deprived eye (reverse occlusion). beyond 1 year of age; only two of six animals recovered sufficient vision to enable

Journal of Physiology (1988), 395, pp. 639-66 639 With 8 text-figures Printed in Great Britain THE EXTENT OF VISUAL RECOVERY FROM EARLY MONOCULAR OR BINOCULAR VISUAL DEPRIVATION IN KITTENS BY DONALD E. MITCHELL From the Psychology Department, Dalhousie University, Halifax, N.S., Canada (Received 27 February 1987) SUMMARY 1. The rate and extent of recovery of vision was studied in the deprived eye of kittens that had been monocularly deprived from near birth for periods that lasted from 6 weeks to 18 months. Recovery was measured in the two situations, where either both eyes were open following the initial deprivation (binocular recovery), or else the non-deprived eye was occluded so as to force the animal to employ its deprived eye (reverse occlusion). 2. Measurements were made of the visual acuity of the deprived eye for gratings at frequent intervals during recovery by means of a simple behavioural technique. 3. The acuity that the deprived eye eventually attained declined with increasing length of deprivation in a manner that could be approximated by a simple exponential decay. Only minimal visual recovery was observed in animals deprived beyond 1 year of age; only two of six animals recovered sufficient vision to enable measurement of visual acuity. In general, animals that were reverse occluded recovered better vision than did those that had both eyes open during recovery. 4. The recovery of vision in the deprived eye of monocularly deprived kittens was compared to that observed following equivalent periods of two forms of binocular deprivation, namely dark-rearing and binocular eyelid suture. 5. The recovery from the two forms of binocular deprivation was quite different. Whereas the extent of recovery from dark-rearing was considerably greater than that observed after equivalent periods of monocular deprivation, the recovery of a limited sample of cats that were binocularly deprived by eyelid suture was worse. 6. These findings suggest that some plasticity remains in the visual pathway for a longer time than indicated by experiments that examine the physiological effects of monocular deprivation on various visual cortical structures. INTRODUCTION The developing visual system of many species of frontal-eyed mammals can be influenced in a dramatic fashion by the nature of the visual input during the first few months of post-natal life (Movshon & Van Sluyters, 1981; Mitchell & Timney, 1984). Much of what is known of environmental influences on post-natal visual development has been derived from studies of the substantial anatomical, physiological and behavioural consequences of monocular deprivation, an extreme form of abnormal

64 D. E. MITCHELL visual input in which one eye is deprived for a time of patterned visual input. Immediately following a period of monocular deprivation imposed from birth on either cats or monkeys, and extending to 5 weeks of age or longer, virtually no cortical cells in layers other than layer IV of cortical area 17 can be excited by visual stimuli presented to the deprived eye (Wiesel & Hubel, 1963, 1965; LeVay, Wiesel & Hubel, 198). At the same time, monocularly deprived kittens appear blind when using only their deprived eye (Giffin & Mitchell, 1978). Subsequent to the original studies of Wiesel & Hubel (1963), a sensitive period was defined for the physiological effects of monocular deprivation which was originally thought to extend to about 4 months of age in kittens (Hubel & Wiesel, 197) and to about a year in monkeys (LeVay et al. 198). The original concept of a sensitive period defined in terms of the effects of periods of monocular deprivation at different ages was broadened by later studies that examined the extent of the subsequent physiological recovery (Blakemore & Van Sluyters, 1974; Blakemore, Garey & Vital-Durand, 1978). Although some functional recovery of influence by the deprived eye can occur in the visual cortex of kittens on simply restoring patterned visual input to the deprived eye (Mitchell, Cynader & Movshon, 1977a; Olson & Freeman, 1978), far greater recovery occurs when the formerly non-deprived eye is occluded at the same time. The procedure of reverse occlusion is particularly effective in monkeys who appear to exhibit little or no functional recovery in the situation where the nondeprived eye remains open during the recovery period (LeVay et at. 198; Blakemore, Vital-Durand & Garey, 1981). For both cats and monkeys reverse occlusion has been found to be effective only when applied during a period of life which was of similar (but somewhat shorter) duration than the sensitive period defined in terms of susceptibility to the effects of a single period of monocular occlusion (Blakemore & Van Sluyters, 1974; Olson & Freeman, 198; LeVay et al. 198). On the basis of these findings the sensitive period can be thought of as a time in life during which functional inputs to cells in the visual cortex can be both disconnected and restored in response to certain extreme conditions of early visual exposure (Movshon, 1976). Prompted by observations of behavioural recovery in cats monocularly deprived to 4 or 5 months of age, Cynader, Timney & Mitchell (198) examined the effects of longer periods of monocular deprivation imposed on kittens at later ages than was the case in previous studies. Obvious shifts of ocular dominance were evident in animals deprived as late as 7 months of age. In a later study that employed a similar rearing regimen, Jones, Spear & Tong (1984) observed physiological effects of monocular deprivation in area 17 as late as 9 months of age. These results indicate that some plasticity remains in the visual cortex for a far longer time than earlier experiments indicated. These findings are reinforced by observations described in this paper of the extent of behavioural recovery from extended periods of monocular deprivation imposed on kittens from birth. These data represent an extension of earlier work (Giffin & Mitchell, 1978) that examined the rate of recovery of vision following brief periods of monocular occlusion. Together, they form part of a larger project that examined factors that promote recovery from the effects of monocular deprivation as a first step towards the elucidation of optimum treatment strategies for various forms of

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION 641 human amblyopia. These measurements of recovery from periods of monocular deprivation of different duration provide a very sensitive probe of residual plasticity within the visual pathway. Although the site and underlying mechanism of this recovery cannot be established from measurements of this sort, nevertheless they permit an estimate to be made of the latest age at which neural connections at any level within the visual pathway are influenced by visual experience. The time course and magnitude of recovery is documented for two extreme recovery conditions, namely reverse occlusion and the situation where both eyes are open following the period of monocular deprivation (binocular recovery). The recovery observed in these two situations is compared to that observed in a more limited sample of animals following equivalent periods of two forms of binocular visual deprivation, namely dark-rearing and binocular deprivation by eyelid closure. METHODS Animala and rearing condition8 Data on the visual recovery from monocular deprivation was obtained over a 12-year period from forty-eight kittens that were reared from birth in a closed laboratory colony. Fourteen of the kittens were reared specifically for this study and provided most of the longitudinal data on recovery from periods of monocular or binocular deprivation lasting 6 months or longer. The remaining kittens were reared for related experiments (Giffin & Mitchell, 1978; Cynader et al. 198; Mitchell, Murphy & Kaye, 1984; Mitchell & Murphy, 1984; Mitchell, Murphy, Dzioba & Home, 1986) and provided much of the data on the initial stages of recovery or else the final acuity that was eventually attained (summarized in Figs 2 and 3). Eleven of these animals were reared for the purpose of the above studies, but for a variety of reasons (such as the presence of a temporary corneal opacity) longitudinal measurements of acuity were not made throughout recovery and hence their data were not included in the published reports of these experiments. The rearing histories of the animals reared specifically for this study, and which are described here for the first time, are listed in Table 1. The recovery from binocular deprivation was documented in nineteen cats that were either darkreared (n = 17), or else subjected to bilateral eyelid suture (n = 2), from before the time of natural eye opening for various periods of time. The rearing histories and detailed descriptions of the results from nine of these animals are provided in earlier papers (Timney, Mitchell & Giffin, 1978; Kaye, Mitchell & Cynader, 1982). Details concerning the rearing histories of the remaining animals are listed in Table 2. The procedures for dark-rearing were as described earlier (Timney et al. 1978). However, since the animals described here were deprived for longer periods than those of the earlier study, efforts were made to ensure that the animals received more extensive handling each day in the darkroom throughout the period of deprivation in order to speed the process of socialization following exposure to light. The monocularly deprived kittens had the eyelids of one eye closed by eyelid suture at about the time of natural eye-opening using conventional surgical procedures that have been described previously (Giffin & Mitchell, 1978). After various periods of deprivation the sutured eyelids were parted to restore patterned visual input to that eye. For some kittens the eyelids of the other eye were closed at the same time (reverse occlusion) in order to force the animal to use the formerly deprived eye, while the remaining kittens had both eyes open after the period of monocular deprivation. Because the vision of the deprived eye was reduced to blindness for some time following termination of the initial period of monocular deprivation, the number of animals that were reverse occluded during recovery was kept to a minimum. The kittens were housed with their mothers in laboratory cages until weaning and thereafter in either a larger custom-built cage (1-2 m long, 9 m wide and -85 m high) that contained as many as five kittens, or else in a pen (3 by 2-4 m) within the colony room. The colony room itself was illuminated with tungsten lighting with a daily cycle of 16 h light and 8 h of darkness. The condition of the sutured eyelids was inspected carefully each morning for any small opening that may have developed overnight. These occurred in only a few cases and were repaired immediately 21 PHY 395

