhousie University, Halifax (Canada)
(Accepted August 28th, 1980) Key words: vestibulo-ocularreflex-- visionreversal-- strobe rearing
SUMMARY Optical reversal of vision brings about adaptive changes in the vestibulo-ocular reflex (VOR) tending to reduce retinal image slip during head movement. The present experiments investigated this form of adaptation in cats whose complement of direction sensitive central visual cells had been substantially reduced by rearing in 8 Hz stroboscopic light. Horizontal vision reversal was produced by dove prisms carried in a skull-mounted mask. A scleral eye coil was used to measure horizontal eye movements. VOR gain and phase were measured in the dark during sinusoidal rotation using test stimuli of 1/8 Hz and 5- or 20°/sec velocity amplitude. Initially, strobe reared cats produced virtually normal VOR in the dark, except for slight but significant exaggeration of the normal phase advancement to be expected at 1/8 Hz. Addition of their familiar strobe illumination produced almost perfect oculomotor compensation. Maintained vision reversal in both strobe and normal illumination produced similar patterns of adaptive change in normal and strobe reared subjects, i.e. all animals exhibited an initial fast, and subsequent much slower, stage of gain attenuation, with similar changes in phase. Thus, strobe rearing did not prevent the development of an essentially normal VOR, nor did it interfere significantly with the ability to adapt in response to vision reversal. Since strobe rearing depletes direction selective visual movement detectors in the cortex and superior colliculi, it is inferred that signals responsible for activating the adaptive process are probably carried mainly in the accessory optic, rather than cortical and collicular, visual system.
INTRODUCTION Previous experiments with humans 7-9, catsl6, 20 and monkeys19 have shown that optical reversal of vision results in marked adaptive changes in the adult vestibulo-
36 ocular reflex (VOR). Since such changes always occurred in a manner tending to reduce retinal image slip during head rotation, it was suggested s that smooth retinal image slip might provide the error signal that activates the adaptive process. Human subjects were therefore exposed to vision reversal in strobe light (4 Hz) of sufficiently short flash duration (5/tsec) to prevent image slip during the flash17. The results demonstrated that even under those conditions goal-directed adaptive changes in the VOR continued to take place. This finding could imply that directionally selective visual cells in the CNS can extract the vector of retinal image displacement from a series of discontinuous (flashed) pattern presentations. One might ask, therefore, whether VOR adaptation would also occur in animals whose complement of directionally selective movement sensitive visual cells has been significantly reduced by having been raised from birth in stroboscopic illuminationÀ, 5. The present experiments investigated this question by comparing the adaptive effects of vision reversal, upon VOR gain and phase, in strobe reared and normal cats. The results indicate that VOR adaptation to reversed vision occurred, in very similar ways, in both normal and strobe reared animals examined in both stroboscopic and normal illumination. METHODS Two male cats ($1 and S~) were reared from birth in 8 Hz stroboscopic illumination of 10/zsec flash duration. After an 18 months period of such visual deprivation, both animals, together with one normal adult control animal (C), were chronically prepared as described in a companion article is, with a head fixation implant and a scleral eye coil in the right eye for measurement of eye movement. The eye movement recording system was calibrated by rotating the field coils relative to the stationary eye. In addition to the head fixation device, the acrylic block carried 3 stereotaxically mounted bolts for attachment of a precision-fitted plastic mask containing light weight dove prisms. The prisms were so arranged as to produce horizontal reversal of vision as previously described 16, providing a binocular field of view of about 50° solid angle. Care was taken to exclude any possibility of direct peripheral vision of the outside world by ensuring a close fit of the plastic mask to the face. The internal surface of the optical mask was coated with 'lamp black' to minimize internal reflections. While S1 was tested specifically for adaptive changes in the VOR due to vision reversal in strobe light, $2 was tested for similar changes during exposure to vision reversal in normal light. Cat C, serving as control to both S1 and $2, was subjected first to a period of adaptation in strobe light lasting 4 days (control for $1), and subsequently, after ten days of recovery with normal vision in normal light, to a period of adaptation in normal light (control for S~). During most of the adaptive regime, the animals were left running free in the laboratory environment with the reversing prisms in place. To expedite the adaptive process animals were routinely subjected to 2-4 h periods of rotational oscillation ('forcing') at 1/8 Hz and 5°/sec velocity amplitude, with head fixed to the turntable. The forcing was done in strobe light with cat $1, and with cat C during the first of C's two adaptive periods; and in normal light with $2 and with C during C's second adaptation period.
37
HEAD
],
,
Fig. 1. An example of compensatory eye movements during sinusoidal head and body oscillation in 8 Hz strobe illumination, performed by strobe reared cat (SD before adaptation to vision reversal. The animal was oscillated sinusoidally at 1/8 Hz, 7 ° amplitude. Despite strobe rearing the VOR gain and phase in this animal were within normal range. However, note the presence of abnormal oscillatory eye movements typical of strobe reared cats TM. Calibration bars 10°, 1 sec.
Adaptive changes in the VOR were estimated by measuring VOR gain (eye angular velocity/head velocity) and phase (relative to normal compensation) during rotational oscillation at 1/8 Hz and 5- or 20°/sec velocity-amplitude with head fixed to the turntable. These conditions were chosen since they have been shown by themselves not to modify the VOR and to lie well within the linear range of the normal VOR a. A computer algorithm 8 was used to extract the slow compensatory components of eye movements, and to synthesize curves of cumulative eye position (C.E.P.) 15. These curves were then fitted by calculated 'least squares' regression sinewaves which were in turn used for the estimation of VOR gain and phase. RESULTS
Fig. 1 shows that a strobe reared cat can generate a normal pattern of VOR in response to rotational stimulation of the vestibular system in the dark. Note, however, the superimposed abnormal oscillatory eye movements of small amplitude, characteristic of strobe reared cats, and described in detail in a companion articlO a. Owing to the relatively high frequency of these oscillations, their presence was not considered to have significantly interfered with measurements of VOR gain and phase at the relatively low test frequency of 1/8 Hz. V O R gain and phase before adaptation to reversed vision T a b l e I s u m m a r i z e s m e a n c o n t r o l v a l u e s o f g a i n a n d phase, for s t r o b e r e a r e d a n d n o r m a l a n i m a l s . It c a n be seen t h a t the gains in the d a r k , for b o t h $1 a n d C, fall well TABLE I
Mean values, and standard errors (in brackets), of VOR gain (eye velocity~head velocity) and phase (re perfect compensation) before adaptation to vision reversal All means derived from 20 rotational cycles, l0 on each of two consecutive days. $I and $2 were reared in 8 Hz strobe light; C was a normally reared control animal.
Subject
S1 $2 C
Dark
Strobe
Gain
Phase-deg.
Gain
Phase-deg.
