Binocular Exposure causes Suppression of the Less Experienced Eye in Cats Previously Reared with Unequal Alternating Monocular Exposure

Transcription

1 Binocular Exposure causes Suppression of the Less Experienced Eye in Cats Previously Reared with Unequal Alternating Monocular Exposure Nino Tumosa,* Stacy Nunberg, Helmut V. B. Hirsch, and Suzannah Bliss Tieman In unequal alternating monocular exposure (unequal AME), each eye receives normal patterned visual input but on alternate days and for unequal periods. It has been shown previously that this imbalance in stimulation produces a deficit in the nasal visual field of the less experienced eye (LEE). The effect of subsequent binocular exposure on these visual field deficits has now been examined. No evidence of recovery was found. Instead, visual fields remained the same or became smaller. In cats reared with little or no imbalance (8 hr/day vs 7 hr/day or 1 hr/day vs 1 hr/day), subsequent binocular exposure had no effect on visual fields. In cats reared with a moderate or large imbalance (8 hr/day vs 4 hr/day or 8 hr/day vs 1 hr/day), subsequent binocular exposure led to a suppression of the LEE: when tested binocularly, these cats rarely responded to targets presented in the monocular field of the LEE. The deficits became progressively more severe throughout the period of binocular exposure, until eventually they could be observed even when the LEE was tested monocularly. Most of these cats were clearly esotropic but not all esotropic cats showed suppression. The degree of suppression was correlated with the degree of the imbalance imposed during unequal AME. Our results suggest that when the eyes are misaligned, binocular exposure does not permit recovery of visual function in a disadvantaged eye, but may exacerbate the existing imbalance. Invest Ophthalmol Vis Sci 24: , 1983 The two eyes have slightly different views of the world that are normally integrated centrally to give a single, three-dimensional representation of the world. This integration cannot be achieved successfully if the eyes are misaligned, as in strabismus, and disturbances of binocular visual perception such as diplopia and visual confusion are likely to result. 1 Attempts to cope with these disturbances can take one of two forms: (1) alternatingfixation usingfirst one eye and then the other, or (2) uniocular fixation using one eye to the exclusion of the other. In the latter case, the eye that is not used becomes amblyopic. 1 Although the most common form of amblyopia in humans is that associated with strabismus, most animal studies of amblyopia have used monocularly deprived (MD) rather than squinting animals, in part From the Neurobiology Research Center, State University of New York at Albany, Albany, New York. * Current address: Department of Anatomy, University of Calgary, Calgary, Alberta, Canada. Supported by NIH Grants EY02609 and EY Submitted for publication: May 4, Reprint requests: Suzannah Bliss Tieman, PhD, Neurobiology Research Center, State University of New York at Albany, Albany, NY because monocular deprivation reliably causes profound amblyopia, 2 " 6 whereas strabismus, as it has been produced in animals, causes ambylopia in only some cases, primarily esotropes. 7 " 10 Further, although amblyopia in humans is thought to be caused by suppression of one eye, there is little or no evidence for comparable suppression in strabismic cats. Thus, the loss of acuity in one eye of some strabismic animals has been hypothesized to result from suppression of that eye.'' However, other explanations are possible, 12 and the suppression hypothesis has received little supporting evidence that the animal does, in fact, suppress the input from the eye with reduced acuity. In this paper we report that we routinely observe suppression of one eye in cats reared with unequal alternating monocular exposure (unequal AME) and later given binocular exposure. We have previously shown that rearing cats with unequal AME places the less experienced eye (LEE) at a disadvantage, and that the monocular visual field of the LEE is restricted to the temporal hemifield We now report that unequal AME cats are strabismic, that their binocular visual fields are restricted, and that both their binocular and monocular visual fields become further restricted with time after onset of binocular exposure /83/0400/496/$ 1.35 Association for Research in Vision and Ophthalmology 496

2 No. 4 BINOCULAR SUPPRESSION IN UNEQUAL AME CATS / Tumoso er ol. 497 Furthermore, the animals respond to fewer targets in the peripheral field ipsilateral to the LEE with both eyes open than they do with the LEE alone, suggesting that with both eyes open the animals suppress input from the LEE. The degree of this suppression is correlated with the extent of the imbalance in stimulation of the two eyes imposed during rearing. Subjects Materials and Methods A total of nine cats, all of which were reared with alternating monocular exposure (AME), were used in this study. One cat was reared with equal AME, and eight cats were reared with unequal AME. All subjects were born in the closed breeding colony maintained by the Department of Biological Sciences at SUNYA. Rearing All of the kittens were placed, with their mothers, into a totally dark room before their eyes had opened. Beginning at days of age, the kittens were brought out into