642 D. E. MITCHELL when noticed. The eyes were checked daily for opacities of the ocular media which might degrade the quality of the retinal image and hence reduce acuity. Corneal clouding associated with a small patch of epithelial erosion developed suddenly in the deprived eye of six animals. Appropriate treatment was initiated immediately and behavioural testing curtailed. The corneal epithelium of three of the animals responded rapidly to treatment within 2 days and all cloudiness of the cornea disappeared after a week. Nevertheless, only the data collected prior to the opacity being noted are reported here. The other three animals did not show any evidence of recovery in 3 days and so were immediately destroyed. TABLE 1. Length of deprivation, recovery condition and visual acuity attained by the deprived eye of fourteen cats that were monocularly deprived from the time of natural eye opening Termination of Recovery condition: deprivation binocular (B) or Visual acuity Name (days of age) reversal (R) (cycles/deg) K8 113 R 4-65 C138 134 R 3-2 C19 141 R 2-12 K141 182 BB 2-7 K214 186 2-5 K287 245 B -8 C49 284 B OD Al 32 B 1-38 K284 364 B OD C35 365 B -22 C57 42 B -16 C5 43 BB OD I 456 BL C51 546 R TABLE 2. Length of deprivation and eventual visual acuity of eight dark-reared (DR) cats and two animals binocularly deprived by lid suture (BLS) Length of Nature of deprivation deprivation Visual acuity Name (days) (DR or BLS) (cycles/deg) K51 42 DR 6-17 K52 42 DR 5.75 K26 122 DR 5.5 HK558 24 DR 2-3 K11 3 DR HK519 3 DR 2- CIO 365 DR 1X57 C12 365 DR 3X92 C71 183 BLS 1X76 C79 211 BLS L Prior to the time of natural eye-opening, those animals that were dark-reared were placed with their mothers (until weaning) in standard laboratory cages in a completely dark room, a description of which is provided elsewhere (Timney et al. 1978). While the animals remained in the dark they were petted for at least 15 min daily in order to help prepare them for the more extensive contact with humans that they would experience once they were brought into an illuminated environment. As a consequence, the animals adjusted rapidly to the extensive handling associated with the daily test of visual capacity on the jumping stand. The two cats that were binocularly deprived by bilateral eyelid suture were reared in the same fashion as their monocularly deprived litter-mates. Whereas the eyelids of both eyes of one of these animals (C79) were parted to allow binocular visual

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION input following the period of deprivation, only the left eye of the other animal (C71) was open during the period of recovery. Behavioural testing Behavioural measurements of the monocular or binocular visual acuity were made on a jumping stand using procedures that have been described extensively in several previous papers (Mitchell, Giffin & Timney, 1977b; Giffin & Mitchell, 1978; Mitchell et al. 1984; Murphy & Mitchell, 1987). The procedures have been modified considerably during the 12 years that have elapsed since this study began in order to decrease the time required to obtain a measurement of acuity, and to improve the repeatability of the measures. The most important of these procedural modifications are described in the most recent papers listed above (Mitchell et al. 1984; Murphy & Mitchell, 1987). However, it should be emphasized that these procedural modifications did not result in any significant change in absolute acuity levels. Measurements of the acuity of normal animals lay within a range of 6-4-8-7 cycles/deg throughout the course of this study. Those animals subjected to extended periods of deprivation remained apparently blind in one or both eyes for long periods and eventually regained only rudimentary vision. Certain operational definitions of rudimentary visual capacities had to be made in order both to document the time of onset of measurable vision in these animals, and to define the nature of the vision that was recovered. In order to describe these operational definitions, and the modifications that had to be made to the general procedure in order to document the time course of visual recovery in animals that could not be trained prior to termination of deprivation (as was the case for all the binocularly deprived animals), it is necessary to review the procedure as it is usually applied on normally reared animals. Typically, animals would be trained to discriminate between a square-wave grating (19-5 x 19-5 cm) and an adjacent uniform grey field of the same size and mean luminance (12 cd/m2) that were placed on the two trapdoors of a jumping stand. The position of the two stimuli would be interchanged from right to left in a pseudorandom order with correct jumps rewarded with pureed chicken liver and petting. Incorrect choices resulted in denial of both these rewards and the kitten was required immediately to repeat the trial. Once the discrimination was learned, measurements of visual acuity could be obtained in a single session by systematically incrementing the spatial frequency of the grating in very small steps between successive blocks of trials until a spatial frequency was reached where the animal was no longer able to maintain a criterion level of performance. The increments in spatial frequency between blocks of trials were approximately equal on a logarithmic scale, being separated by I of an octave for gratings having periods between 1 and 4 mm, 1,L of an octave for gratings having periods between 3 and 1 mm, and of an octave for periods below 3 mm (corresponding to spatial frequencies of approximately 4 cycles/deg or higher at the largest observation distance of 7 cm). As soon as the animal had learned the basic discrimination in the manner described by Giffin & Mitchell (1978), measurements of acuity could begin. A typical testing session began with gratings of a very low spatial frequency that was several octaves from the animal's threshold. Usually the animal received only one trial at each of the very low spatial frequencies unless an error was made, in which case the kitten received additional trials until it either made five consecutively correct responses or else made seven correct jumps in a maximum of ten trials. Within approximately an octave of threshold the minimum number of trials at each spatial frequency (when performance was flawless) was increased to three. For the three or four spatial frequencies closest to threshold the minimum number of trials per block was increased once more to five. Sessions were terminated as soon as the animal was no longer able to attain criterion performance as defined above within the maximum number of trials (ten) allowed at any spatial frequency. Where it was suspected that the animal's motivational state was poor, the procedure was repeated once, or even twice more in the same session beginning with a spatial frequency one octave lower than that on which the animal previously failed. Typically performance fell from flawless to chance within two incremental steps in spatial frequency, equivalent to an increase of approximately 12 %. As a consequence thresholds could be titrated very precisely. Thresholds were defined conservatively as the highest spatial frequency for which criterion performance could be maintained. At this spatial frequency the animal's performance was at best flawless and at worst 7 % correct in the situation where the maximum ten trials were required for that particular block of trials. This procedure resulted in thresholds that were highly repeatable between sessions, even in the usual situation where the measurements were made by different people. 21-2 643