0.82 (0.03) 0.63 (0.02) 0.95 (0.02)
7.9 (2.0) 9.7 (2.2) 3.7 (0.8)
0.97 (0.03) 0.83 (0.02)
0.3 (1.6) --0.7 (1.0)
38 within the range of 1.16--0.74 (mean 0.89, S.E. 0.14) determined for a group of 11 normal cats by Robinson 20. The gain for $2 lay somewhat outside this range, but was still within its 95 ~ confidence limit. This was presumably due to individual variability. Exposure to strobe light during oscillation of the animals, modified the oculomotor gain of both strobe reared cats significantly (P < 0.01), effectively raising the oculomotor response of S~ to the goal of perfect tracking with a gain of I. The response of $2, while also raised toward this functional goal, remained a small, but significant, amount below the level for perfect compensation. With regard to phase, the slight advancement with respect to perfect compensation, shown by control cat C in the dark, is quite normal for these test conditions3, 20. However, the dark value for $2 is significantly greater than this control value (P < 0.02). Taken together with the increased value for $1 (0.05 < P < 0.1) it seems likely there was a real tendency for an abnormal degree of phase advancement in the responses of the strobe reared animals. Note also that the variability in eye-head phase relation is somewhat larger for both strobe reared animals than for C (see also Figs. 3 and 4). Exposure to strobe light during turntable oscillation much improved the mean phase in both strobe reared animals, which then achieved values indistinguishable from the ideal zere phase. This functional improvement was accompanied by a reduced variability (Table I).
Effects of vision reversal upon VOR gain Fig. 2A shows a comparison of the time course of VOR gain attenuation during maintained vision reversal in strobe light for $1 (filled symbols) and C (open symbols). The salient features for both subjects are: (1) the initial rapid reduction of gain, within the first few hours, to about 60-70 ~ of control value; and (2) the subsequent much slower decline at an approximate rate of some 5-10~/day. Since over the first week or so the process of long-term gain reduction tends to follow a quasi-exponential time course, the data have been plotted on semi-log coordinates, and regression lines fitted to the data points obtained after the first 5 h of forcing. The line equations with correlation coefficients are given in Table II. The yintercept values of the two regression lines for $1 and C in strobe light are statistically indistinguishable (P > 0.2), indicating that the ultimate magnitude of the rapid adaptive component of attenuation was similar in the strobe reared and control animals. Table III shows time constants for the fast (rh) and slow (z~) adaptive effects, zh Was estimated from the first gain measurements on the first day of vision reversal, after 4-5 h forced oscillations in strobe light. Actual estimates of zh in the table were calculated according to the relation zh ---- t/(lnl--ln2), where In1 is the natural log of 100, and ln~.is the natural log of the normalized gain measured t h after commencing forcing with the prisms on. This latter value was corrected for the small effect of the slow process represented by the plotted straight regression lines in Fig. 2. The data indicate a somewhat more rapid fast effect in C than in $1, although the short time constants of both animals are of the same order of magnitude, and much shorter than the values for ~ra.
39
loo~
A
801~ • 6o]g-_..
=
i\" 40J~
r<
(.9
"-.. ,,
"'-..I
~: 2O O z
STROBE
o
~
~,
DAYS
100
B
8o, 6O z < (9 4 0 a u.I N .J <
0~.~.~.~
0
~ 2o
o
LIGHT
z
0
o 0
o
~
~,
~
DAYS
~
1'o
Fig. 2. Time course of VOR adaptation to vision reversal. All data points indicate VOR gain (eye velocity/head velocity; ordinate) measured in the dark, and normalized relative to mean control values ([3). Linear regression lines (equations in Table II) have been calculated for control animal C (solid lines, open symbols), and for the two strobe reared animals $1 and $2 (interrupted lines, filled symbols). A: comparison of adaptation in strobe light, of strobe reared $1 with normal C. B: comparison of adaptation in normal light, of strobe reared Sz with normal C. TABLE I1
Equations and correlation coefficients for regression lines in Fig. 2. Correlation coefficients are significant at P < 0.01, except for $1 in strobe light which is significant at P < 0.05.
Cat
Adapting condition
C
Strobe
S1 C $2
No. of data points 5 9
Light
11 5
Regression line y (In norm. gain) x (days)
Correlation coefficient
y = 3.97-0.27 x (Fig. 2A, solid line) y = 4.15-0.13 X (Fig. 2A, dashed line) y = 4.16-0.12 x (Fig. 2B, solid line) y = 4.31-0.14 x (Fig. 2B, dashed line)
0.97 0.69 0.83 0.99
40 TABLE III Time constants for fast (rn) and slow (ra) components of VOR adaptation to vision reversal in both stroboscopic and normal illumination.
$1, $2 and C are experimental animals used in the present study; L represents a normally reared cat from previous experiments~6. Corresponding average data for 3 human subjects are also given9. Subject
Cat C $1 C S2 L Man (average of 3)
Time constant
Illumination
Short (rn, hours)
Long (ra, days)
8.9 14.6 I 1.7 13.3 9.8 7.8
3.8 7.8 8.1 7.3 8.6 l 1.5
Strobe Normal
The two regression lines in Fig. 2A have significant correlation coefficients (0.69, P < 0.05 for $1 ; 0.97, P < 0.01 for C; see Table II) and hence demonstrate the presence of additional long-term adaptive changes having time constants (za, Table III) at least an order of magnitude greater than those of the initial short-term adaptation described above. Although the long-term decline in VOR gain appears faster in C than in $1, the slopes of the two regression lines are statistically indistinguishable (P :> 0.05), suggesting a similar long-term adaptive process operating in both the strobe reared and the normal cats. Fig. 2B shows a comparison of the time course of gain attenuation during maintained vision reversal for $2 (filled symbols) and C in normal light. Again, gain changes in the strobe reared ($2) and the control (C) animals follow similar time courses: an initial rapid fall (time constant zh 13.3 and 11.7 h) to some 70 ~ of initial value was followed by a much more gradual decline (time constant ~ra 7.3 and 8.1 days, Table III). The two regression lines have highly significant correlation coefficients (Table II; 0.83 and 0.99, P < 0.01) and statistically indistinguishable slopes (P > 0.5) and yintercepts (P > 0.2), indicating that, just as in strobe illumination, the initial fast, and subsequent slow, adaptive processes operatiDg in normal light in the two animals are indistinguishable. These results are in line with those obtained from previous animal experiments in normal light (Cat L, Table Ili; data reanalyzed from Melvill Jones and Davies16). Within this context, it is particularly noteworthy that data from previous human experiments reveal a similar dichotomy of adaptive time courses with similar time constants (Table III; data reanalyzed from Gonshor and Melvill Jonesg). Combining all data points obtained in strobe light (Fig. 2A, $1 and C), and comparing these with similarly lumped data obtained in normal light (Fig. 2B, $2 and C), showed no significant differences in either the y-intercepts (P :> 0.4), or the slopes (P > 0.5), of the two resulting regression lines. Consequently, VOR adaptation to vision reversal proceeded in similar fashion in both strobe and normal light conditions. Effects o f vision reversal upon V O R phase
Since VOR adaptation to vision reversal may involve simultaneous changes in
41 phase as well as gain, it is convenient to display these two parameters in the form of polar plots. Figs. 3 and 4 show several examples of such plots where the length and angle of each vector represents the gain and phase, respectively, obtained from a single rotational cycle in a test series of 10 such cycles. The first notable feature seen in both figures is that, before adaptation to vision reversal, both strobe reared cats exhibited a larger variability in the dark tested phase as compared to the control animal (see also Table I). The effects of the early (rapid) adaptation led to an exaggeration of this particular phenomenon. At the end of each experiment, that is after the slow adaptive process had taken effect, an important difference emerged between the strobe reared and control animals, whereby significant phase advance of the dark-tested response occurred in the control animal, but not in the strobe reared ones. Interestingly, this latter difference did not apparently carry over into the adapted oculomotor response observed in the presence of reversed vision (strobe and normal light). Indeed, from the data shown, it would seem that in these
S1
C
BEFORE ADAPTATION (DARK)
ADAPTATION
-1
÷1
P ~ ~ ' ~
(DARK)
-1
J
-1
'°-'
LATE ADAPTATION
+1
-1
__
l +1
-1
+1
~ ,
+1
....