the light for daily periods of monocular exposure: one eye was occluded with an opaque scleral contact occluder that was placed in the eye before the animal was brought out into the light. For all animals, the eye that was exposed alternated on successive days. Equal AME: One animal was exposed for equal periods with each eye. This animal was exposed for 1 hr/day (AME 1/1). Unequal AME: Eight animals were exposed for unequal periods with each eye. For two cats from one litter, the imbalance in stimulation was slight (8 hr/ day vs 7 hr/day; AME 8/7). The left eye was exposed for 8 hr and, on alternate days, the right eye for 7 hr. For four cats from two litters, the imbalance was moderate (8 hr/day vs 4 hr/day; AME 8/4). The left eye was exposed for 8 hr and the right eye was exposed for 4 hr. For two cats from one litter, the imbalance was large (8 hr/day vs 1 hr/day; AME 8/1). The right eye was exposed for 8 hr and the left eye for 1 hr. Quantification of the imbalance: We quantified the degree of the imbalance in the following manner. We took the difference in the duration of stimulation of the two eyes and divided it by the longer of the two durations. That is, I = (A B)/A, where A is the longer duration and B is the shorter. Thus, for AME 1/ 1, which imposes no imbalance, 1 = 0; for AME 8/7, which imposes a slight imbalance, I = 0.125; for AME 8/4, which imposes a moderate imbalance, I = 0.50; and for AME 8/1, which imposes a large imbalance, I = Binocular Exposure Late-exposed group: Two of the AME 8/4 cats (#8 and #9) received 9 months of unequal AME. They then received 2 weeks of binocular exposure followed by 3 weeks of alternating binocular, monocular exposure: to give the 4-hr eye an advantage, the 8-hr eye was occluded every other day for 24 hr/day. Thereafter, they received binocular exposure except when being tested monocularly. These animals were the pilot subjects in a study of recovery from the effects of unequal AME. After we finished testing these cats, we redesigned the experiment to study systematically the exacerbation of visual field deficits in unequal AME cats given subsequent binocular exposure. Early-exposed group: All of the other AME cats (the AME 1/1 cat, both AME 8/7 cats, AME 8/4 #10 and #12, and both AME 8/1 cats) received 12 to 16 weeks of unequal AME. They were then brought out of the dark into the continuously lighted animal facility where they remained for the rest of the experiment. From this point on they received binocular exposure. Contact occluders were placed in the animals' eyes only when the animals were being tested monocularly. Visual Field Perimetry Just before the onset of binocular exposure and at selected intervals thereafter, the animals were tested for their ability to orient to targets in visual space. 13 " 16 A complete description of our testing procedure appears elsewhere. 14 Briefly, the animals were encouraged to fixate a target (a piece of beef kidney on a wire) presented straight ahead at a distance of 40 cm. While the animal was fixating this target, a novel stimulus (another piece of beef kidney on another wire) was introduced at a distance of 20 cm along one of the guidelines that were placed every 15 to the left or right of the fixation line (0 ). A positive response was recorded when, upon being released, the cat turned and immediately approached the novel stimulus. A negative response was recorded if the cat approached the fixation object or if it scanned the field before approaching the novel stimulus. If the animal turned toward the novel stimulus but then turned back toward and approached the fixation stimulus, the response was recorded as plus/ minus since the animal apparently saw the novel stimulus but chose not to respond normally to it. The novel stimulus was presented at each of 15 positions from 105 left to 105 right except at 0. For the trials at 0, only the fixation object was presented: On these trials, failure to approach directly the fixation object was scored as a negative response. Each

3 498 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1983 Vol. 24 AME 1/1 # w**k«left BINOC RIGHT Fig. 1. Visual fields of an AME 1/1 cat at different times after the onset of binocular exposure. Above each set of graphs are given the age of the animal at the onset of binocular exposure and the number of weeks since that time. The visualfieldsare represented by polar plots that show the number of responses to stimuli presented at every 15 of the visual field. The part-circle represents a level of 12 positive responses in 12 trials. These visual fields are essentially normal and do not change with time. animal was tested binocularly and monocularly with each eye 12 times at each position. Within each block of 15 trials, the order in which the different positions were tested was determined by a table of random permutations. Before the onset of binocular exposure, all animals were tested monocularly with each eye on separate days. After the onset of binocular exposure, each animal was tested binocularly and then monocularly with each eye. Which eye was tested first was determined randomly. Interocular Alignment After the animals had received at least 2 months of binocular exposure, we photographed the eyes of each animal to determine the interocular alignment. 