644 D. E. MITCHELL Because the periods of monocular deprivation were so long, the animals were fully trained long before vision was restored to the deprived eye. Also, because of the long period of deprivation the deprived eye remained blind for some time after normal visual input was restored. Consequently the procedure employed to document the recovery of vision of the deprived eye necessarily had to incorporate procedures employed in initial training, particularly in the situation where the animal was reverse occluded during recovery. In the alternative condition, where both eyes were open during recovery, the vision of the deprived eye was tested after inserting a large opaque contact lens occluder in the non-deprived eye. A drop of local anaesthetic (proparacaine hydrochloride, 5%) was first instilled into the eye in order to eliminate discomfort. On initial testing, the animals appeared blind when forced to employ their blind eye. Formal tests for visually triggered placing, visual startle responses, or obstacle avoidance were all negative. On the jumping stand blindness was defined operationally in the manner described by Giffin & Mitchell (1978). Briefly, the jumping platform was lowered so that it was only a few centimetres above the trapdoor surfaces on top of the jumping stand. One of the trapdoors was opened while a grating having the largest period (4 mm) was placed on the other. The animal was prompted from the jumping platform from which it would typically emerge in a very cautious manner by first placing one paw onto the wooden divider that separated the two trapdoors. On emerging further it would usually begin to reach with its free paw for the closed (safe) surface. If it encountered nothing with this paw it would then take a further step in order to free the other paw which it used to search for the trapdoor on the other side. On encountering a closed door with its paw it emerged slowly onto this surface. At first it was obvious that the animal could only distinguish the closed from the open trapdoor by use of such tactile cues. However, with the passage of time the animal began to step onto the closed door immediately without searching for it first with one paw. At this point the jumping platform was elevated gradually so that the animal was now forced to jump rather than walk onto the grating, an ability that required some visual ability. The ability to make this discrimination without error on 2 consecutive days was taken as an operational definition for the onset of vision. As soon as the animal was able to distinguish a closed from an open door on the jumping stand by use of visual cues alone, the open door was closed and a grey uniform field of the same mean luminance as the grating was placed on top in order to test formally for the presence of pattern vision. As in initial training, the animal was rewarded only if it jumped toward the grating. When the period of initial deprivation was reasonably short, this discrimination was possible either immediately, or within a few days of the animal first discriminating an open from a closed door. However, some animals subjected to very long periods of monocular deprivation eventually recovered only this ability and were never able to solve the rudimentary pattern discrimination, while other animals deprived for somewhat shorter periods were able to solve the latter discrimination only after several months had elapsed. Since the discrimination of an open from a closed door could be done theoretically on the basis of vision as crude as the ability to discern simple luminance differences, certain animals were formally tested on the jumping stand for their ability to make a luminance discrimination between a white (luminance 25 cd/m2) and a black stimulus of the same dimensions and having a luminance of only 5 cd/m2, in order to document more precisely the nature of the vision that was recovered. Cats were classified as having passed this discrimination if they made twenty consecutively correct choices on two successive sessions. On the Figures that follow, the ability to solve this formal test of luminance discrimination is signified by the letter 'L' to distinguish it from the less immediately informative ability to discriminate an open from a closed door on the jumping stand (signified by the letters OD in the Figures that follow). Testing of visual acuity was terminated when the acuity of the deprived eye had remained stable for at least a month (2 months when the period of deprivation was long). Measurements of contrast sensitivity functions were made on two animals (C229 and C288) once the visual acuity of the deprived eye had stabilized. Details of the procedures employed are provided in earlier papers (Mitchell et al. 1984; Murphy & Mitchell, 1987). Stimuli were generated as a split-screen display on a large oscilloscope display monitor (Joyce Electronics, Cambridge). At the usual observation distance of 57 cm, the sinusoidal gratings (luminance 135 cd/m2) that formed the positive stimulus (adjacent to a uniform field of the same average luminance that served as the negative stimulus) were 13 deg wide and 18 deg high. Consequently, at the lowest spatial frequency ( 37 cycles/deg) on which measurements were made (C229), only 3-5 cycles of the grating were visible which may have resulted in artificially low contrast sensitivities.

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION 645 RESULTS Monocular deprivation Irrespective of whether the non-deprived eye was open or closed following an early period of monocular deprivation, the recovery of vision in the deprived eye exhibited two obvious trends with increasing length of deprivation. First, it took progressively 6- C282 C127 m4 S... _ : * t555*; o C 2 Br O. B. 4 8 12 16 2 Recovery time (days) Fig. 1. The recovery of visual acuity of the deprived eye of three animals that were monocularly deprived from near birth until either 42 (C282), 94 (C127) or 32 (Al) days of age. Both eyes were open following the early period of deprivation. During the period designated by the letter 'B' on the ordinate the animal appeared blind on the jumping stand. Throughout the period indicated by the horizontal line corresponding to zero on the ordinate, Al was capable only of discriminating an open from a closed door, but appeared incapable of passing a formal test of pattern vision. Between the times indicated by open arrows, the non-deprived eye of C127 was occluded so as to force the animal to employ the initially deprived eye. longer for animals to exhibit signs of vision with their deprived eye, and second, the vision that this eye eventually attained decreased from acuity levels that were close to normal for very short periods of deprivation, to only rudimentary light perception when deprivation was extended to a year or longer. These trends are evident in the data from the three representative animals displayed in Fig. 1, that were monocularly deprived from near birth until either 42 (C282), 94 (Cl27) or 32 (Al) days of age at which time the deprived eye was opened to allow visual input to both eyes during recovery. The first of these animals (C282) exhibited signs of pattern vision with its deprived eye after 5 days, while the animal deprived to 94 days did not show any indication of pattern vision for 17 days. Finally, the animal deprived to 32 days appeared completely blind with its deprived eye for 6 days at which point it first showed evidence of an ability to discriminate an open from a closed door on the jumping stand. No further improvement in vision was evident for more than 2 months at which time it was able to pass a formal test of pattern vision for the first time. Thereafter the vision of this eye improved very gradually to eventually attain %*

646 D. E. MITCHELL an acuity of only 1-35 cycles/deg. By contrast, the vision of the deprived eye of the other two animals improved substantially more, to an acuity in excess of 5 cycles/ deg for C282, while C127 achieved an acuity of 3-45 cycles/deg. During the period indicated by the open arrows, the non-deprived eye of the latter animal was sutured closed in order to ascertain whether forced use of the deprived eye might lead to some further improvement in the vision of this eye. Clearly no such improvement occurred during this 8 week period of reverse occlusion. 1 8 ( > ) 6 -~ '4 -. U_ X2 - X - ~~~~ - OD- B-1 I 2 4 6 Period of deprivation (days) Fig. 2. The visual acuity eventually achieved by the deprived eye of thirty-seven monocularly deprived cats as a function of the length of the prior period of deprivation. Filled symbols show the final acuities of seventeen animals that were reverse occluded following the initial period of monocular deprivation, while open symbols indicate results from the twenty animals that had both eyes open during recovery. Data are shown for all fourteen cats of Table 1, and twenty-three animals reared during the course of previously published studies (Giffin & Mitchell, 1978; Cynader et al. 198; Mitchell et al. 1984). The deprived eye of five animals that were deprived for long periods never acquired any pattern vision. One appeared completely blind in this eye (signified by the letter 'B' on the ordinate), three were able to discriminate only an open from a closed door on the jumping stand (OD), while the fifth animal could, in addition, solve a simple luminance discrimination with its deprived eye (L). The square symbols show data of Smith & Holdefer (1985) obtained from six animals that were monocularly deprived until they were either 6 or 12 months old and which subsequently were either reverse occluded (filled symbols) or else received binocular visual exposure during recovery. The data on recovery from all the monocularly deprived animals are summarized in Figs 2 and 3 which display plots of either the time to recover first signs of vision (Fig. 3), or the final acuity attained (Fig. 2), as a function of the length of the period of deprivation. Results are shown for animals that were reverse occluded (filled symbols) as well as for those that had both eyes open during recovery (open symbols). Figure 2 also includes data from six animals (square symbols) from a study of Smith & Holdefer (1985) who employed similar procedures (including the use of a jumping stand) and conditions to document the extent of recovery from periods of

