DEVELOPMENTAL EFFECTS OF BINOCULAR CONTOUR DEPRIIIATION AND LIGHT DEPRIVATION 0N EYE ALIGNMENT IN KITTENS Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY DIANE L. SMOLEN 1977 _ _ ,.~4 “2‘ I)ate 0-7639 This is to certify that the thesis entitled DEVELOPMENTAL EFFECTS OF BINOCULAR CONTOUR DEPRIVATION AND LIGHT DEPRIVATION 0N EYE ALIGNMENT IN KITTENS presented by DIANE L. SMOLEN has been accepted towards fulfillment of the requirements for PH . D. Jegree in PSYCHOLOGY g Major prof or July 28, I977 ABSTRACT DEVELOPMENTAL EFFECTS OF BINOCULAR CONTOUR DEPRIVATION AND LIGHT DEPRIVATION ON EYE ALIGNMENT IN KITTENS By Diane L. Smolen Kittens, at eye Opening, exhibit a large divergent strabismus and appear to be behaviorally blind. In this case the eyes are directed temporarily rather than straight ahead. Development of interocular eye alignment occurs in kittens between the fourth and seventh postnatal weeks. During this time period the animal also begins to respond positively to visual stimuli. Several studies have indicated that some functional aspects of the kitten visual system are highly dependent upon environmental stimulus exposure during early development. This study is primarily concerned with the effects of binocular contour deprivation and light deprivation on the development of interocular eye alignment in kittens. TO assess the role of light deprivation, kittens were reared a) with limited exposure in scotopic illumination, b) with limited exposure in photopic illumination and c) in total darkness. In both the scotopic and photopic illumination half of the kittens were binocularly deprived of contour to assess the role of contour in the development of interocular alignment. Littermates were exposed to contour. Exposure periods were for four hours. each day from the third postnatal week until the third month (the deprivation period). At the end of the third postnatal month interocular eye alignment Diane L. Smolen was measured using a modified version of the corneal reflex technique. Seven normal three month Old kittens served as controls to establish a normal range of interocular eye alignment from which to compare a suspected strabismus. The results suggest that contour seems to be the primary contributing stimulus factor in the development of interocular eye alignment. The level of illumination (scotOpic or photOpic) does not seem to be important. Secondly, subsequent normal visual experience following deprivation did not reveal a re-alignment or recovery. The data suggest that simultaneous contour exposure during the critical period is necessary for the development of interocular eye alignment in kittens. Kittens deprived of contour did not develop normal interocular eye alignment. The data support the hypothesis that eye alignment and perhaps full binocular integration are dependent upon the stimulus environment. DEVELOPMENTAL EFFECTS OF BINOCULAR CONTOUR DEPRIVATION AND LIGHT DEPRIVATION ON EYE ALIGNMENT IN KITTENS BY Diane L. Smolen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1977 6);. ‘7 Cé Io T0 5 . HOWARD BARTLEY ii ACKNOWLEDGMENTS I am grateful to Dr. James L. Zacks, my academic advisor and Chairman of my dissertation committee, who provided guidance and encouragement throughout my doctoral training. I would like to thank Dr. Richard J. Ball for his assistance and expertise regard- ing the use Of contact lenses as visual occluders. I would also like to thank Drs. John 1. Johnson and Charles Tweedle for many pertinent discussions and insights during this undertaking. In addition I would like to thank Allen Vieth for his critical comments, discussions and personal support during my graduate training. TABLE OF CONTENTS LIST OF TABLES ...................... LIST OF FIGURES ...................... INTRODUCTION ....................... The Effects of Monocular and Binocular Deprivation During the Critical Period .............. Selective Environmental Deprivation: Neural Plasticity ...................... Specific Background ................. METHOD . . . . ..................... Subjects ....................... Contact Lenses .................... Room Illumination ................... Rearing Conditions .................. Assessment of Interocular Alignment ......... RESULTS ......................... The Origin of Strabismus ............... Clinical Implications ................. Interocular Eye Alignment: A Critical Period Phenomenon ...................... The Development of Interocular Eye Alignment ..... Qualifications .................... REFERENCES ........................ iv 24 69 Table II III IV LIST OF TABLES Analysis of variance summary indicating contour as the significant factor .............. Duncan's Multiple Range test summarizes multiple comparisions between group means ...... Analysis of variance summaries for each group over test days suggests stability of measurement ................................. Analysis of variance summary from data on kittens on day 5 after deprivation...... ....... Duncan's Multiple Range sumnarizing the multiple comparisions between group means combining all measurements for the entire photographic series ....................... ..... Page 35 37 44 54 LIST OF FIGURES Figure Page l. The distance (in mm) between the corneal light reflex and the center of the pupil in each eye was measured and summed. This measure described as the interpupilary distance (IPD) minus the inter-reflex distance (1RD) is positive when the animal is divergent and negative when the animal is convergent .................................. 3O 2. The mean interocular eye alignment (inter- pupilary distance minus inter-reflex distance) is depicted for individual animals +lSE. Each panel (A to F) represents an addithnal deprivation condition from Normal. High illumination with Contour experience. High illumination Contour Deprived, Low illumination Contour Deprived to Dark Reared . The normal range depicted fbr across group comparisions is the group mean :JSD ...................................... 33 3. Within each panel the data are shown for a litter. In panels A and C data are shown fOr kittens who were raised in high illum- ination, in panels 8,0 and E the data are shown fOr kittens reared in low illumination and in panel F the data are depicted for dark reared kittens. Within each panel (A-D) it can be seen that littermates who were contour deprived had eye alignments that were clearly different from animals who were contour experienced. The data presented in panels C and D provide a genetic control across light conditions depicting contour deprivation as the Significant factor ...................... 39 vi Figure Page 4. Mean interocular eye alignment is depict- . ed as the mean difference between inter- reflex and interpupillary distance for each group for days 2,3,4 and 5 of test- ing. The same trends that existed on day l appear to hold over the five day test period. The High Contour Deprived, Low Contour Deprived and Dark Reared kittens Show eye alignments that are not normal .................................... 42 5. In figure 5A and 8 individual means are shown over the five day period for kittens reared in Low illumination. The data Shown in A are from Contour Deprived kittens, the data shown in B are litter- mates that experienced Contour. The normal range is illustrated fbr comp- arision. With the exception of CD , the Contour Deprived kittens are sgen to have abnormal eye alignments compared to both the normal range and to littermates that were Contour experienced .................. 47 6. In figure 6 the individual means are shown for each test session for animals raised in Hight illumination. In A the data are shown for kittens that were Contour Deprived. In B the data are shown for littermates that were exposed to Contour. Kitten HCD exhibits eye alignments that are clearl within the normal range ............ 50 7. Figure 7 shows individual means for each kitten on each test day. All kittens were Dark Reared. Eye alignments are not particularly different from kitten to kitten and are clearly outside the normal range .......................................... 52 vii Figure Page 8. Figure 8 depicts the data for three groups of animals following normal visual exp- erience. The first data point for each animal is the mean difference for all five test sessions. The second and third data points are mean differences follow- ing at least one month Of normal visual experience. The data indicate that once eye alignment was disrupted that the recovery did not occur. In none of the recovery measurements was a , previously strabismic animal found to have normal interocular eye alignment .......... 57 9. Figure 9 shows the data for three normal kittens. These kittens were photo- graphed at 4 weeks, 5 weeks, 6 weeks and 7 weeks. The fixation target was at l2 cm. This is the time period of interocular eye alignment. Note that these kittens were not observed to have eye alignments that fell within the normal range for the first seven weeks ............................... 