17 For comparison, we also measured the interocular alignment in ten normally reared cats. The cats faced a photoflood lamp that was positioned at least 4 m away. They were photographed using a 35 mm camera with a 50 mm macro lens onto Tri-X film at ASA 400. The film plane was approximately 60 cm from the cat. We photographed each cat several times. We printed the photographs at 1.3 to 1.4 times life size and then measured (1) the distance between the corneal reflections of the floodlamp and (2) the interpupillary distance. We then calculated the ratio of the distance between the reflexes to the interpupillary distance,* a variable that is comparable across cats regardless of magnification differences. 23 ' 24 The vergence ratios given in the results are averages of ratios determined from two to four photographs. Results Visual Field Perimetry In general, the monocular visual fields at the beginning of binocular exposure resembled those reported previously The visual field of each eye of the equal AME cat was normal, as was the visual field of the more experienced eye (MEE) of each unequal AME cat, whereas the field of the less experienced eye (LEE) was restricted: All unequal AME cats showed a nasal field deficit in the LEE. For animals reared with little or no imbalance, the binocular field was easily predicted from the two monocular fields, and the visual fields remained unchanged with binocular exposure. However, for most of the AME 8/ 4 cats and AME 8/1 cats the binocular field was less extensive than would be predicted from the monocularfields,and further, both the binocular field and the monocular field of the LEE became more restricted with time. AME 1/1 cat: The AME 1/1 cat showed essentially normal fields (Fig. 1). The visual field of each eye extended 120, from 90 temporal to 30 nasal, and the binocular field extended 180. This pattern of results remained unchanged throughout the period of binocular exposure. * This method measures the alignment of the pupillary axes, from which the alignment of the visual axes must be inferred. Variability in the relationship between visual axis and pupillary axis 18 thus limits the accuracy with which we can determine interocular alignment. However, we could not use Olson and Freeman's 18 more accurate method of inferring interocular alignment in awake cats, which involves recording from binocular units in Area 17, 18 for two reasons. First, there are very few binocular cells in Area 17 of AME cats Second, the correspondence between the receptive fields in the left and right eye may be anomalous in our AME cats, as it has been reported to be in some cats with monocular paralysis 21 or strabismus. 22

4 No. 4 BINOCULAR SUPPRESSION IN UNEQUAL AME CATS / Tumoso er ol. 499 AME 8/7 #1 AME 8/7 #2 ag*: 12+0 we«ks ag«: 12+0 w««k«fig. 2. Visual fields for two AME 8/7 cats at different times after the onset of binocular exposure. In each case, the field of the MEE is normal in extent while that of the LEE is restricted. Each cat made fewer responses at 30 and 15 nasal with its 7-hr eye than with its 8-hr eye. By week 3, neither cat responded to any target in the nasalfieldof its LEE. Although in general, the binocular fields can be predicted from the monocular fields, AME 8/7 #1 at first showed some tendency to suppress the LEE. This tendency disappeared with time. MEE BINOC LEE MEE BINOC LEE AME 8/7 cats: The visual fields of the two AME 8/7 cats are shown in Fig. 2. At the start of binocular exposure, when the animals were 12 weeks old, they showed a normal visual field in the 8-hr eye and a nasal deficit in the LEE. This nasal field deficit was less severe than it was in the AME 8/4 cats and the AME 8/1 cats (Figs. 4-6), but it was more severe than we reported previously for AME 8/7 cats tested at 8 weeks postnatal. 14 By the third week of binocular exposure, 15 weeks postnatal, our AME 8/7 cats resembled AME 8/4 and AME 8/1 cats in that they never responded to targets in the nasal field of the LEE. To determine whether this aggravation of the nasal field loss was a function of the binocular exposure or of the animal's age, we reared an additional three AME 8/7 cats that received continued unequal AME until they were 14 weeks of age. We tested their visual fields repeatedly from 8 wks of age, and we plotted the number of responses to targets presented in the nasal field of the LEE (Fig. 3). Note that by 14 weeks of age, all three kittens ceased responding to targets in the nasal field. Thus, the development of the complete nasal field deficit that we obseved in our AME 8/7 cats was not a function of the binocular experience. The binocular visual field of AME 8/7 #2 was normal at the start of binocular exposure and remained normal. The binocular visual field of AME 8/7 #1 at first was slightly restricted on the side of the LEE, but recovered with continued exposure. Thus, for the o AME 8/7 cats binocular exposure had little effect on the visual fields. AME 8/4 cats (early group): The visual fields of these AME 8/4 cats are shown in Figure 4. The mon or feio cr LU GD 10 AGE IN WEEKS Fig. 3. Loss of responsiveness to targets in the nasal field of the LEE of AME 8/7 cats. The number of responses to targets presented 15 and 30 nasal to the LEE are plotted as a function of age for five AME 8/7 cats. The black circles represent data for cats given binocular exposure beginning at 12 weeks of age. The open circles represent data for cats given only AME throughout. It can be seen that whether or not it received any binocular exposure, each animal ceased responding to targets in the nasal field of the 7-hr eye by the age of 14 or 15 weeks. 16