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION 647 deprivation that extended to either 6 or 12 months of age. Besides the trend that was mentioned earlier for the extent of visual recovery to decline with increasing length of deprivation, it was also apparent from Fig. 2 that animals that were reverse occluded recovered somewhat better vision in their deprived eye than did those that 1 C 1 V E4--.._ (u 1 @ * / o. 1 1 5 1 5 1 Period of deprivation (days) Fig. 3. The time (T) to exhibit the first sign of vision on the jumping stand as a function of the length (t) of a prior period of monocular deprivation. Symbols are as defined for Fig. 2. The data for both conditions of recovery have been fitted by the function: T = 7 t1f675. Data are shown for forty-six animals, including thirteen from Table 1, that were reared for previously published studies. and thirty-three had both eyes open following the initial period of monocular deprivation. This was particularly obvious in the case of animals deprived for only brief periods of time, a conclusion in agreement with the findings of an earlier study (Giffin & Mitchell, 1978). Despite the difference between the final outcome from these two recovery conditions, the initial stages of recovery in these two situations (Fig. 3) were remarkably comparable (see also Giffin & Mitchell, 1978), an impression that was confirmed by the lack of any statistically significant difference between the results shown in Fig. 3 for the two recovery conditions. The final acuities achieved by animals in both recovery conditions declined with increasing length of deprivation in a manner that could be fitted satisfactorily by an exponential having the form S(t) = A exp (-at),

648 D. E. MITCHELL where S(t) is the spatial resolution eventually achieved in the deprived eye following monocular deprivation to t days of age, and A and a are two constants. For animals that had both eyes open during recovery, A was 7-92 cycles/deg and a was -86, while the corresponding values for the data from animals that were reverse occluded were 1-88 cycles/deg and 94, respectively. The data of Fig. 3, which displays on logarithmic scales the time to first recover vision as a function of the length of deprivation, are well fitted (with a regression coefficient of 93) by the expression: T = 7 t1675, where T is the time to first discriminate an open from a closed door on the jumping stand, and t is the length of deprivation as before. A single equation was fitted to the data from both recovery conditions since the early stages of recovery in the two situations appeared closely comparable (see also Giffin & Mitchell, 1978). This impression was confirmed by the lack of any statistically significant difference between the results from the two recovery conditions. Following visual deprivation to only 4-6 weeks of age, the deprived eye showed signs of vision in 1-5 days. However, after periods of monocular deprivation that lasted 12 or 18 months, animals appeared blind for at least 4, and as long as 12, months. As noted earlier (Giffin & Mitchell, 1978), the vision of the deprived eye of animals monocularly deprived to 2 months of age or less can show substantial recovery of grating acuity, even in the situation where both eyes are open during recovery. However, it should be recognized that the good grating acuity of the deprived eye may provide a false impression of the status of form perception of this eye as a whole since these animals may demonstrate much larger deficits on other measures of spatial resolution, such as vernier acuity or contrast sensitivity. Because procedures that enable measurements of such visual capacities by use of a jumping stand have been developed just recently, data on contrast sensitivity or vernier acuity have so far been collected from only a few animals (Mitchell et al. 1984, 1986). Representative results of measurements of contrast sensitivity functions for both eyes of two animals that were monocularly deprived until either 42 (C229) or 6 (C288) days of age are shown in Fig. 4. Both animals received binocular visual input during the period of recovery that followed the early deprivation. Although the visual acuity of the deprived eye ofc229 recovered in 2 months to 4-7 cycles/deg, a value that was only 3% lower than the acuity of the non-deprived eye (6-6 cycles/deg) at that time, the contrast sensitivities of the two eyes differed by a factor of about 1 at all spatial frequencies. An even more remarkable divergence between the different measures of spatial resolution was evident in the results from the second animal (C288) where the contrast sensitivities of the two eyes differed by about a factor of 2, as compared to less than a twofold difference between their visual acuities. Preliminary results suggest that the discrepancies between measures of vernier and grating acuity are at least as large as those observed in Fig. 4 between the latter measure and contrast sensitivity. Interestingly, substantial discrepancies between different measures of spatial resolution are a characteristic feature of amblyopia in humans (e.g. Levi & Klein, 1982; Howell, Mitchell & Keith, 1983). As is evident from Table 1 and Fig. 2, animals deprived to 1 months of age or

.649 VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION longer show only limited recovery of vision in the deprived eye. Including the data of Smith & Holdefer (1985), five of the nine animals deprived to between 3 and 456 days of age recovered some pattern vision with the deprived eye (defined as the ability to discriminate a grating of very low spatial frequency from a uniform field of the same mean luminance). However, none of these animals recovered more 5 - C229 C288 1~ - -~~~~~~~~~~~ 1 ~ ~~~~~~~~~ o 1 5 1 1 5. Spatial frequency (cycles/deg) Fig. 4. Contrast sensitivity functions for both the deprived () and non-deprived eye (*) of two cats that had been monocularly deprived to either 42 (C229) or 6 (C288) days of age and which subsequently received binocular visual exposure. The open and filled arrows represent the grating acuities for square-wave gratings of respectively, the deprived and non-deprived eyes. than rudimentary grating acuity, that ranged from -2 cycles/deg for C35, to 1-38 cycles/deg for Al, who were deprived respectively, to 12 and 1 months of age. For reasons mentioned earlier, all but one of the animals (C51) that were deprived to 6 months of age or longer had both eyes open during recovery, a situation that in general leads to less recovery of vision by the deprived eye than that which occurs following reverse occlusion where the animal is forced to use its deprived eye throughout recovery. In an effort to determine if the latter procedure might promote further recovery over that which occurred during binocular recovery, the formerly non-deprived eye of two animals was sutured closed after a prolonged period of binocular recovery. One of these animals (C49), that was monocularly deprived to 284 days of age, and subsequently had both eyes open for the first 7 months of recovery, acquired nothing more than the ability to discriminate an open from a closed door on the jumping stand during this time. The vision of this eye showed no further improvement in the first 8 weeks following reverse occlusion initiated after 21 days of binocular recovery. However, during the next 3 months the vision of this eye improved gradually to an acuity of 1-25 cycles/deg, after which it remained unchanged.