60 viii INTRODUCTION A majority of the research on sensory development has been concerned with the effects of environmental manipulation on the structure and function of the kitten visual system. Much Of the current interest in the problem originated with a series of investigations by Hubel and Wiesel. Perhaps the most interesting finding was the delineation of a sensitive or critical period. By varying the time periods of monocular deprivation and comparing the resulting functional deficits, Hubel and Wiesel (1970) deter- mined the period of susceptibility, or the “critical period", to be during the 4,5 and sixth weeks. Deprivation during this sensitive period was fbund to have profOund effects on the functional and morphological integrity of the developing visual system. However, in the adult cat deprivation was found to have no effect. A myriad of deprivation methods have sub- sequently been employed to investigate this finding of neural plasticiy-- induced neural modification following short term stimulus deprivation. These techniques range from total light and fbrm deprivation, to selective stimulus exposure such as horizontal or vertical stripes, to artificially induced strabismus and astigmatism. The discussion begins with a review of the classic work of Hubel and Wiesel describing the functional deficits induced by monocular and binocular eye lid suture. The effects of deprivation on the Lateral Geniculate and the Superior Colliculus will also be included in this section. Behavioral indications of perceptual deficits are also discussed. In the 2 second section the effects of selective environmental depriVation are discussed. Although somewhat controversial these data provide dramatic examples of neural plasticity. The final section details the Specific background and rationale for the present study. Included in this discussion will be a review of the literature suggesting that cats are binocular animals, a depiction of the neurophysiological mechanism presumed to underly binocular single vision and stereopsis, and the development of interocular eye alignment in kittens, presumed to be a necessary prerequisite for binocular single vision. It is at this point that the literature from the precedihg sections can be brought to bear upon the concerns Of the present study. This investigation was designed to independently evaluate the effects of contour deprivation and light deprivation on the development of interocular eye alignment in kittens. The study also suggests some consequences of binocular deprivation on the binocular substrate of the developing kitten visual system. THE EFFECTS OF MONOCULAR AND BINOCULAR OCCLUSION DURING THE SENSITIVE PERIOD Cortical Cell Responses Following Deprivation In the adult cat a majority of cortical cells are binocularly activated. That is. two receptive fields exist for that cell, both having similar receptive field properties and both are in corresponding locations in the visual field. A cell's receptive field is the small part of the visual field in which stimulation affects that cell's firing rate. Cortical cells respond poorly to diffuse stimuli (responding best to edges or bar-like stimuli), respond to movement and exhibit stimulus orientation preferences often called that cell's receptive field axis. A binocular cell responds optimally (defined 3 as the maximum number of spikes and/or maximum peak firing rate) when both eye's receptive fields are stimulated and the most effective stimulus configuration in one eye is also effective in the other eye. Thus the stimulus orientation and the direction and rate of stimulus movement must be nearly identical for both receptive fields. Although a majority of the neurons are binocular, a single cortical cell may respond better (with a higher response rate) to the stimulation of one eye cOmpared to the response following stimulation of the other eye. One eye tends to dominate. The distribution of ocular dominances for cortical cells ranges from cells dominated completely by the stimulation Of the contralateral eye to cells dominated completely by the ipsilateral eye to cells that respond equally to the stimulation of either eye. Histological reconstruction of electrode tracks indicates that cells are organized in functional columns according to Similar orientation preferences, disparities and ocular dominances. These columnar organizations are independent of each other. That is, a column may contain, for example, cells with similar orientation preferences but not all the same Optimum disparities or ocular dominances (Hubel and Wiesel, 1963,1969;Blakemore, I970). The term receptive field disparity refers to the distribution of locations of receptive fields for one eye after the corresponding receptive field pairs for the other eye have been artificially superimposed. This mechanism suggests that for cortical cells in a localized area, that receptive field pairs (having similar receptive field properties) will be correspondent or superimposed at different depths. After monocular deprivation, achieved by eyelid suture during the sensitive period, almost all of the neurons responded to the experienced, non- deprived eye (Wiesel and Hubel, 1963). The visually deprived eye 4 seemed to be functionally disconnected from cortical cells. A similar effect was observed after suturing the nictitating membrane and after occluding the eye with opaque and translucent contact lenses. Ganz, Fitch and Satterberg (1968) have reported that 38% of the cells recorded from respond to stimulation of the deprived eye. However, these cells were found to lack many of the normal receptive field characteristics. They were not directionally sensitive, were easily fatigued and they responded to unusually large stimuli. In binocularly deprived kittens, (Wiesel and Hubel, 1965) cells in the striate cortex were fbund to be responsive to both eyes but with a slight reduction in the number Of cells driven equally by the two eyes. Out of 139 cells, only 57 were normal. 