5 500 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1983 Vol. 24 AME 8/4 #10 ag*: 12+0 w««ka AME 8/4 #12 ag«: 12+0 w*«k«fig. 4. Visual fields for two AME 8/4 cats (early group). The solid lines represent positive responses to stimuli at each position; the dotted lines represent plus/ minus responses. In each case, the field of the MEE is normal in extent, whereas that of the LEE is restricted to the temporal hemifield. The binocularfieldsare usually smaller than what would be predicted from the monocular fields, as if the cat were suppressing the input from the LEE. Eventually (after 40 weeks of binocular exposure), the effects of this suppression can be seen even when the animal is tested monocularly. MEE BINOC LEE MEE BINOC LEE ocular visual fields were characteristic of AME 8/4 cats 14 and remained so throughout the first 16 weeks of binocular exposure. However, except for the initial data of AME 8/4 #10, the binocular visualfieldswere smaller than would be predicted from the monocular visual fields. They were restricted on the side of the LEE. The animals acted as if they could not see the targets with the LEE when both eyes were open, although clearly they could see such targets with the LEE when the MEE was occluded (Fig. 4, right column). The animals thus appeared to be suppressing the input from the LEE. This binocular visual field deficit did not recover with continued binocular exposure; if anything, it got worse. After 40 weeks of binocular exposure, the effects of the suppression could be seen even when the animals were tested monocularly: there was an added deficit at the temporal margin of the field of the LEE. Thus, in the early exposed group of AME 8/4 cats, binocular exposure resulted in a partial restriction of the binocular visual field and, eventually, a temporal restriction of the monocular field of the LEE. AME 8/4 cats (late group): The visualfieldsof these AME 8/4 cats are shown in Figure 5. Again, at the beginning of binocular exposure, the visualfieldswere characteristic of AME 8/4 cats. For AME 8/4 #9, they remained so, and the binocular field could be predicted from the monocular fields. For AME 8/4 #8, however, the binocular field was always restricted on the side of the LEE. Further, the monocular field of the LEE became restricted at its temporal margin and did so sooner than it had for the early group of AME 8/4 cats (10 and 16 weeks of binocular exposure vs 40 weeks of binocular exposure). Thus, one of the late exposed group of AME 8/4 cats showed strong suppression both binocularly and monocularly; the other cat showed none. AME 8/1 cats: The visual fields of the AME 8/1 cats are shown in Figure 6. For ease of comparison, they have been plotted in mirror-image fashion, as if the right eye were the LEE. The initial monocular fields were characteristic of AME 8/1 cats However, with binocular visual experience the monocular and binocular visual fields rapidly became more and more severely restricted. After 26 weeks of binocular exposure, the binocular field was equivalent to the monocular field of the MEE, suggesting that neither animal was using its LEE at all. Even when the animal was tested with the LEE alone, the effects of the suppression were very clear: the field of the LEE was

6 No. 4 BINOCULAR SUPPRESSION IN UNEQUAL AME CATS / Tumoso er ol. 501 Fig. 5. Visual fields for two AME 8/4 cats (late group) These animals were not tested binocularly at 0 weeks or monocularly at 5 and 7 weeks. The solid lines represent positive responses to stimuli at each position; the dotted lines represent plus/minus responses. In each case, the field of the MEE is normal whereas the field of the LEE is restricted to the temporal hemifield. The fields of AME 8/4 #9 do not change with time, and the binocular field can be predicted from the monocular fields. For AME 8/4 #8, however, both the binocular field and the monocular field become restricted with time. AME 8/4 #8 ago: 38+0 weeks AME 8/4 #9 a O «: 38+0 u««kc MEE BINOC LEE MEE BINOC LEE AME 8/1 #16 age: 12+0 weeks AME 8/1 #17 age: 12+0 weeks Fig. 6. Visual fields for two AME 8/1 cats. For ease of comparison, the visual fields have been plotted in mirror-image fashion, as if the right eye were the LEE, although, in fact, it was the left eye that was the LEE. In each case, the field of the MEE is normal in extent, whereas that of the LEE is restricted to the temporal hemifield. Both the binocularfieldand thefieldof the LEE become more restricted with time until eventually, the binocular field is equivalent to the monocularfieldof the MEE, and the field of the LEE is restricted to the central 30 of the temporal hemifield. MEE BINOC LEE MEE BINOC LEE