65 D. E. MITCHELL A similar improvement in vision was evident in the deprived eye of the second animal (C35) that was initially deprived to 1 year of age. At the time the nondeprived eye was sutured closed after 247 days of binocular recovery, the vision of the deprived eye had recovered to only -15 cycles/deg (see Fig. 5), where it had C35 2 - <V%<D _<D D P<c 2-1 2 3 " * Period of deprivation (days) Fig. 5. The recovery of visual acuity of the deprived eye of a cat (C35) that was monocularly deprived to 1 year of age and which subsequently had both eyes open. The animal appeared blind with its deprived eye on the jumping stand (B) for more than 3 months, but eventually acquired pattern vision. After 8 months the deprived eye had recovered an acuity of only 15 cycles/deg. During the period between the two arrows the formerly non-deprived eye was closed by eyelid suture. In this period the acuity of the deprived eye improved to 1-45 cycles/deg. However, upon restoring vision to the other eye, the vision of the deprived eye declined rapidly to the level it had attained before reverse occlusion. remained constant for the previous 3 months. Following occlusion of the nondeprived eye, the vision of the deprived eye apparently dropped for 3 weeks, possibly because sufficient patterned light passed through the eyelids of the non-deprived eye to cause suppression of the vision of the deprived eye for a time. However, eventually the vision of the deprived eye began to improve to values (1-45 cycles/deg) that were well in excess of those achieved during the long preceding period of binocular recovery. Interestingly, the recovery promoted during the short period of occlusion of the non-deprived eye did not appear to be permanent; on restoring visual input to this eye, the vision of the deprived eye dropped rapidly in only 11 days to the level it had reached during binocular recovery. Binocular deprivation The general pattern of recovery of animals dark-reared to 4 or 6 months of age has been described earlier (Timney et al. 1978). In agreement with these earlier findings, the extent of visual recovery from extended periods of dark-rearing was greater than that observed following equivalent periods of monocular deprivation. For example, kittens dark-reared to 4 months of age develop in just 4 months grating acuity that can approach normal levels (Timney et al. 1978; Kaye et al. 1982). By contrast, following an equivalent period of monocular deprivation the visual acuity of the deprived eye recovers to levels that are approximately one-half of normal

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION 651 values (Fig. 2). A similar trend was evident in the results from animals that were dark-reared for even longer periods of time. However, with increasing length of deprivation there was also a much greater variation in the degree of recovery exhibited by different animals. This latter characteristic is exemplified by the findings from two litter-mates (C1 and C12) that were dark-reared to 1 year of age 2 -C1 1 _ *### #~~~~d ) "a ~ 1 -._ C 1 s b U I I, I 4 3 C12 2- C/) OL@@ 8 12 2 24 Recovery time (days) Fig. 6. The recovery of visual acuity of two litter-mates (C1 and C12) that had been reared in total darkness from birth until 1 year of age. During the period indicated by the continuous horizontal line the animals appeared blind on the jumping stand. and whose results are displayed in Fig. 6. Both animals appeared blinxd on the jumping stand for about 2 months following introduction to illuminated surroundings. Signs of pattern vision were manifested by C1 about 3 weeks before its litter-mate, but for the next 3 months the vision of both animals improved at about the same rate. However, whereas the vision of CIO showed no further improvement after it had been in the light for 4 months, the visual acuity of C12 suddenly increased by a factor of two after 7 months of visual exposure. The data for all seventeen dark-reared animals are shown by filled symbols in Figs 7 and 8, which show, respectively, the time required for the animals to first show signs of vision on the jumping stand and the final acuity they eventually attained. The data of Fig. 7 have been fitted (with a regression coefficient of 9,O) by the expression: T = 57 t"199) where, as before, T is the time to first discriminate an open from a closed door on the jumping stand, and t is the length of dark rearing. As evident in comparisons of data from individual animals (e.g. Figs 1 and 6), as well as in the overall summaries (Figs 3 and 7), recovery from dark-rearing tended to be faster than that which occurred after equivalent periods of monocular deprivation. However, a similar statement does not appear applicable to the limited data from two animals subjected to the other form of binocular deprivation, namely bilateral lid suture. One of these

652 D. E. MITCHELL animals (C79), that was binocularly deprived to 211 days of age, and subsequently tested binocularly, first exhibited signs of vision after 139 days but never acquired any useful pattern vision in 22 months of regular formal testing. However, it did pass a formal test of the ability to make luminance discriminations after 24 days. The 5 1 -o E._ 1 I 1 I 1 1 1 1 1 Period of deprivation (days) Fig. 7. The time (T) required by seventeen dark-reared animals (@) to exhibit the first signs of vision on the jumping stand (the ability to discriminate an open from a closed door), as a function of the length of the period of deprivation (t). Also shown are data from two animals that were binocularly deprived by eyelid suture (). The data for the darkreared animals have been fitted by the function: T = -56 t19"9. The seventeen dark-reared animals include all eight cats from Table 2, and nine reared for previously published studies (Timney et al. 1978; Kaye et al. 1982). other animal (C71) which was binocularly deprived to 183 days of age, and subsequently had only the left eye opened, first showed signs of vision after 163 days and eventually recovered an acuity of 1-75 cycles/deg. Both animals took approximately 4 times longer to pass a formal test of vision than did the animals that were dark-reared for equivalent periods (Fig. 7). The latter two animals differed from dark-reared animals in at least two other respects. Whereas the pupils of the latter animals remained obviously constricted for some time (as long as 4 weeks for the two animals that were deprived for a year) after the animals were introduced to illuminated surroundings, the pupils of the lid-sutured animals appeared of normal size from the time the lids were first opened. In addition, one of the latter animals 1

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION 653 (C79) exhibited a very pronounced pendular nystagmus throughout the recovery period. In addition to recovering faster, dark-reared animals acquired better vision than did animals that were monocularly deprived for equivalent periods. This point is 1- _s~ 5-- )~~~~~~~~~~~N > ss* _ 1* N.W 5 >~~~~~~~~~~~~ C N -1 L ~~~~~ 2 4 Period of deprivation (days) Fig. 8. The acuity that was eventually attained by the dark-reared cats of Fig. 8 as a function of the length of deprivation. The continuous line represents an exponential fit to the data for the dark-reared animals. The dashed line is the exponential fit to the data from the monocularly deprived animals of Fig. 2 that were reverse occluded during recovery. The open symbols show results from the two animals that had been deprived by binocular lid suture. One of these animals never acquired better vision than the ability to make a simple luminance discrimination on the jumping stand (L). made evident in Fig. 8 in which the final acuity of all the binocularly deprived animals are displayed. These data have been fitted by the exponential: S(t) = 8-73 exp (--378 t), where S(t) is the spatial resolution that was finally achieved, and t the length of deprivation (in days). The dashed line is the best-fitting exponential to the data from monocularly deprived animals that were reverse occluded throughout recovery. Limited data from two animals that were binocularly deprived by eyelid suture (open symbols) suggest that this form of deprivation may produce effects as severe as those induced by monocular deprivation. DISCUSSION Studies that documented the behavioural consequences of early monocular deprivation in cats followed close on the heels of the original descriptions of the physiological effects of this form of deprivation on the visual cortex (Ganz & Fitch, 1968; Dews & Wiesel, 197; Rizzolatti & Tradardi, 1971; Chow & Stewart, 1972;