-1
t
.......
,
i
t
(8 HZ STROBE
WIT. PRISMS):,
.
.
.
.
.
.
.
.
.
.
.
.
.
Fig. 3. Adaptation to vision reversal in strobe light: polar plots of VOR gain and phase obtained from $1 and C before adaptation to reversed vision, and in the early (hours) and late (days) stages of adaptation in strobe light. The gain and phase for each cycle of rotational oscillation (1/8 Hz, 5°/sec) is represented, respectively, by the length and the angle of a given vector. There are 10 vectors in each diagram (this is not always evident due to occasional superposition of individual lines). Perfect oculomotor compensation without vision reversal is represented by a vector of + 1 magnitude and zero phase relative to the + 1 coordinate. Phase advancement is represented by anti-clockwise displacement of the vector, with perfect reversed compensation denoted by the --1 point. Note that the bottom plots were obtained in 8 Hz strobe lighting, with reversing prisms in place. All other plots represent results obtained in the dark.
42
C
S2 BEFORE ADAPTATION (DARK)
EARLY ADAPTATION (DARK)
(DARK)
/-'
i
-1
t
+1
! -1
I
-1
'
'
'
'
+I
-1
I
+I
L
+1
-1
+I
+1
-1
+1
a ~"',
,
t
,
I
LATE ADAPTATION
(LIGHT~ WITH
PRISMS)
_II
Fig. 4. Adaptation to vision reversal in normal light : polar plots of VOR gain and phase, obtained from S~ and C before and during adaptation to reversed vision in normal light. Details as in Fig. 3, except for the bottom plots which were obtained in normal light with prisms in place. conditions both strobe reared cats performed similarly to the parallel control animal (Figs. 3 and 4). Thus, strobe rearing had apparently not prejudiced the ability of the adapted animals to combine both visual and vestibular drives toward the ideal goal of reversed tracking. Nevertheless, at this relatively early stage of the adaptive process, the realization of the ideal condition for reversed tracking was far from achieved, as indicated by the very wide dispersion of phase and low gain seen in both strobe reared and control animals. This contrasts with previous long-term experiments on cats, where virtually perfect reversed tracking was eventually found to occur under similar stimulus conditions in normal light 8. It is worth noting that, in a series of parallel experiments with normal and strobe reared cats, concerned with the influence of VOR adaptation on optokinetic tracking, marked changes in oculomotor responses to monocular nasal and temporal optokinetic stimulation were observed in the adapted animals 14. Consequently, the adaptive changes observed here were presumably not confined to the vestibular system. DISCUSSION The original hypothesis, regarding the dependence o f V O R adaptatioP on retinal image slip, suggested that movement sensitive visual cells with directional preference
43 were important elements in the mediation of the adaptive process. This, in turn, implied that animals lacking a full complement of directionally selective visual cells should be deficient in VOR adaptation. Thus, cats reared in total darkness and lacking direction selective neurons in both the visual cortex and the superior colliculus2, 22, exhibit a lowered VOR gain of about 0.3~).5 (as measured in the dark). These animals are also deficient in their ability to adaptively modify this gain, even under conditions which are known to promote such adaptation in normal animals10. The present experiments were an attempt to examine the role of direction selective visual cells after selective deprivation of exposure to moving stimuli. The most important conclusion emerging from the present results is that both strobe reared and normal cats showed very similar patterns of adaptation of their VOR to optical reversal of vision, in that both types of animals exhibited an initial fast, and a subsequent much slower, stage of gain attenuation. Furthermore, each of the two stages proceeded at similor rates in strobe reared and control animals, and independent of whether adaptation occurred in strobe or normal light. Since the directional component of the implied retinal error signal would be of paramount importance in the adaptive process, the fact that the deprived animals performed indistinguishably from the normal controls provides strong evidence against the involvement of cortical and collicular motion sensitive cells in normal VOR adaptation. This in turn, suggests a predominant utilization of the accessory optic system in the adaptive process, and by implication, a sparing of directional sensitivity of this system in strobe reared cats. These latter conclusions are in line, on the one hand, with the observations of Ito et al. 13 in rabbits, that ablation of cerebral cortex has no effect upon VOR adaptation; and on the other hand, with the suggestion that projections from the accessory optic tract (AOT), via olivary climbing fiber input to vestibular cerebellum, may be responsible for the modulation of the synaptic efficacy responsible for VOR adaptation 6,12. Interestingly, direction selectivity of cells in the nucleus of the optic tract remains unchanged after lesions of the visual cortex11. The present results also demonstrate another important aspect of VOR gain attenuation, common to both normal and strobe reared cats, and to humans: reference to Table III shows that adaptation to vision reversal proceeds, in both animals and humans, as a two time constant process, with the initial fast period measured in hours, and the subsequent slow period in days. These two time courses are quite distinct from one another (P ,~ 0.001), and therefore presumably represent two quite different processes - an observation which confirms an earlier suggestion based on an experimentally derived model of VOR adaptation in humans 4, to the effect that at least two processes, with different time constants, act simultaneously to produce the overall adaptive phenomenon. The similar values for cats and humans shown in Table III strongly suggests that similar fast and slow adaptive neural mechanisms are at play in primates and carnivores. Despite the similarities in adaptive VOR gain attenuation discussed above, reference to Figs. 3 and 4 and Table I indicates that the dark tested VOR before adaptation differed in two significant ways between strobe reared and control animals. First, the phase advancement in strobe reared animals was larger than would normally