45 showed no orientation preferences and 37 cells were not responsive to visual stimulation at all. In agreement with these data, Chow and Stewart (1972) reported similar deficits in receptive field properties following binocular deprivation. In summary, both monocular and binocular deprivation have been found to have profound effects on the functional integrity of the developing visual cortex. Monocular deprivation, severely reduces binocular cortical interactions since the deprived eye seems to become ineffectual in activating cortical cells. Binocular deprivation also disrupts binocular cortical interactions, although less severely than monocular deprivation. In addition this form of deprivation seems to affect the receptive field properties of these cells. 5 Dorsal Lateral Geniculate Nucleus: Morphological and Functional Deficits Following Deprivation The Lateral Geniculate Nucleus (LGN) in the cat has three layers A, A1 and B. A and B receive retinal input from the contralateral eye and A1 receives retinal input from the ipsilateral eye. Following monocular deprivation a severe reduction in cell size was noted in those layers that received inputs from the deprived eye. By varying the method of occlusion so as to increase the amount of light entering the deprived eye, the atrOphy was less pronounced. This reduction in cell size was also observed in binocularly deprived animals (Wiesel and Hubel, 1963). To explain the presence of atrophy following monocular occlusion, Hubel and Wiesel have suggested that the deprived eye may not be able to compete with the experienced eye for synaptic sites at the cortical level. This supposition was supported by data (Wiesel and Hubel, 1963) indicating that the functional properties of LGN cells were not affected by visual deprivation. Guillery and his associates have pursued the competition hypothesis. If the competition hypothesis is correct, then morphologically the binocular segments of the deprived lamini ought to remain unaffected. The binocular segment of the LGN is that area that receives retinal input from the binocular visual field. The monocular segment receives retinal input from the ipsilateral monocular visual field (Sanderson, 1971). If visual deprivation affects only cell growth, then both monocular and binocular segments ought to be equally affected . Replicating the Hubel and Wiesel study, animals were monocularly deprived (Guillery and Stelzner. 1970). The binocular segments of the deprived lamini showed a 28% reduction in cell size while the 6 monocular segment was not found to be significantly different from normal. In squirrels where the monocular and binocular segments of the Lateral Geniculate are more clearly delineated similar results were obtained (Guillery and Kaas, 1974). More recently other investigators have looked more closely at the functional characteristics of LGNd cells both in the normal animal and following visual deprivation. Functionally, cells in the LGN exhibit concentric receptive field organization and for most cells, receptive fields can be plotted for both eyes (Singer, 1970; Freund, 1973). The dominant eye provides the excitatory input and concentric organization while the non-dominant eye contributes an inhibitory influence. A cell's maintained discharge is reduced to zero fbllowing stimulation of the non-dominant eye's receptive field. Simultaneous stimulation of both receptive fields causes a 30% reduction in the response. Hamasaki and Winters (1973) have found that 60% of the LGN cells encountered are abnormal when the animal has been visually deprived. These units showed normal receptive field organizations, normal sized receptive field centers, normal response functions and normal spatial summation. They differed from normal in that the maximum transient firing rates were lower, latencies were longer, the slopes of the intensity response functions were flatter, and peripheral inhibition seemed to be reduced. Sherman and Sanderson (1972) encountered fewer cells in the deprived lamini and the number of cells with binocular interactions was affected slightly. Binocular deprivation reduced the percentage of cells found to exhibit binocular interactions. Visual deprivation has been found to selectively affect a particular cell type (Y-Type cells) in the LGN Y I- TY res; and firm .