654 D. E. MITCHELL Ganz, Hirsch & Tieman, 1972; Ganz & Hafner, 1974). However, for the most part these early studies examined only the ability of deprived animals to make luminance discriminations, or to solve pattern discrimination tasks of various levels of complexity with their deprived eye. Little attention was paid in early investigations to the effects of deprivation on such basic aspects of vision as spatial or temporal resolution. By contrast, documentation of the nature of the deficits in visual acuity and contrast sensitivity have formed the major focus of most studies of the behavioural consequences of monocular deprivation in the last 1 years (Giffin & Mitchell, 1978; Wilkinson, 198; Lehmkuhle, Kratz & Sherman, 1982; Harwerth, Smith, Boltz, Crawford & Von Noorden, 1983; Smith & Holdefer, 1985). With the exception of an earlier study from this laboratory (Giffin & Mitchell, 1978), most of these studies examined the consequences of only a limited range of deprivation conditions. The results reported here represent an attempt to explore in a systematic fashion -the consequences of periods of deprivation imposed from near birth for durations that vary in length from 6 weeks to as long as 18 months. The present study differs from earlier investigations in another important respect. By virtue of the behavioural technique employed here it was possible to not only define the longterm consequences of different periods of deprivation, but also it enabled documentation of the time course of the visual recovery. The extent of recovery of visual acuity by the deprived eye declined with increasing length of deprivation in a regular manner that, to a first approximation, could be described by a simple exponential decay that had a time constant (the length of deprivation necessary to reduce acuity by a factor of 1/e) of 17 days in the situation of binocular recovery, and 116 days for animals that were reverse occluded. Only one of eight animals deprived for 8 months or longer recovered acuity in its deprived eye in excess of 1D cycles/deg, while four animals never passed formal tests of pattern discrimination on the jumping stand (Fig. 2). A similar finding was obtained by Smith & Holdefer (1985) from two animals (see Fig. 2) that were deprived to approximately 1 year of age and tested with a similar procedure to that employed here. One of these (MD-BE-1) that subsequently had both eyes open, recovered an acuity in its deprived eye of -3 cycles/deg, while the other animal (RS-3), that was reverse occluded during recovery, achieved slightly better acuity of -8 cycles/deg. Other groups that have examined the consequences for vision of long periods of monocular occlusion have employed quite different behavioural testing procedures. Despite these procedural differences, the findings of these studies are quite comparable to those shown in Fig. 2. Berkley (1981) measured the acuity of the deprived eye of four cats that were monocularly deprived until they were about 1 year old and then allowed binocular visual input for the next 19 months. During this period only two of the cats recovered measurable acuity in their deprived eye. One cat (cat Squint) attained an acuity of about 5 cycles/deg, while the other (cat Hannibal) only achieved minimal acuity of 41 cycles/deg. All four animals were then reverse occluded for 1-3 months during which time the acuity of cat Squint improved to 9 cycles/deg and that of cat Hannibal to about -6 cycles/deg. One of the two cats that did not recover measurable acuity during binocular recovery barely achieved minimal acuity of 41 cycles/deg during the subsequent short period of reverse

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION 655 occlusion. Lehmkuhle et al. (1982) and Kratz & Lehmkuhle (1983) measured contrast sensitivity functions for both eyes of two cats that were monocularly deprived to either 8 or 16 months of age and then allowed binocular visual input. Despite the twofold difference in the length of deprivation between the two animals, the contrast sensitivity functions for the deprived eye of each animal were remarkably similar. The visual acuity of the deprived eye for gratings of 6 contrast was about 1. cycles/deg for both animals. Finally, Loop, Smyly, Millican & Greifer (1981) reported acuities of between 3 and 1P cycles/deg for the deprived eye of four cats that had been monocularly deprived to 19-5 months of age and subsequently reverse occluded. The acuity of the deprived eye was found to decrease substantially as the dimensions of the stimuli were reduced; so much so that when this eye was tested with stimuli having the same dimensions as those employed for testing the vision of the non-deprived eye, the deprived eye appeared pattern-blind. The findings from two animals reported by Berkley (1981), as well as two of three animals reported here (C35, C49, but not C127), indicate that some animals that have both eyes open following early monocular deprivation can exhibit some further recovery of vision in their deprived eye if they are then reverse occluded for a short period of time. The fact that this is not a universal finding, together with the lack of any further improvement in one animal (C127 of Fig. 1) that had already recovered good acuity during the period of binocular recovery, raises the possibility that some animals may simply learn to interpret better the poor pattern vision of the deprived eye when forced to use it exclusively. Alternatively, this improvement during reverse occlusion may reflect an increase in the number or efficacy of neural connections with the deprived eye in the visual cortex, or less specifically, an ability of the deprived eye to gain information from neurones or structures that cannot be accessed, even for very brief periods, during binocular recovery. The time required for animals to first exhibit signs of vision with their deprived eye increased with length of deprivation in a manner that could be described by a simple power law. Nevertheless once some vision was acquired, the subsequent recovery to the eventual best acuity appeared remarkably similar in all animals irrespective of the length of deprivation. This point is made evident in comparisons of the recovery of the animals of Figs 1 and 5. No matter whether deprivation extended to only 42 days (C282, Fig. 1) or 1 year of age (C35, Fig. 5), the visual acuity of the deprived eye attained its final value about 2 or 3 months from the time that it first exhibited signs of pattern vision. Where the period of deprivation was long (e.g. Al of Fig. 1), the comparatively rapid recovery of pattern vision was preceded by a substantial period of time during which the animal may only have been capable of the ability to make formal luminance discriminations. Site of recovery In view of demonstrations of substantial recovery from the physiological effects of monocular deprivation in the visual cortex (Blakemore & Van Sluyters, 1974; Mitchell et al. 1977a), it could be argued quite reasonably that the visual recovery observed in Fig. 2 may simply be a reflection of these physiological events. However, this is by no means the only possible explanation for the recovery. For example, it is known that cats with extensive lesions of cortical areas 17 and 18 manifest grating

656 D. E. MITCHELL acuities of at least 2 cycles/deg (Berkley & Sprague, 1979; Kaye, Mitchell & Cynader, 1981; Lehmkuhle et al. 1982). The fact that monocular deprivation produces visual deficits in the deprived eye that are far more profound implies that either the animal cannot utilize with this eye those structures that mediate good vision in the absence of areas 17 and 18 when these areas are physically present, or else that these structures are themselves strongly influenced by monocular deprivation. Moreover, the severity of the permanent visual deficits implies that the effects of monocular deprivation on these structures must also be long-lasting. Although it is possible that the behavioural recovery described here reflects partial recovery from the physiological effects of deprivation on structures other than areas 17 and 18, an experiment of Spear & Ganz (1975) suggests that this is not the case. They found that various pattern discriminations recovered after monocular deprivation to 4 months of age were lost following lesions of the primary visual cortical areas. Sensitive periods The data of Fig. 2 indicate that some animals retain the capacity to recover measurable visual acuity with their deprived eye after periods of monocular deprivation that extend to a year of age or even longer. A similar conclusion can be drawn from earlier studies, including some that employed different methods and responses from the animals which may have placed less demands on visuomotor coordination than was required for the execution of visually guided jumps on a jumping stand (Berkley, 1981; Lehmkuhle et al. 1982; Smith & Holdefer, 1985). Parenthetically, the visual demands of the latter response may have contributed to the differences that were observed between animals in the extent of recovery from long periods of deprivation. Some animals may have been less able or motivated than others to employ rudimentary vision to search for, and jump towards, the correct stimulus on the jumping stand. Nevertheless, the fact that visual recovery was observed in four of eight (Fig. 2) cats that were deprived to 1 year of age or longer, suggests that some plasticity must remain in the visual system for at least this long. Although it is by no means certain that the visual recovery observed after monocular deprivation is mediated by events in area 17, recent electrophysiological experiments indicate that the sensitive period defined by the ability to alter the distribution of ocular dominance in this area extends far longer than originally thought (Hubel & Wiesel, 197) to between 8 and 1 months of age (Cynader et al. 198; Jones et al. 1984). If neural plasticity can be made evident at these late ages by use of such comparatively crude measures, then it is not unreasonable to assume that other more sensitive physiological measures that probe, in a quantitative fashion, the nature of the binocular interactions of individual cells, may reveal residual plasticity even later in life. Binocular deprivation In agreement with earlier findings (Timney et al. 1978; Timney, Mitchell & Cynader, 198; Mower, Caplan & Letsou, 1982), the extent of visual recovery from dark-rearing was greater than that which occurred after equivalent periods of monocular deprivation. Such a result fits well with expectations based upon the very