44 be expected under the test conditions. This could imply a decrease due to strobe rearing in the overall VOR time constant which is now thought to depend on interaction between peripheral and central dynamics 21. Since it seems unlikely that the peripheral component has undergone any change, and following Robinson's argument 21, the observed phase advancement would seem to implicate central components, specifically a reduced positive feedback gain in the 'optokinetic loop'. In view of this prediction it is intriguing to note that the optokinetic gain tested with prolonged (1 min) unidirectional stimulation in the strobe reared animals appeared significantly reduced, as reported elsewhere 14. Second, in strobe reared animals there was increased variability of phase within a given test series of 10 cycles of sinusoidal oscillation. It seems unlikely that this could be attributed to the presence of spontaneous eye oscillations per se, both on account of the small amplitude and frequency of these eye oscillations (Fig. 1), and because there was no significant difference (P ,~ 0.001) between the correlation coefficients of sinewaves fitted to the curves of cumulative eye position (CEP. see Methods) of strobe reared and control cats. It is conceivable that neurological 'jitter', originating from the strobe rearing environment, might contribute to the enlarged phase dispersion. It is also interesting to note that in the late stage of adaption, the dark tested VOR appears to have been, on the average, less phase shifted in the strobe reared cats than in the control one (Figs. 3 and 4). Nevertheless, in both normal and strobe illumination, the VOR adapted strobe reared cats did as well as the control animal in the attempt to produce reversed ocular compensation (Figs. 3 and 4). Lastly, it is interesting to note that, whereas the adaptive response in the present experiments was as great in strobe light as in normal light (for both strobe reared and normal cats), in the previous human experiments 17 the rate of adaption in strobe light was only about half that in normal light. This apparent discrepancy might be attributable to either a species differences; and/or to a difference in strobe flash frequency, which was 8 Hz in the present experiments, but only 4 Hz in the previous human study. ACKNOWLEDGEMENTS The present work was supported by the Canadian MRC, the Canadian NSERC, and USPHS Grant EY02248 to M.C. The scleral eye coil implantations were done by Dr. D. Guitton. The invaluable technical assistance of Mr. T. O'Farrel and Mr. W. Ferch is gratefully acknowledged. REFERENCES 1 Cynader, M. and Chernenko, G., Abolition of direction selectivityin the visual cortex of the cat, Science, 193 (1976) 504-505. 2 Cynader, M., Berman, N. and Hein, A., Recovery of function in cat visual cortex following prolonged deprivation, Exp. Brain Res., 25 (1976) 139-156. 3 Davies, P. R. T., Neural Adaptation in Humans and Cats Sube]cted to Long-term Optical Reversal of Vision: an Experimental and Analytical Study of Plasticity, Ph.D. Thesis, McGill University, Montreal (1979). 4 Davies, P. R. T. and Melvill Jones, G., An adaptive neural model compatible with plastic changes induced in the human vestibulo-ocularreflexby prolonged optical reversal of vision, Brain Research, 103 (1976) 546--550.
45 5 Flandrin, J. M., Kennedy, H. and Amblard, B., Effects of stroboscopic rearing on the binocularity of cat superior colliculus neurons, Brain Research, 101 (1976) 576-581. 6 Ghelarducci, B., Ito, M. and Yagi, N., Impulse discharges from flocculus Purkinje cells of alert rabbits during visual stimulation combined with horizontal head rotation, Brain Research, 87 (1975) 66-72. 7 Gonshor, A. and Melvill Jones, G., Plasticity in the adult human vestibulo-ocular reflex arc, Proc. Can. Fed. Biol. Soc., 14 (1971) 11. 8 Gonshor, A. and Melvill Jones, G., Short-term adaptive changes in the human vestibulo-ocular reflex arc, J. Physiol. (Loncl.), 256 (1976) 361-379. 9 Gonshor, A. and Melvill Jones, G., Extreme vestibulo-ocular adaptation induced by prolonged optical reversal of vision, J. Physiol. (Lond.), 256 (1976) 381-414. 10 Harris, L. R. and Cynader, M. S., Attempts to modify the vestibulo-ocular reflex of normal and dark-reared cats, Neurosci. Abstr., 5 (1979) abstract 2646, p. 787. 11 Hoffmann, K.-P., Behrend, K. and Schoppmann, A., Visual responses of neurons in the nucleus of the optic tract of visually deprived cats, NeuroscL Abstr., 3 (1977) abstract 1790, p. 563. 12 Ito, M., Neural design of the cerebellar motor control system, Brain Research, 40 (1972) 81-84. 13 Ito, M., Shiida, T., Yagi, N. and Yamamoto, M., Visual influence on rabbit's horizontal vestibuloocular reflex presumably effected via the cerebellar flocculus, Brain Research, 65 (1974) 170-174. 14 Mandl, G., Melvill Jones, G. and Cynader, M., Modification of OKN by maintained vision reversal in normal and strobe reared cats, Neurosci. Abstr., 1980, abstract 158.1, p. 473. 15 Meiry, J. L., The Vestibular System and Human Dynamic Space Orientation, NASA CR-628, Washington, D.C., 1966. 16 Melvill Jones, G. and Davies, P. R. T., Adaptation of cat vestibulo-ocular reflex to 200 days of optically reversed vision, Brain Research, 103 (1976) 551-554. 17 Melvill Jones, G. and Mandl, G., Effects of strobe light on adaptation of vestibulo-ocular reflex (VOR) to vision reversal, Brain Research, 164 (1979) 300-303. 18 Melvill Jones, G., Mandl, G., Cynader, M. and Outerbridge, J. S., Eye oscillations in strobe reared cats, Brain Research, 209 (1981). 19 Miles, F. A. and Eighmy, B. B., Long-term adaptive changes in primtae vestibuloocular reflex. I. Behavioral observations, J. NeurophysioL, 43 (1980) 1406-1425. 20 Robinson, D. A., Adaptive gain control of the vestibuloocular reflex by the cerebellum, J. Neurophysiol., 39 (1976) 954-969. 21 Robinson, D. A., Vestibular and optokinetic symbiosis: an example of explaining by modelling. In R. Baker and A. Berthoz (Eds.), Control of Gaze by Brain Stem Neurons, Elsevier, Amsterdam, 1977, pp. 49-58. 22 Wickelgren, B. G. and Sterling, P., Influence of visual cortex on receptive fields in superior colliculus of the cat, J. Neurophysiol., 32 (1969) 16-23.