7 502 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / April 1983 Vol. 24 TESTED BINOCULARLY CO LU en u_ o en Id CD TESTED MONOCULARLY x x AME I/I AME 8/7 O O AME 8/4 A A AME 8/I 1 I I I I WEEKS OF BINOCULAR EXPOSURE 40 Fig. 7. Suppression as a function of binocular exposure. The number of responses to targets presented 45, 60, 75, and 90 ipsilateral to the LEE is plotted on the ordinate (0 is straight ahead). Plus/minus responses are given halfweight. For the purpose of this and subsequent analyses, we arbitrarily treated the right eye of the AME 1/ 1 cat as its LEE. In general, the number of responses decreased with time. This trend is clearer for the data obtained under monocular testing conditions (lower graph) than for those obtained under binocular testing conditions (upper graph). restricted to the central 30 of the temporal hemifield. Thus, for the AME 8/1 cats, binocular exposure created an almost complete suppression of the LEE. Time Course The performance of our animals deteriorated with time after the onset of binocular exposure. This is shown in Figure 7. For each animal, this figure plots as a function of weeks of binocular exposure the number of correct responses to targets presented at 45, 60, 75, and 90 ipsilateral to the LEE. Data obtained under conditions of binocular testing are plotted in the upper graph and data obtained under conditions of monocular testing are plotted in the lower graph. It can be seen that the performance with both eyes open was generally worse and more variable than the performance with the MEE occluded. Further, although the performance of the AME 1/1 cat and the AME 8/7 cats remained stable, the performance of the AME 8/1 cats and all but one of the AME 8/ 4 cats deteriorated with time. This trend was especially clear in the monocular data, where the curves for the two AME 8/1 cats and one of the late group of AME 8/4 cats are nearly superimposed. As a control to show that the increasingly poorer performance was a function of the length of the binocular exposure rather than the age of the animal, we

8 No. 4 BINOCULAR SUPPRESSION IN UNEQUAL AME CATS / Tumosa er ol. 503 tested three AME 8/1 cats that had never received binocular exposure, but had continued to receive unequal AME. We tested these animals monocularly at 24 weeks of age and compared their data to the data that we obtained from the other two AME 8/1 cats after 10 weeks of binocular exposure, when they were 22 weeks old. We found that the control AME 8/1 cats responded significantly more often to targets presented at 45, 60, 75, and 90 ipsilateral to the LEE than did the binocularly exposed AME 8/1 cats (t(3) = 4.37, p < 0.025, two-tailed). Further, when we compared the responsiveness to temporal targets with the 8-hr eye and with the 1-hr eye, we found that the difference in performance between the 8-hr eye and the 1-hr eye was much greater for the binocularly exposed AME 8/1 cats than for the control AME 8/ 1 cats (t(3) = p < 0.001, two-tailed). Thus 24 weeks of unequal AME resulted in better performance of the LEE than did 12 weeks of unequal AME followed by 10 weeks of unrestricted binocular exposure. Imbalance and Suppression In general, the greater the imbalance imposed during rearing, the greater the suppression. This is illustrated in Figure 8 where we show the extent to which the imbalance imposed during rearing predicts the number of responses to targets presented from 45 to 90 temporal on the side of the LEE. The values plotted are for binocular data obtained after 16 weeks of binocular exposure. Clearly, the greater the imbalance, the poorer the performance (r = 0.81, p < 0.01). The rearing of our cats varied along several parameters, other than imbalance, which might be correlated with suppression. First, the animals differed in the total amount of exposure received during rearing: from 2 hrs per 2 days (AME 1/1) to 15 hrs per 2 days (AME 8/7). However, neither the animal that received the least total exposure nor the two animals that received the most total exposure showed any sign of suppression, suggesting that total amount of exposure is not a critical parameter. Further, the rearing of our cats also differed with respect to age at onset of AME, age at onset of binocular exposure, and whether there was any forced used of the LEE. It was the two late-exposed cats, AME 8/4 #8 and #9, that were the most aberrant with respect to each of these three variables. To some extent then, we can evaluate the contribution of these variables to the relationship between suppression and imbalance by eliminating the data from these two animals (which are plotted in Fig. 8 as asterisks). When we do this, the correlation of suppression with Id if) CO L«J o CD 10 0, IMBALANCE 1.0 Fig. 8. Suppression as a function of imbalance. The number of responses to targets presented 45, 60, 75, and 90 ipsilateral to the LEE are plotted as a function of the degree of imbalance imposed during rearing (0 is straight ahead). The data plotted are those obtained under binocular testing conditions after 16 weeks of binocular exposure. The data for the late-exposed cats are plotted as asterisks. The line shown is the best-fitting straight line through all nine data points. imbalance becomes even stronger (r = 0.96, p < 0.001). We thus conclude that the correlation we observe between suppression and imbalance is not produced by variations in these other variables. Interocular Alignment The vergence ratios for the nine AME animals and the ten normal control animals, also from our breeding colony, are given in Table 1. The vergence ratios of the normal cats were all less than 1.0, indicating that their pupillary axes diverged, as has been reported before In contrast, the vergence ratios of the AME cats centered around 1.0, indicating that Table 1. Vergence ratios for ten normally reared and nine AME cats Cat N #1 N #2 N#3 N#4 N#5 N#6 N#7 N#8 N#9 N#10 Mean Normal Vergence ratio Cat AME 1/1 #8 AME 8/7 #1 AME 8/7 #2 AME 8/4 #8 AME 8/4 #9 AME 8/4 #10 AME 8/4 #12 AME 8/1 #16 AME 8/1 #17 Mean AME Vergence ratio