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION 657 different physiological consequences of these two forms of deprivation (Wiesel & Hubel, 1965). Since the original description of the physiological effects of binocular deprivation, the more substantial sequelae of monocular deprivation have been attributed to a process of binocular competition. According to this notion, geniculate afferents from the non-occluded eye of monocularly deprived animals gain control of postsynaptic space on cells in the visual cortex (at the expense of afferents from the deprived eye) by virtue of the competitive advantage that this eye holds during development. Despite the absence of binocular competition, the physiological effects of binocular deprivation on the visual cortex can be severe. There is general consensus that a high proportion of cortical cells respond poorly or even become unresponsive to visual stimuli, and lack the selectivity for orientation and direction of motion that characterize normal cells (reviewed in Movshon & Van Sluyters, 1981; Mitchell & Timney, 1984). It is also apparent that the physiological effects of dark-rearing and binocular lid suture are not equivalent; the latter procedure produces a much higher proportion of visually unresponsive and abnormally responsive cortical cells than observed in dark-reared animals, as well as a reduced proportion of binocular cells (Mower, Berry, Burchfiel & Duffy, 1981). In addition, the extent of recovery from the physiological effects of these two forms of binocular deprivation are very different. Whereas a high proportion of cortical cells of dark-reared animals recover their stimulus specificity (such as orientation selectivity), the cortical recovery in binocular lid-sutured animals is far less dramatic (Cynader, Berman & Hein, 1976; Cynader & Mitchell, 198; Mower et al. 1981). The differences between the consequences of dark-rearing and binocular lid suture have been attributed by Mower et al. (1981) to the fact that the former represents complete deprivation to all visual stimulation, while the latter permits diffuse visual stimulation by virtue of the fact that the eyelids are not opaque (Crawford & Marc, 1976), a point that has also been demonstrated by behavioural (Loop & Sherman, 1977) and physiological measures (Spear, Tong & Langsetmo, 1978). Presumably the reduced visual recovery that occurs after binocular lid suture results from specification of many cortical cells by the low spatial frequency form information transmitted through the eyelids. On the other hand, the absence of visual stimulation during dark-rearing is known to leave the cortex in a state that can still respond to its first visual input, even when it occurs at 1 months of age (Cynader & Mitchell, 198).While it is tempting to attribute the recovery of vision after the two forms of binocular deprivation to the physiological recovery observed in the cortex following introduction of normal visual input, it is important to recognize that these rearing procedures may also lead to deficits in the control of involuntary eye movements which may place a ceiling on the level of vision that can be recovered. Intermittent spontaneous nystagmus has been demonstrated in cats that were dark-reared to at least 11 months of age (Harris & Cynader, 1981), and one of the two lid-sutured animals of the present study (C79) exhibited a pendular nystagmus that was very obvious to casual inspection. The presence of such oculomotor disturbances may account for the greater severity of the deficits observed in C71 and C79 than those reported in an earlier study (Smith, Lorber, Stanford & Loop, 198).

658 D. E. MITCHELL Clinical implications Particularly when the period of monocular deprivation was short, the recovery of vision in the deprived eye that occurred with reverse occlusion was greater than that which occurred with binocular recovery (Fig. 2). Nevertheless, the visual recovery promoted by reverse occlusion was never complete, and moreover, in subsequent work it has been shown to be only temporary; much of the improvement in the vision of the deprived eye is lost following termination of reverse occlusion (Mitchell et al. 1984; Murphy & Mitchell, 1987). Because reverse occlusion closely resembles fulltime occlusion therapy for human amblyopia, an additional study was initiated recently to explore the effectiveness of various alternative recovery conditions in an effort to find a procedure with a better final outcome (Mitchell et al. 1986). Interestingly, occlusion procedures that effectively combine on a daily basis the two procedures studied here in just the right proportions, can lead to complete and permanent recovery of vision in the deprived eye of animals monocularly deprived to 6 weeks of age. Such an outcome is observed only with certain restricted regimens of part-time occlusion of the non-deprived eye in which this eye is occluded for the first part of each day (for between 5 and 7% of the time) followed by binocular visual exposure for the remainder. It remains to be seen whether in animals deprived for much longer periods any regimen of part-time occlusion can promote a greater degree of recovery than was observed here (Fig. 2) in animals that received either full-time occlusion (reverse occlusion), or no occlusion (binocular recovery) following extended periods of early monocular deprivation. This work was supported by grants from the Natural Sciences and Engineering Council of Canada (A 766), and the Medical Research Council of Canada (MA-7732 and PG-29). Some of the binocularly deprived animals described here were involved in studies performed in collaboration with Drs B. Timney and M. Kaye. Jane Anne Horne provided invaluable mathematical and statistical advice. I wish to thank Cindy Trask and Heather Dzioba for their dedicated care of the animals. REFERENCES BERKLEY, M. A. (1981). Animal models of visual development: behavioral evaluation of some physiological findings in cat visual development. In Development ofperception, vol. 2, The Visual System, ed. ASLIN, R. N., ALBERTS, J. R. & PETERSON, M. R., pp. 197-215. New York: Academic Press. BERKLEY, M. A. & SPRAGUE, J. (1979). Striate cortex and visual acuity functions in the cat. Journal of Comparative Neurology 187, 679-72. BLAKEMORE, C., GAREY, L. J. & VITAL-DURAND, F. (1978). The physiological effects of monocular deprivation and their reversal in the monkey's visual cortex. Journal of Physiology 283, 223-262. BLAKEMORE, C. & VAN SLUYTERS, R. C. (1974). Reversal of the physiological effects of monocular deprivation in kittens: further evidence for a sensitive period. Journal of Physiology 237, 195-216. BLAKEMORE, C., VITAL-DURAND, F. & GAREY, L. J. (1981). Recovery from monocular deprivation in the monkey. I. Recovery of physiological effects in the visual cortex. Proceedings of the Royal Society B 213, 399-423. CHow, K. L. & STEWART, D. L. (1972). Reversal of structural and functional effects of long-term visual deprivation in cats. Experimental Neurology 34, 49-433. CRAWFORD, M. L. J. & MARC, R. E. (1976). Light transmission of cat and monkey eyelid. Vision Research 16, 323-324.