35
ADAPTABILITY OF THE VESTIBULO-OCULAR REFLEX TO VISION REVERSAL IN STROBE REARED CATS
G. MANDL,G. MELVILLJONES and M. CYNADER Department of Physiology, McGill University, Montreal and (M.C.) Department of Psychology, Dalhousie University, Halifax (Canada)
(Accepted August 28th, 1980) Key words: vestibulo-ocularreflex-- visionreversal-- strobe rearing
SUMMARY Optical reversal of vision brings about adaptive changes in the vestibulo-ocular reflex (VOR) tending to reduce retinal image slip during head movement. The present experiments investigated this form of adaptation in cats whose complement of direction sensitive central visual cells had been substantially reduced by rearing in 8 Hz stroboscopic light. Horizontal vision reversal was produced by dove prisms carried in a skull-mounted mask. A scleral eye coil was used to measure horizontal eye movements. VOR gain and phase were measured in the dark during sinusoidal rotation using test stimuli of 1/8 Hz and 5- or 20°/sec velocity amplitude. Initially, strobe reared cats produced virtually normal VOR in the dark, except for slight but significant exaggeration of the normal phase advancement to be expected at 1/8 Hz. Addition of their familiar strobe illumination produced almost perfect oculomotor compensation. Maintained vision reversal in both strobe and normal illumination produced similar patterns of adaptive change in normal and strobe reared subjects, i.e. all animals exhibited an initial fast, and subsequent much slower, stage of gain attenuation, with similar changes in phase. Thus, strobe rearing did not prevent the development of an essentially normal VOR, nor did it interfere significantly with the ability to adapt in response to vision reversal. Since strobe rearing depletes direction selective visual movement detectors in the cortex and superior colliculi, it is inferred that signals responsible for activating the adaptive process are probably carried mainly in the accessory optic, rather than cortical and collicular, visual system.
INTRODUCTION Previous experiments with humans 7-9, catsl6, 20 and monkeys19 have shown that optical reversal of vision results in marked adaptive changes in the adult vestibulo-
36 ocular reflex (VOR). Since such changes always occurred in a manner tending to reduce retinal image slip during head rotation, it was suggested s that smooth retinal image slip might provide the error signal that activates the adaptive process. Human subjects were therefore exposed to vision reversal in strobe light (4 Hz) of sufficiently short flash duration (5/tsec) to prevent image slip during the flash17. The results demonstrated that even under those conditions goal-directed adaptive changes in the VOR continued to take place. This finding could imply that directionally selective visual cells in the CNS can extract the vector of retinal image displacement from a series of discontinuous (flashed) pattern presentations. One might ask, therefore, whether VOR adaptation would also occur in animals whose complement of directionally selective movement sensitive visual cells has been significantly reduced by having been raised from birth in stroboscopic illuminationÀ, 5. The present experiments investigated this question by comparing the adaptive effects of vision reversal, upon VOR gain and phase, in strobe reared and normal cats. The results indicate that VOR adaptation to reversed vision occurred, in very similar ways, in both normal and strobe reared animals examined in both stroboscopic and normal illumination. METHODS Two male cats ($1 and S~) were reared from birth in 8 Hz stroboscopic illumination of 10/zsec flash duration. After an 18 months period of such visual deprivation, both animals, together with one normal adult control animal (C), were chronically prepared as described in a companion article is, with a head fixation implant and a scleral eye coil in the right eye for measurement of eye movement. The eye movement recording system was calibrated by rotating the field coils relative to the stationary eye. In addition to the head fixation device, the acrylic block carried 3 stereotaxically mounted bolts for attachment of a precision-fitted plastic mask containing light weight dove prisms. The prisms were so arranged as to produce horizontal reversal of vision as previously described 16, providing a binocular field of view of about 50° solid angle. Care was taken to exclude any possibility of direct peripheral vision of the outside world by ensuring a close fit of the plastic mask to the face. The internal surface of the optical mask was coated with 'lamp black' to minimize internal reflections. While S1 was tested specifically for adaptive changes in the VOR due to vision reversal in strobe light, $2 was tested for similar changes during exposure to vision reversal in normal light. Cat C, serving as control to both S1 and $2, was subjected first to a period of adaptation in strobe light lasting 4 days (control for $1), and subsequently, after ten days of recovery with normal vision in normal light, to a period of adaptation in normal light (control for S~). During most of the adaptive regime, the animals were left running free in the laboratory environment with the reversing prisms in place. To expedite the adaptive process animals were routinely subjected to 2-4 h periods of rotational oscillation ('forcing') at 1/8 Hz and 5°/sec velocity amplitude, with head fixed to the turntable. The forcing was done in strobe light with cat $1, and with cat C during the first of C's two adaptive periods; and in normal light with $2 and with C during C's second adaptation period.
37
HEAD
],
,
Fig. 1. An example of compensatory eye movements during sinusoidal head and body oscillation in 8 Hz strobe illumination, performed by strobe reared cat (SD before adaptation to vision reversal. The animal was oscillated sinusoidally at 1/8 Hz, 7 ° amplitude. Despite strobe rearing the VOR gain and phase in this animal were within normal range. However, note the presence of abnormal oscillatory eye movements typical of strobe reared cats TM. Calibration bars 10°, 1 sec.
Adaptive changes in the VOR were estimated by measuring VOR gain (eye angular velocity/head velocity) and phase (relative to normal compensation) during rotational oscillation at 1/8 Hz and 5- or 20°/sec velocity-amplitude with head fixed to the turntable. These conditions were chosen since they have been shown by themselves not to modify the VOR and to lie well within the linear range of the normal VOR a. A computer algorithm 8 was used to extract the slow compensatory components of eye movements, and to synthesize curves of cumulative eye position (C.E.P.) 15. These curves were then fitted by calculated 'least squares' regression sinewaves which were in turn used for the estimation of VOR gain and phase. RESULTS
Fig. 1 shows that a strobe reared cat can generate a normal pattern of VOR in response to rotational stimulation of the vestibular system in the dark. Note, however, the superimposed abnormal oscillatory eye movements of small amplitude, characteristic of strobe reared cats, and described in detail in a companion articlO a. Owing to the relatively high frequency of these oscillations, their presence was not considered to have significantly interfered with measurements of VOR gain and phase at the relatively low test frequency of 1/8 Hz. V O R gain and phase before adaptation to reversed vision T a b l e I s u m m a r i z e s m e a n c o n t r o l v a l u e s o f g a i n a n d phase, for s t r o b e r e a r e d a n d n o r m a l a n i m a l s . It c a n be seen t h a t the gains in the d a r k , for b o t h $1 a n d C, fall well TABLE I
Mean values, and standard errors (in brackets), of VOR gain (eye velocity~head velocity) and phase (re perfect compensation) before adaptation to vision reversal All means derived from 20 rotational cycles, l0 on each of two consecutive days. $I and $2 were reared in 8 Hz strobe light; C was a normally reared control animal.