9 504 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1983 Vol. 24 the pupillary axes of these cats did not diverge, and thereby suggesting that their visual axes converged. Thus, most of the AME animals appeared to be esotropic. For only one animal (AME 8/4 #8) did the vergence ratio lie within the normal range ( ). For every other AME animal, the vergence ratio was greater than normal, and the mean vergence ratio for the AME cats (0.995) was significantly different from that for the normal cats (0.957, t(17) = 5.21, p< 0.001). Esotropia, as inferred from our measure, appears to be a function of the alternating exposure rather than the overall imbalance in stimulation. There was no correlation between the vergence ratio and the degree of imbalance imposed during rearing (r = 0.028, ns). Further, the degree of esotropia could not be used to predict the presence or the magnitude of suppression. The one AME cat whose vergence ratio was within the normal range (AME 8/4 #8) was a cat that suppressed (Fig. 5), whereas the four AME cats that did not suppress all had vergence ratios that were higher than normal. Also, there was little correlation between a cat's vergence ratio and the number of responses that it made to targets presented from 45 to 90 ipsilateral to the LEE (r = -0.03, ns). Discussion We have shown that binocular visual exposure results in an increase in the extent of the deficits in visual fields of cats reared with unequal alternating stimulation of the two eyes during early postnatal life. Unequal AME per se results in a loss of responsiveness to targets presented in the nasal portion of the visual field of the LEE. Subsequent binocular exposure usually leads to suppression of most or all responsiveness with the LEE, both when the animal is tested binocularly, and, eventually, when the animal is tested monocularly with the LEE. The fact that three AME 8/1 cats that received only AME 8/1 exposure performed better with the LEE than did two AME 8/1 cats of the same age that had received 10 weeks of binocular exposure suggests that binocular exposure is worse than continued imbalance. Time Course The severity of the visual field deficits increased with time after the beginning of binocular exposure. Deficits were initially more evident when the animal was tested binocularly but their severity increased with time both when the animal was tested binocularly and when the animal was tested monocularly with the LEE. This effect was most dramatic in the AME 8/1 cats who had received an extreme imbalance in stimulation of the two eyes during early postnatal life. There is a suggestion in the data from binocular testing of an initial (4- to 6-week) period of improvement in visual fields, followed by a steady deterioration that proceeds at different rates for different groups of animals. The animal's poor performance at the time they are first tested binocularly (0 weeks) and shortly thereafter (2 weeks; see Fig. 7) may reflect an initial confusion produced by their first experience with simultaneous stimulation of the two eyes. Since the animals are probably esotropic, this simultaneous stimulation is likely to result in diplopia. As the animal becomes familiar with this experience its performance may show a temporary improvement. Subsequent deterioration in performance is presumably due to suppression of the input from the LEE. Our finding that suppression becomes worse with time is reminiscent of a report by Glass, 25 who used visually evoked potentials to study suppression in a monocularly deprived (MD) cat. He found that the response evoked by stimulation of the deprived eye was suppressed by binocular exposure, becoming progressively smaller with time. It could be reinstated by suturing together the lids of the nondeprived eye. Thus, in both Glass's study and our own, suppression increased with continued binocular exposure. Relation to Previous Work Several studies have provided evidence for suppression in cats who had one or both eyes rotated or one eye paralyzed. 21 One study found, as we did, peripheral visual field deficits that were present when the animal was tested binocularly, but not when it was tested monocularly. 26 Two studies 21 ' 27 have examined suppression by training cats to perform visual discriminations with both eyes open and then testing them with each eye alone. Any animal that could perform the discrimination with only one of its two eyes was assumed to have suppressed the other eye during the original binocular training. In each of these studies, in every case where suppression was obtained, the animals' eyes were misaligned, and in most cases, one eye was clearly at a disadvantage. However, no attempt was made in these studies to vary the degree of the disadvantage, and it was not possible to predict the occurrence or extent of the suppression. Results of several studies have thus provided evidence of suppression in cats under conditions of binocular exposure when the eyes are misaligned and one eye is at a disadvantage. Since the eyes of MD cats are usually misaligned 17 and the deprived eye is clearly at a disadvantage, one would expect the deprived eye to be suppressed, as Glass found using evoked potentials. 25 Indeed, we observed suppression