VISUAL RECOVERY AFTER MONOCULAR DEPRIVATION CYNADER, M., BERMAN, N. & HEIN, A. (1976). Recovery of function in cat visual cortex following prolonged deprivation. Experimental Brain Research 25, 139-156. CYNADER, M. & MITCHELL, D. E. (198). Prolonged sensitivity to monocular deprivation in darkreared cats. Journal of Neurophysiology 43, 126-14. CYNADER, M., TIMNEY, B. N. & MITCHELL, D. E. (198). Period of susceptibility of kitten visual cortex to the effects of monocular deprivation extends beyond six months of age. Brain Research 191, 545-55. DEWS, P. B. & WIESEL, T. N. (197). Consequences of monocular deprivation on visual behaviour in kittens. Journal of Physiology 26, 435-455. GANZ, L. & FITCH, M. (1968). The effects of visual deprivation on perceptual behavior. Experimental Neurology 22, 638-66. GANZ, L. & HAFNER, M. E. (1974). Permanent perceptual and neurophysiological effects of visual deprivation in the cat. Experimental Brain Research 2, 67-87. GANZ, L., HIRSCH, H. V. B. & TIEMAN, S. B. (1972). The nature of perceptual deficits in visually deprived cats. Brain Research 44, 547-568. GIFFIN, F. & MITCHELL, D. E. (1978). The rate of recovery of vision after early monocular deprivation in kittens. Journal of Physiology 274, 511-537. HARRIS, L. R. & CYNADER, M. (1981). The eye movements of the dark-reared cat. Experimental Brain Research 44, 41-56. HARWERTH, R. S., SMITH, E. L., BOLTZ, R. L., CRAWFORD, M. L. J. & VON NOORDEN, G. K. (1983). Behavioral studies on the effect of abnormal early visual experience in monkeys: Spatial modulation sensitivity. Vision Research 23, 151-151. HOWELL, E. R., MITCHELL, D. E. & KEITH, C. G. (1983). Contrast thresholds for sine gratings of children with amblyopia. Investigative Ophthalmology and Visual Science 24, 782-787. HUBEL, D. M. & WIESEL, T. N. (197). The period of susceptibility to the physiological effects of unilateral eye closure in kittens. Journal of Physiology 26, 419-436. JONES, K. R., SPEAR, P. D. & TONG, L. (1984). Critical periods for effects of monocular deprivation: differences between striate and extrastriate cortex. Journal of Neuroscience 4, 2543-2552. KAYE, M., MITCHELL, D. E. & CYNADER, M. (1981). Selective loss of binocular depth perception after ablation of cat visual cortex. Nature 293, 6-62. KAYE, M., MITCHELL, D. E. & CYNADER, M. (1982). Depth perception, eye alignment and cortical ocular dominance of dark-reared cats. Developmental Brain Research 2, 37-53. KRATZ, K. E. & LEHMKUHLE, S. (1983). Spatial contrast sensitivity of monocularly deprived cats after removal of the non-deprived eye. Behavioural Brain Research 7, 261-266. LEHMKUHLE, S., KRATZ, K. E. & SHERMAN, S. M. (1982). Spatial and temporal sensitivity of normal and amblyopic cats. Journal of Neurophysiology 48, 372-387. LEVAY, S., WIESEL, T. N. & HUBEL, D. H. (198). The development of ocular dominance columns in normal and visually deprived monkeys. Journal of Comparative Neurology 191, 1-51. LEVI, D. M. & KLEIN, S. (1982). Differences in vernier discrimination for gratings between strabismic and anisometropic amblyopes. Investigative Ophthalmology and Visual Science 23, 398-47. Loop, M. S. & SHERMAN, S. M. (1977). Visual discriminations during eyelid closure in the cat. Brain Research 128, 329-339. Loop, M. S., SMYLY, E. C., MILLICAN, C. L. & GREIFER, C. F. (1981). Acuity, luminance, and monocular deprivation in the cat. Behavioural Brain Research 2, 323-334. MITCHELL, D. E., CYNADER, M. & MOVSHON, J. A. (1977 a). Recovery from the effects of monocular deprivation in kittens. Journal of Comparative Neurology 176, 53-64. MITCHELL, D. E., GIFFIN, F. & TIMNEY, B. (1977b). A behavioural technique for the rapid assessment of the visual capabilities of kittens. Perception 6, 181-193. MITCHELL, D. E. & MURPHY, K. M. (1984). The effectiveness of reverse occlusion as a means of promoting visual recovery in monocularly deprived kittens. In Development of Visual Pathways in Mammals, ed. STONE, J., DREHER, B. & RAPAPORT, D. H., pp. 381-392. New York: Alan R. Liss. MITCHELL, D. E., MURPHY, K. M., DZIOBA, H. A. & HORNE, J. A. (1986). Optimization of visual recovery from early monocular deprivation in kittens: implications for occlusion therapy in the treatment of amblyopia. Clinical Vision Sciences 1, 173-177. 659

66 D. E. MITCHELL MITCHELL, D. E., MURPHY, K. M. & KAYE, M. G. (1984). The permanence of the visual recovery that follows reverse occlusion of monocularly deprived kittens. Investigative Ophthalmology and Visual Science 25, 98-917. MITCHELL, D. E. & TIMNEY, B. (1984). Postnatal development of function in the mammalian visual system. In Handbook of Physiology, section I: The Nervous System, vol. 3, part 1, Sensory Processes, ed. DARIAN-SMITH, I., pp. 57-555. Bethesda: American Physiological Society. MOVSHON, J. A. (1976). Reversal of the physiological effects of monocular deprivation in the kitten's visual cortex. Journal of Physiology 261, 125-174. MOVSHON, J. A. & VAN SLUYTERS, R. C. (1981). Visual neuronal development. Annual Review of Psychology 32, 477-522. MOWER, G. D., BERRY, D., BURCHFIEL, J. L. & DUFFY, F. M. (1981). Comparison of the effects of dark rearing and binocular suture on development and plasticity of cat visual cortex. Brain Research 22, 255-267. MOWER, G. D., CAPLAN, C. J. & LETSOU, G. (1982). Behavioral recovery from binocular deprivation in the cat. Behavioural Brain Research 4, 29-215. MURPHY, K. M. & MITCHELL, D. E. (1987). Reduced visual acuity in both eyes of monocularly deprived kittens following a short or a long period of reverse occlusion. Journal of Neuroscience 7, 1526-1536. OLSON, C. R. & FREEMAN, R. D. (1978). Monocular deprivation and recovery during sensitive period in kittens. Journal of Neurophysiology 41, 65-74. OLSON, C. R. & FREEMAN, R. D. (198). Profile of the sensitive period for monocular deprivation in kittens. Experimental Brain Research 39, 17-21. RIZZOLATTI, G. & TRADARDI, V. (1971). Pattern discrimination in monocularly reared cats. Experimental Neurology 33, 181-194. SMITH, D. C. & HOLDEFER, R. N. (1985). Binocular competitive interaction and recovery of visual acuity in long-term monocularly deprived cats. Vision Research 25, 1783-1794. SMITH, D. C., LORBER, R., STANFORD, L. R. & LooP, M. S. (198). Visual acuity following binocular deprivation in the cat. Brain Research 183, 183-193. SPEAR, P. D. & GANZ, L. (1975). Effects of visual cortex lesions following recovery from monocular deprivation in the cat. Experimental Brain Research 23, 181-21. SPEAR, P. D., TONG, L. & LANGSETMO, A. (1978). Striate cortex neurons of binocularly deprived kittens respond to visual stimuli through the closed eyelids. Brain Research 155, 141-146. TIMNEY, B., MITCHELL, D. E. & CYNADER, M. (198). Behavioral evidence for prolonged sensitivity to effects of monocular deprivation in dark-reared cats. Journal of Neurophysiology 43, 141-154. TIMNEY, B., MITCHELL, D. E. & GIFFIN, F. (1978). The development of vision in cats after extended periods of dark-rearing. Experimental Brain Research 31, 547-56. WIESEL, T. N. & HUBEL, D. H. (1963). Single-cell responses in striate cortex of kittens deprived of vision in one eye. Journal of Neurophysiology 26, 13-117. WIESEL, T. N. & HUBEL, D. H. (1965). Comparison of the effects of unilateral and bilateral eye closure on cortical unit responses in kittens. Journal of Neurophysiology 28, 129-14. WILKINSON, F. (198). Reversal of the behavioral effects of monocular deprivation as a function of age in the kitten. Behavioural Brain Research 1, 11-123.