‘JIII sing: If Y. I II Nan. 7 (Sherman, Hoffman and Stone, 1972). In the normal animal three types of ganglion cells have been classified, X-Type cells, Y-Type cells and W-Type cells, on the basis of responses to particular stimuli, response characteristics (sustained or transient), conduction velocities and projection sites (Enroth-Cugell and Robson, 1966; Cleland, Dubin and Levick, 1971; Stone and Hoffman, 1972). X-Type cells and Y-Type cells are reported to project to the LGNd while W-Type cells are found to project to the Superior Colliculus along with some Y-Type axon collaterals. Selective decrements in the number of Y-Type cells were found depending upon whether the animal had been monocularly or binocularly deprived. Animals that had been monocularly deprived were fbund to have significantly reduced numbers of Y-Type cells in the binocular segments Of the deprived lamini, while the normal complement of Y-Type cells was observed in the monocular segment of deprived lamini. In the nondeprived or experienced binocular segments, a Slightly greater- than-normal complement of Y-Type cells seemed to exist. In binocularly deprived animals, both the monocular and binocular segments were found to have reduced numbers of Y-Type cells. The criteria for distinguishing X and Y type cells were conduction velocity and response characteristics. The retina, however, appears to remain unaffected by visual deprivation in the cat (Sherman and Stone, 1973). After both monocular and binocular occlusion the normal proportions of X-Type and Y-Type cells were observed. Latencies and retinal histology were also normal. These retinal data would seem to lend support to the synaptic competition hypothesis. Visual deprivation does not affect the system at the periphery. with SI II :‘III 8 However, the actual locus of synaptic competition remained to be clarified. By suturing the lids of one eye and placing a discrete lesion in binocular retinal fields of the experienced eye (for the duration of the critical period), the effects of competition should be reduced in the corresponding binocular segments of the sutured eye's projections. Results were not conclusive, but the effect (a decrease in cell size) in the deprived eye's binocular layers Of LGN that corresponded to the lesioned area was slightly reduced (Guillery, 1972). Sherman, Guillery Kaas and Sanderson (1974) have looked at the critical segment behaviorally, electrOphysiologically and morphologically. The critical segment is defined as those areas of the deprived eye's projections that correspond to the lesioned area and its projections in the experienced eye's pathway. Stimulation of the deprived eye proved relatively ineffective at eliciting cortical single unit responses. Morphologically, the sparing of large cells within the critical segment of the deprived eye was not pronounced. However, behaviorally, the animal was able to respond visually to stimuli in both the monocular portion of the visual field and in the critical segment of the binocular visual field. In summary, visual deprivation via monocular and binocular eye lid suture selectively affects the Y-Type cells of the LGN. This would seem to be the functional correlate of the reduction in cell size previously observed by Hubel and Wiesel. Monocular deprivation reduced the number of Y-Type cells in binocular segments of deprived lamina, although normal representation was Observed in monocular segments. Binocular deprivation results in a reduction in the number of Y-Type cells found in both monocular and binocular segments 9 Of the deprived lamini. The retina remains unaffected. Some form of synaptic competition between the deprived and experienced eye for cortical post-synaptic sites, changing secondarily the properties of LGN cells, almost certainly exists. Behavioral Correlates of Perceptual Deficits Behavioral testing of visually deprived kittens has revealed accompanying perceptual deficits. Dews and Wiesel (1970) teSted for acuity, using horizontal stimuli with and without "gaps“ or breaks in the lines, perceptual motor co-ordination, using a water barrier test and luminance, using a light-dark discrimination. They reported a decrease in acuity and a failure to learn the water barrier task in monocularly deprived cats tested using the deprived eye only. The kittens were able to make the light-dark discrimination. Ganz and Fitch (1968) tested monocularly deprived animals for deficits in acuity using the autokinetic nystagmus response, placing responses, visually guided placement, gross depth perception, behavior in an open field and pattern discrimination. Immediately fbllowing deprivation, the kittens showed no placing response using the deprived eye and strong startle responses when placed in an open field. These deficits were not evident when the kittens were tested using the experienced eye. After one to two weeks in a normal visual environment, when tested using the deprived eye, visual acuity was still found to be reduced, the animal's performance on a pattern discrimination task was only slightly better than chance and only gross indications of depth perception were present. Other investigators have reported (Ganz, Hirsch and Tieman, 1972) that monocularly deprived animals are incapable of learning a pattern discrimination task but that binocularly deprived animals learn n.