10 No. 4 BINOCULAR SUPPRESSION IN UNEQUAL AME CATS / Tumoso er ol. 505 of the deprived eye in one of two MD cats given prolonged binocular exposure (unpublished observations). Surprisingly, however, Sherman 28 found no evidence of suppression when testing the visual fields of three MD cats after at least one year of binocular exposure. It is possible that in many MD cats, the inputs from the deprived eye are sufficiently reduced, especially in the region of binocular overlap, 1629 " 31 that binocular exposure does not cause diplopia, and, thus, that there is no stimulus for suppression. Conditions Contributing to Suppression The first contributing condition to suppression is probably the misalignment of the eyes. All but one of our AME cats appeared to be esotropic, confirming earlier reports 23 ' 32 that AME produces strabismus. Even the one AME cat with apparently normal alignment may have had a slight esotropia that our methods were unable to detect. Thus, strabismus may be a necessary condition for the development of binocular suppression. It does not, however, appear to be a sufficient condition: Four of our cats (AME 1/ 1 #8, AME 8/7 #1, AME 8/7 #2, and AME 8/4 #9) showed no evidence of suppression, and all four had vergence ratios outside the normal range (Table 1). The second contributing condition to suppression is undoubtedly that one eye has been placed at a disadvantage. We created the disadvantage by varying the relative amount of exposure of the two eyes during rearing. The extent of the suppression was correlated with the size of the imbalance: the greater the imbalance in stimulation during rearing, the greater the suppression (Fig. 8). The strabismus undoubtedly gives rise to inputs that cannot be readily fused by the animal. To eliminate the resulting conflicts, the animal will probably use one eye at a time. It may alternate between the eyes equally, it may alternate between the two eyes unequally, or it may use one eye to the exclusion of the other. We suggest that the unequal stimulation of the two eyes during rearing has increased the probability that the animal will favor its MEE and, thus, alternate between the two eyes unequally, or even use the MEE to the exclusion of the LEE. 33 Both strabismus and AME reduce the number of cells in area 17 that respond to stimulation of either eye. 719 ' 20 This loss in binocularity is not sufficient to produce the suppression we observed in our unequal AME cats since neither the AME 1/1 cat nor the AME 8/7 cats suppressed. The loss in binocularity may not even be necessary, since Peck et al 26 reported that two of their cats with binocular rotations suppressed one eye and Crewther et al 34 have found many binocular cells in cats with binocular rotations. Further, Buchtel et al 21 reported that their "suppressors" did not show as much of a loss in binocularity as their "alternators" did. An imbalance in stimulation changes the relative number of cells activated by each eye so that the MEE activates more cells than the LEE. This effect is greater in AME 8/1 cats than it is in AME 8/4 cats. 20 It is possible that the relative number of cells driven by stimulation of an eye determines the relative frequency with which that eye will be used. This would explain ourfinding that both AME 8/4 cats and AME 8/1 cats suppress the LEE, but that the suppression is greater in the AME 8/1 cats than it is in the AME 8/4 cats. Relation to Clinical Studies In this paper we have presented evidence that the LEE of an unequal AME cat is suppressed under conditions of binocular exposure, presumably to eliminate the diplopia produced by the misalignment of the two eyes. This suppression is first seen only with binocular testing, although it eventually is apparent even when the LEE is tested alone. In these respects, the suppression observed in unequal AME cats resembles that observed in human strabismic patients. 1 ' 33 However, the "suppression scotomas" of the cats were peripheral and although peripheral suppression scotomas are occasionally seen in humans, central suppression scotomas are more common. 1 Nevertheless, some of our observations may be of clinical significance. It is possible, for example, that the reason that one of our AME 8/4 cats (AME 8/4 #9) never showed evidence of suppression may be related to the fact that this was one of only two cats that received three weeks of exposure during which it was periodically forced to use only its LEE, and during this period the LEE was used at least as much as the MEE. Thus, a reversal in the relative exposure of the two eyes immediately following the initial imbalance in stimulation during early postnatal life may help to prevent suppression of the LEE. This observation is consistent with the clinician's strategy of patching one eye as a treatment for amblyopia 133 and suggests that unequal AME cats may be a useful animal model for the experimental study of such clinical procedures. Key words: suppression, visualfields,amblyopia, strabismus, unequal alternating exposure, cats, behavior. Acknowledgments The authors thank P. Caruccio, E. A. O'Neill, and D. Zimmerman for help rearing and testing the animals, L. P. Welch for help preparing the manuscript, R. Loos and R. Speck for help preparing the illustrations, and D. Tieman for help in every phase of this project.