Subject
S1 $2 C
Dark
Strobe
Gain
Phase-deg.
Gain
Phase-deg.
0.82 (0.03) 0.63 (0.02) 0.95 (0.02)
7.9 (2.0) 9.7 (2.2) 3.7 (0.8)
0.97 (0.03) 0.83 (0.02)
0.3 (1.6) --0.7 (1.0)
38 within the range of 1.16--0.74 (mean 0.89, S.E. 0.14) determined for a group of 11 normal cats by Robinson 20. The gain for $2 lay somewhat outside this range, but was still within its 95 ~ confidence limit. This was presumably due to individual variability. Exposure to strobe light during oscillation of the animals, modified the oculomotor gain of both strobe reared cats significantly (P < 0.01), effectively raising the oculomotor response of S~ to the goal of perfect tracking with a gain of I. The response of $2, while also raised toward this functional goal, remained a small, but significant, amount below the level for perfect compensation. With regard to phase, the slight advancement with respect to perfect compensation, shown by control cat C in the dark, is quite normal for these test conditions3, 20. However, the dark value for $2 is significantly greater than this control value (P < 0.02). Taken together with the increased value for $1 (0.05 < P < 0.1) it seems likely there was a real tendency for an abnormal degree of phase advancement in the responses of the strobe reared animals. Note also that the variability in eye-head phase relation is somewhat larger for both strobe reared animals than for C (see also Figs. 3 and 4). Exposure to strobe light during turntable oscillation much improved the mean phase in both strobe reared animals, which then achieved values indistinguishable from the ideal zere phase. This functional improvement was accompanied by a reduced variability (Table I).
Effects of vision reversal upon VOR gain Fig. 2A shows a comparison of the time course of VOR gain attenuation during maintained vision reversal in strobe light for $1 (filled symbols) and C (open symbols). The salient features for both subjects are: (1) the initial rapid reduction of gain, within the first few hours, to about 60-70 ~ of control value; and (2) the subsequent much slower decline at an approximate rate of some 5-10~/day. Since over the first week or so the process of long-term gain reduction tends to follow a quasi-exponential time course, the data have been plotted on semi-log coordinates, and regression lines fitted to the data points obtained after the first 5 h of forcing. The line equations with correlation coefficients are given in Table II. The yintercept values of the two regression lines for $1 and C in strobe light are statistically indistinguishable (P > 0.2), indicating that the ultimate magnitude of the rapid adaptive component of attenuation was similar in the strobe reared and control animals. Table III shows time constants for the fast (rh) and slow (z~) adaptive effects, zh Was estimated from the first gain measurements on the first day of vision reversal, after 4-5 h forced oscillations in strobe light. Actual estimates of zh in the table were calculated according to the relation zh ---- t/(lnl--ln2), where In1 is the natural log of 100, and ln~.is the natural log of the normalized gain measured t h after commencing forcing with the prisms on. This latter value was corrected for the small effect of the slow process represented by the plotted straight regression lines in Fig. 2. The data indicate a somewhat more rapid fast effect in C than in $1, although the short time constants of both animals are of the same order of magnitude, and much shorter than the values for ~ra.
39
loo~
A
801~ • 6o]g-_..
=
i\" 40J~
r<
(.9
"-.. ,,
"'-..I
~: 2O O z
STROBE
o
~
~,
DAYS
100
B
8o, 6O z < (9 4 0 a u.I N .J <
0~.~.~.~
0
~ 2o
o
LIGHT
z
0
o 0
o
~
~,
~
DAYS
~
1'o
Fig. 2. Time course of VOR adaptation to vision reversal. All data points indicate VOR gain (eye velocity/head velocity; ordinate) measured in the dark, and normalized relative to mean control values ([3). Linear regression lines (equations in Table II) have been calculated for control animal C (solid lines, open symbols), and for the two strobe reared animals $1 and $2 (interrupted lines, filled symbols). A: comparison of adaptation in strobe light, of strobe reared $1 with normal C. B: comparison of adaptation in normal light, of strobe reared Sz with normal C. TABLE I1
Equations and correlation coefficients for regression lines in Fig. 2. Correlation coefficients are significant at P < 0.01, except for $1 in strobe light which is significant at P < 0.05.
Cat
Adapting condition
C
Strobe
S1 C $2
No. of data points 5 9
Light
11 5
Regression line y (In norm. gain) x (days)
Correlation coefficient
y = 3.97-0.27 x (Fig. 2A, solid line) y = 4.15-0.13 X (Fig. 2A, dashed line) y = 4.16-0.12 x (Fig. 2B, solid line) y = 4.31-0.14 x (Fig. 2B, dashed line)
0.97 0.69 0.83 0.99
40 TABLE III Time constants for fast (rn) and slow (ra) components of VOR adaptation to vision reversal in both stroboscopic and normal illumination.
$1, $2 and C are experimental animals used in the present study; L represents a normally reared cat from previous experiments~6. Corresponding average data for 3 human subjects are also given9. Subject
Cat C $1 C S2 L Man (average of 3)
Time constant
Illumination
Short (rn, hours)
Long (ra, days)
8.9 14.6 I 1.7 13.3 9.8 7.8
3.8 7.8 8.1 7.3 8.6 l 1.5
Strobe Normal
The two regression lines in Fig. 2A have significant correlation coefficients (0.69, P < 0.05 for $1 ; 0.97, P < 0.01 for C; see Table II) and hence demonstrate the presence of additional long-term adaptive changes having time constants (za, Table III) at least an order of magnitude greater than those of the initial short-term adaptation described above. Although the long-term decline in VOR gain appears faster in C than in $1, the slopes of the two regression lines are statistically indistinguishable (P :> 0.05), suggesting a similar long-term adaptive process operating in both the strobe reared and the normal cats. Fig. 2B shows a comparison of the time course of gain attenuation during maintained vision reversal for $2 (filled symbols) and C in normal light. Again, gain changes in the strobe reared ($2) and the control (C) animals follow similar time courses: an initial rapid fall (time constant zh 13.3 and 11.7 h) to some 70 ~ of initial value was followed by a much more gradual decline (time constant ~ra 7.3 and 8.1 days, Table III). The two regression lines have highly significant correlation coefficients (Table II; 0.83 and 0.99, P < 0.01) and statistically indistinguishable slopes (P > 0.5) and yintercepts (P > 0.2), indicating that, just as in strobe illumination, the initial fast, and subsequent slow, adaptive processes operatiDg in normal light in the two animals are indistinguishable. These results are in line with those obtained from previous animal experiments in normal light (Cat L, Table Ili; data reanalyzed from Melvill Jones and Davies16). Within this context, it is particularly noteworthy that data from previous human experiments reveal a similar dichotomy of adaptive time courses with similar time constants (Table III; data reanalyzed from Gonshor and Melvill Jonesg). Combining all data points obtained in strobe light (Fig. 2A, $1 and C), and comparing these with similarly lumped data obtained in normal light (Fig. 2B, $2 and C), showed no significant differences in either the y-intercepts (P :> 0.4), or the slopes (P > 0.5), of the two resulting regression lines. Consequently, VOR adaptation to vision reversal proceeded in similar fashion in both strobe and normal light conditions. Effects o f vision reversal upon V O R phase
Since VOR adaptation to vision reversal may involve simultaneous changes in
41 phase as well as gain, it is convenient to display these two parameters in the form of polar plots. Figs. 3 and 4 show several examples of such plots where the length and angle of each vector represents the gain and phase, respectively, obtained from a single rotational cycle in a test series of 10 such cycles. The first notable feature seen in both figures is that, before adaptation to vision reversal, both strobe reared cats exhibited a larger variability in the dark tested phase as compared to the control animal (see also Table I). The effects of the early (rapid) adaptation led to an exaggeration of this particular phenomenon. At the end of each experiment, that is after the slow adaptive process had taken effect, an important difference emerged between the strobe reared and control animals, whereby significant phase advance of the dark-tested response occurred in the control animal, but not in the strobe reared ones. Interestingly, this latter difference did not apparently carry over into the adapted oculomotor response observed in the presence of reversed vision (strobe and normal light). Indeed, from the data shown, it would seem that in these
S1
C
BEFORE ADAPTATION (DARK)
ADAPTATION
-1
÷1
P ~ ~ ' ~
(DARK)
-1
J
-1
'°-'
LATE ADAPTATION
+1
-1
__
l +1
-1
+1
~ ,
+1
....