‘ A .4; ' , .u 'ug. n'. ‘PI, “ 10 the task, although more slowly than normal animals. However, Rizzolatti and Tradardi (1971) report that monocularly deprived animals can make pattern discriminations but that the discrimination was dependent upon peculiar head movements. Using a visual perimetry test, Sherman (1973) has reported that monocularly deprived animals responded to (oriented to) stimuli presented to the deprived eye in the monocular crescent of the visual field (60-900). In this task the animal is presumably fixating a target at O0 and the visual field is divided into 15° sectors for testing. On a similar task, binocularly deprived animals were observed to respond to stimuli presented in only the ipsilateral hemifield (0-900). Stimuli presented in the contralateral 0-300 were ignored. The normal animal responds to a stimulus presented in the contralateral 30-0o field and to the ipsi- lateral hemifield for each eye. Shenman attributes these behavioral deficits in the monocularly deprived kitten to be related to the functional abnormalities, the loss of the Y-Type cell, found to exist in the binocular segments of the geniculo-striate system. For the deprived eye, only the monocular segments of the geniculO-striate system remain unaltered. He suggests that binocularly deprived kittens are able to make visual responses relying on the normal retino-tectal pathway. The geniculo-striate system is assumed to be impaired. These assumptions can be supported by A) know- ing the normal inputs to and functional characteristics of the Superior Colliculus and B) Observing behaviorally and functionally the deficits found to exist in animals with cortical lesions and comparing these data with the data Showing similar effects on the Superior Colliculus following deprivation. II Superior Colliculus: Functional Deficits Following_Deprivation In the normal animal the Superior Colliculus (SC) receives an almost completely crossed retinal-tectal input consisting of Y-Type cell axons (9%) and W-Type cell axons (73%) and an indirect input of Y-Type cell axons (18%) via the striate cortex. A majority of the cells of the SC respond binocularly and are responsive to horizontal movement (Wickelgren—Gordon, 1972; Sterling and Wickelgren1969). Both the binocularity and the directional sensitivity of cells in the SC have been attributed to the cortical, ipsi- lateral Y-Type cell input. Following cortical ablation a severe decrease in both binocular responses and the number of units responsive to movement is found (Wickelgren and Sterling, 1969). In cases where striate cortex was lesioned unilaterally, neurons Of the SC contralateral to the lesion maintained direttional sensitivity and binocular response properties. Behaviorally, animals with cortical ablations are Observed to respond to stimuli presented to only the ipsilateral hemifield (Sprague, 1966). Replicating this study, Sherman (1974) lesioned areas 17,18 and 19 and produced similar behavioral deficits. These behavioral data are similar to the data reported following binocular deprivation and decortication (Sherman, 1974). Kittens deprived of visual experience during the critical period via unilateral lid suture were found to have functional collicular abnormalities similar to those animals with unilateral cortical lesions. In the SC of these animals almost all cells were driven by the non-deprived or normal eye. Hoffman and Sherman (1974) found a decrease in the Y-Type cell (cortical indirect) input to the Superior Colliculus ipsilateral to the deprived eye following monocular deprivation. The experienced eye seemed to have an increase, over normal proportions, of the Y-Type cell indirect input. The ocular dominance distributions obtained were found to be somewhat 12 dominated by the ipsilateral or normal eye with some proportion of binocular cells remaining. Collicular~ units ipsilateral to the deprived eye were found to respond, overwhelmingly, to the non-deprived or contralateral eye with only an occasional binocular response. Either bilateral or contralateral lesioning of areas 17,18 and 19 was found to increase