11 506 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / April 1983 Vol. 24 References 1. Duke-Elder S and Wybar K: Ocular Motility and Strabismus. In System of Ophthalmology, Vol. 6, Duke-Elder S, editor. St. Louis, CV Mosby, 197.3, pp Cleland BG, Mitchell DE, Gillard-Crewther S, and Crewther DP: Visual resolution of retinal ganglion cells in monocularlydeprived cats. Brain Res 192:261, Dews PB and Wiesel TN: Consequences of monocular deprivation on visual behaviour in kittens. J Physiol 206:437, Ganz L and Fitch M: The effect of visual deprivation on perceptual behavior. Exp Neurol 22:638, Ganz L, Hirsch HVB, and Tieman SB: The nature of perceptual deficits in visually deprived cats. Brain Res 44:547, Giffin F and Mitchell DE: The rate of recovery of vision after early monocular deprivation in kittens. J Physiol 274:511, Hubel DH and Wiesel TN: Binocular interaction in striate cortex of kittens reared with artificial squint. J Neurophysiol 28:1041, Jacobson SG and Ikeda H: Behavioural studies of spatial vision in cats reared with convergent squint: Is amblyopia due to arrest of development? Exp Brain Res 34:11, von Griinau MW and Singer W: Functional amblyopia in kittens with unilateral exotropia. II. Correspondence between behavioral and electrophysiological assessment. Exp Brain Res 40:305, von Noorden GK and Dowling JE: Experimental amblyopia in monkeys. II. Behavioral studies in strabismic amblyopia. Arch Ophthalmol 84:215, Singer W, von Griinau M, and Rauschecker J: Functional amblyopia in kittens with unilateral exotropia. I. Electrophysiological assessment. Exp Brain Res 40:294, Ikeda H and Tremain KE: Amblyopia occurs in retinal ganglion cells in cats reared with convergent squint without alternating fixation. Exp Brain Res 35:559, Tumosa N, Tieman SB, and Hirsch HVB: Unequal alternating monocular deprivation causes asymmetric visualfieldsin cats. Science 208:421, Tumosa N, Tieman SB, and Hirsch HVB: Visual field deficits in cats reared with unequal alternating monocular exposure. Exp Brain Res 47:119, Sprague JM and Meikle TH Jr: The role of the superior colliculus in visually guided behavior. Exp Neurol 11:115, Sherman SM: Visual field defects in monocularly and binocularly deprived cats. Brain Res 49:25, Sherman SM: Development of interocular alignment in cats. Brain Res 37:187, Olson CR and Freeman RD: Eye alignment in kittens. J Neurophysiol 41:848, Blake R and Hirsch HVB: Deficits in binocular depth perception in cats after alternating monocular deprivation. Science 190:1114, Tieman DC, McCall MA, and Hirsch HVB: Physiological effects of unequal alternating monocular exposure. J Neurophysiol 49:804, Buchtel HA, Berlucchi G, and Mascetti, GG: Behavioural and electrophysiological analysis of strabismus in cats. In Aspects of Neural Plasticity, Vital-Durand F and Jeannerod M, editors. Paris, INSERM, 1975, pp Olson CR: Spatial localization in cats reared with strabismus. J Neurophysiol 43:792, Packwood J and Gordon B: Stereopsis in normal domestic cat, Siamese cat, and cat raised with alternating monocular occlusion. J Neurophysiol 38:1485, Bruce CJ, Isley MR, and Shinkman PG: Visual experience and development of interocular orientation disparity in visual cortex. J Neurophysiol 46:215, Glass JD: Plasticity of cat visual cortex. Exp Neurol 70:446, Peck CK, Barber G, Pilsecker CE, and Wark RC: Visual field deficits in cats reared with cyclodeviations of the eyes. Exp Brain Res 41:61, Peck CK and Wark RC: A relationship between visual suppression and amblyopia in cats with cyclodeviations of the eyes. Exp Brain Res 44:317, Sherman SM: Permanence of visual perimetery deficits in monocularly and binocularly deprived cats. Brain Res 73:491, Wiesel TN and Hubel DH: Comparison of the effects of unilateral and bilateral eye closure on cortical unit responses in kittens. J Neurophysiol 28:1029, Shatz CJ and Stryker MP: Ocular dominance in layer IV of the cat's visual cortex and the effects of monocular deprivation. J Physiol 281:267, Tieman SB: Loss of geniculocortical terminals from deprived laminae in monocularly deprived cats. Soc Neurosci Abstr 5:810, Blake R, Crawford MLJ, and Hirsch HVB: Consequences of alternating monocular deprivation on eye alignment and convergence in cats. Invest Ophthalmol 13:121, Jampolsky A: Unequal visual inputs in strabismus management: a comparison of human and animal strabismus. In Symposium on Strabismus. Transactions of the New Orleans Academy of Ophthalmology. St. Louis, CV Mosby, 1978, pp Crewther SG, Crewther DP, Peck CK, and Pettigrew JD: Visual cortical effects of rearing cats with monocular or binocular cyclotorsion. J Neurophysiol 44:97, 1980.