-1
t
.......
,
i
t
(8 HZ STROBE
WIT. PRISMS):,
.
.
.
.
.
.
.
.
.
.
.
.
.
Fig. 3. Adaptation to vision reversal in strobe light: polar plots of VOR gain and phase obtained from $1 and C before adaptation to reversed vision, and in the early (hours) and late (days) stages of adaptation in strobe light. The gain and phase for each cycle of rotational oscillation (1/8 Hz, 5°/sec) is represented, respectively, by the length and the angle of a given vector. There are 10 vectors in each diagram (this is not always evident due to occasional superposition of individual lines). Perfect oculomotor compensation without vision reversal is represented by a vector of + 1 magnitude and zero phase relative to the + 1 coordinate. Phase advancement is represented by anti-clockwise displacement of the vector, with perfect reversed compensation denoted by the --1 point. Note that the bottom plots were obtained in 8 Hz strobe lighting, with reversing prisms in place. All other plots represent results obtained in the dark.
42
C
S2 BEFORE ADAPTATION (DARK)
EARLY ADAPTATION (DARK)
(DARK)
/-'
i
-1
t
+1
! -1
I
-1
'
'
'
'
+I
-1
I
+I
L
+1
-1
+I
+1
-1
+1
a ~"',
,
t
,
I
LATE ADAPTATION
(LIGHT~ WITH
PRISMS)
_II
Fig. 4. Adaptation to vision reversal in normal light : polar plots of VOR gain and phase, obtained from S~ and C before and during adaptation to reversed vision in normal light. Details as in Fig. 3, except for the bottom plots which were obtained in normal light with prisms in place. conditions both strobe reared cats performed similarly to the parallel control animal (Figs. 3 and 4). Thus, strobe rearing had apparently not prejudiced the ability of the adapted animals to combine both visual and vestibular drives toward the ideal goal of reversed tracking. Nevertheless, at this relatively early stage of the adaptive process, the realization of the ideal condition for reversed tracking was far from achieved, as indicated by the very wide dispersion of phase and low gain seen in both strobe reared and control animals. This contrasts with previous long-term experiments on cats, where virtually perfect reversed tracking was eventually found to occur under similar stimulus conditions in normal light 8. It is worth noting that, in a series of parallel experiments with normal and strobe reared cats, concerned with the influence of VOR adaptation on optokinetic tracking, marked changes in oculomotor responses to monocular nasal and temporal optokinetic stimulation were observed in the adapted animals 14. Consequently, the adaptive changes observed here were presumably not confined to the vestibular system. DISCUSSION The original hypothesis, regarding the dependence o f V O R adaptatioP on retinal image slip, suggested that movement sensitive visual cells with directional preference
43 were important elements in the mediation of the adaptive process. This, in turn, implied that animals lacking a full complement of directionally selective visual cells should be deficient in VOR adaptation. Thus, cats reared in total darkness and lacking direction selective neurons in both the visual cortex and the superior colliculus2, 22, exhibit a lowered VOR gain of about 0.3~).5 (as measured in the dark). These animals are also deficient in their ability to adaptively modify this gain, even under conditions which are known to promote such adaptation in normal animals10. The present experiments were an attempt to examine the role of direction selective visual cells after selective deprivation of exposure to moving stimuli. The most important conclusion emerging from the present results is that both strobe reared and normal cats showed very similar patterns of adaptation of their VOR to optical reversal of vision, in that both types of animals exhibited an initial fast, and a subsequent much slower, stage of gain attenuation. Furthermore, each of the two stages proceeded at similor rates in strobe reared and control animals, and independent of whether adaptation occurred in strobe or normal light. Since the directional component of the implied retinal error signal would be of paramount importance in the adaptive process, the fact that the deprived animals performed indistinguishably from the normal controls provides strong evidence against the involvement of cortical and collicular motion sensitive cells in normal VOR adaptation. This in turn, suggests a predominant utilization of the accessory optic system in the adaptive process, and by implication, a sparing of directional sensitivity of this system in strobe reared cats. These latter conclusions are in line, on the one hand, with the observations of Ito et al. 13 in rabbits, that ablation of cerebral cortex has no effect upon VOR adaptation; and on the other hand, with the suggestion that projections from the accessory optic tract (AOT), via olivary climbing fiber input to vestibular cerebellum, may be responsible for the modulation of the synaptic efficacy responsible for VOR adaptation 6,12. Interestingly, direction selectivity of cells in the nucleus of the optic tract remains unchanged after lesions of the visual cortex11. The present results also demonstrate another important aspect of VOR gain attenuation, common to both normal and strobe reared cats, and to humans: reference to Table III shows that adaptation to vision reversal proceeds, in both animals and humans, as a two time constant process, with the initial fast period measured in hours, and the subsequent slow period in days. These two time courses are quite distinct from one another (P ,~ 0.001), and therefore presumably represent two quite different processes - an observation which confirms an earlier suggestion based on an experimentally derived model of VOR adaptation in humans 4, to the effect that at least two processes, with different time constants, act simultaneously to produce the overall adaptive phenomenon. The similar values for cats and humans shown in Table III strongly suggests that similar fast and slow adaptive neural mechanisms are at play in primates and carnivores. Despite the similarities in adaptive VOR gain attenuation discussed above, reference to Figs. 3 and 4 and Table I indicates that the dark tested VOR before adaptation differed in two significant ways between strobe reared and control animals. First, the phase advancement in strobe reared animals was larger than would normally
44 be expected under the test conditions. This could imply a decrease due to strobe rearing in the overall VOR time constant which is now thought to depend on interaction between peripheral and central dynamics 21. Since it seems unlikely that the peripheral component has undergone any change, and following Robinson's argument 21, the observed phase advancement would seem to implicate central components, specifically a reduced positive feedback gain in the 'optokinetic loop'. In view of this prediction it is intriguing to note that the optokinetic gain tested with prolonged (1 min) unidirectional stimulation in the strobe reared animals appeared significantly reduced, as reported elsewhere 14. Second, in strobe reared animals there was increased variability of phase within a given test series of 10 cycles of sinusoidal oscillation. It seems unlikely that this could be attributed to the presence of spontaneous eye oscillations per se, both on account of the small amplitude and frequency of these eye oscillations (Fig. 1), and because there was no significant difference (P ,~ 0.001) between the correlation coefficients of sinewaves fitted to the curves of cumulative eye position (CEP. see Methods) of strobe reared and control cats. It is conceivable that neurological 'jitter', originating from the strobe rearing environment, might contribute to the enlarged phase dispersion. It is also interesting to note that in the late stage of adaption, the dark tested VOR appears to have been, on the average, less phase shifted in the strobe reared cats than in the control one (Figs. 3 and 4). Nevertheless, in both normal and strobe illumination, the VOR adapted strobe reared cats did as well as the control animal in the attempt to produce reversed ocular compensation (Figs. 3 and 4). Lastly, it is interesting to note that, whereas the adaptive response in the present experiments was as great in strobe light as in normal light (for both strobe reared and normal cats), in the previous human experiments 17 the rate of adaption in strobe light was only about half that in normal light. This apparent discrepancy might be attributable to either a species differences; and/or to a difference in strobe flash frequency, which was 8 Hz in the present experiments, but only 4 Hz in the previous human study. ACKNOWLEDGEMENTS The present work was supported by the Canadian MRC, the Canadian NSERC, and USPHS Grant EY02248 to M.C. The scleral eye coil implantations were done by Dr. D. Guitton. The invaluable technical assistance of Mr. T. O'Farrel and Mr. W. Ferch is gratefully acknowledged. REFERENCES 1 Cynader, M. and Chernenko, G., Abolition of direction selectivityin the visual cortex of the cat, Science, 193 (1976) 504-505. 2 Cynader, M., Berman, N. and Hein, A., Recovery of function in cat visual cortex following prolonged deprivation, Exp. Brain Res., 25 (1976) 139-156. 3 Davies, P. R. T., Neural Adaptation in Humans and Cats Sube]cted to Long-term Optical Reversal of Vision: an Experimental and Analytical Study of Plasticity, Ph.D. Thesis, McGill University, Montreal (1979). 4 Davies, P. R. T. and Melvill Jones, G., An adaptive neural model compatible with plastic changes induced in the human vestibulo-ocularreflexby prolonged optical reversal of vision, Brain Research, 103 (1976) 546--550.
45 5 Flandrin, J. M., Kennedy, H. and Amblard, B., Effects of stroboscopic rearing on the binocularity of cat superior colliculus neurons, Brain Research, 101 (1976) 576-581. 6 Ghelarducci, B., Ito, M. and Yagi, N., Impulse discharges from flocculus Purkinje cells of alert rabbits during visual stimulation combined with horizontal head rotation, Brain Research, 87 (1975) 66-72. 7 Gonshor, A. and Melvill Jones, G., Plasticity in the adult human vestibulo-ocular reflex arc, Proc. Can. Fed. Biol. Soc., 14 (1971) 11. 8 Gonshor, A. and Melvill Jones, G., Short-term adaptive changes in the human vestibulo-ocular reflex arc, J. Physiol. (Loncl.), 256 (1976) 361-379. 9 Gonshor, A. and Melvill Jones, G., Extreme vestibulo-ocular adaptation induced by prolonged optical reversal of vision, J. Physiol. (Lond.), 256 (1976) 381-414. 10 Harris, L. R. and Cynader, M. S., Attempts to modify the vestibulo-ocular reflex of normal and dark-reared cats, Neurosci. Abstr., 5 (1979) abstract 2646, p. 787. 11 Hoffmann, K.-P., Behrend, K. and Schoppmann, A., Visual responses of neurons in the nucleus of the optic tract of visually deprived cats, NeuroscL Abstr., 3 (1977) abstract 1790, p. 563. 12 Ito, M., Neural design of the cerebellar motor control system, Brain Research, 40 (1972) 81-84. 13 Ito, M., Shiida, T., Yagi, N. and Yamamoto, M., Visual influence on rabbit's horizontal vestibuloocular reflex presumably effected via the cerebellar flocculus, Brain Research, 65 (1974) 170-174. 14 Mandl, G., Melvill Jones, G. and Cynader, M., Modification of OKN by maintained vision reversal in normal and strobe reared cats, Neurosci. Abstr., 1980, abstract 158.1, p. 473. 15 Meiry, J. L., The Vestibular System and Human Dynamic Space Orientation, NASA CR-628, Washington, D.C., 1966. 16 Melvill Jones, G. and Davies, P. R. T., Adaptation of cat vestibulo-ocular reflex to 200 days of optically reversed vision, Brain Research, 103 (1976) 551-554. 17 Melvill Jones, G. and Mandl, G., Effects of strobe light on adaptation of vestibulo-ocular reflex (VOR) to vision reversal, Brain Research, 164 (1979) 300-303. 18 Melvill Jones, G., Mandl, G., Cynader, M. and Outerbridge, J. S., Eye oscillations in strobe reared cats, Brain Research, 209 (1981). 19 Miles, F. A. and Eighmy, B. B., Long-term adaptive changes in primtae vestibuloocular reflex. I. Behavioral observations, J. NeurophysioL, 43 (1980) 1406-1425. 20 Robinson, D. A., Adaptive gain control of the vestibuloocular reflex by the cerebellum, J. Neurophysiol., 39 (1976) 954-969. 21 Robinson, D. A., Vestibular and optokinetic symbiosis: an example of explaining by modelling. In R. Baker and A. Berthoz (Eds.), Control of Gaze by Brain Stem Neurons, Elsevier, Amsterdam, 1977, pp. 49-58. 22 Wickelgren, B. G. and Sterling, P., Influence of visual cortex on receptive fields in superior colliculus of the cat, J. Neurophysiol., 32 (1969) 16-23.