the effectiveness of the previously deprived eye for activating collicular units (Wickelgren, 1969). Animals deprived of early visual experience via binocular eye lid suture were found to have collicular cells driven only by the contralateral eye and only 12% of the units were directionally sensitive (Sterling and Wickelgren, 1970). In summary the effects of monocular and binocular deprivation on the Superior Colliculus are as follows. After binocular eye lid suture, the colliculus functions much the same as in the decorticate animal, relying on intact retino-tectal pathways. This would seem to indicate that the retino-tectal input dominates in this condition. In monocular eye lid suture the visual cortex still exerts some influence over the colliculus, perhaps through the monocular Y-Type cell indirect pathway. Recovery From Visual Deprivation The effects of eyelid suture on the physiology of the developing visual system appear, in most cases, to be permanent. Varying recovery periods from three to eighteen months, and in some cases reverse suturing (to force the use of the previously deprived eye) Wiesel and Hubel (1965) showed no evidence of functional recovery, although Dews and Wiesel (1970) reported slight improvements. behaviorally following reverse suturing. Blakemore and Van Sluyters (1974) have monocularly deprived animals and then reverse sutured at 5,6,8,10 and 14 weeks. The only case in which reverse suturing was found to alter the cortical ocular dominance distribution 13 was in favor of the initially deprived eye was when reversal occured during the "critical period" at 5,6 and 8 weeks. In these animals the ocular dominance distributions shifted away from the experienced (now sutured ) eye toward the deprived eye (now Open). If reversal occured at fourteen weeks cortical responses favored the experienced eye. The behavioral data concerned with the recovery of perceptual deficits are somewhat contradictory. Ganz and Fitch (1968) report slight improvements following reverse suturing and Others find that monocularly deprived animals are able to make some pattern judgments, using the deprived eye after many trials (Chow and Stewart, 1972). Recent data reported at the annual Association for Research in Vision and Ophthalmology (Sarasota, 1976) suggested that behavioral assessment of the visual capacity of a kitten following early deprivation may not be consistent with the profound functional deficits. D.E. Mitchell (paper presented) reported that kittens dark reared or binocularly deprived for either four months or 45 days developed normal visual acuity (tested with square wave gratings) after three months of normal visual experience. Kittens that were monocularly deprived and then reverse sutured at 45 days achieved normal acuity levels within 3 to 4 days of normal experience. SELECTIVE ENVIRONMENTAL DEPRIVATION: NEURAL PLASTICITY Recording from visually inexperienced kittens at 8.16 and 20 days, Hubel and Wiesel (1963) found cortical cells to have adult-like functional characteristics. Single units were reported to have orientation preferences and were binocularly activated. These data led Hubel and Wiesel to conclude that visual experience was not the necessary variable 14 for functional development but perhaps a factor in the maintenance of the system. Contradictory data (Barlow and Pettigrew, 1971) Show a lack of specificity or adult-like properties in young kittens. Disparity preferences and probably orientation preferences were not observed. The young animal seemed to have "broadly tuned" responses when compared to the Optimum response preferences of the adult cat. The controversy between environmental maintenance of innate functional organization and neural plasticity--induced neural modification following short term deprivation has led to a number of recent studies. The data suggest that although the kitten's visual system seems to be extremely sensitive to environmental exposure during the "critical period" its plasticity is somewhat constrained by innate or inherent neural organization. For example, both visually inexperienced and dark reared animals seem to have bincoular cells. Yet cortical cells in dark reared animals Show no obvious preference for edges or Spot stimuli in contrast to the normal animal. Also kittens reared in a planetarium like environment show optimal single unit responses to very small spot stimuli (Pettigrew and Freeman, 1973) in contrast to the normal animal where Optimum reSponses are obtained from bar or edge like stimuli. Selective deprivation has been used to look at the extent to which the neural system can be modified by restricted exposure during the critical period. Kittens exposed only to horizontal or vertical lines during the critical period have been found to have cortical cells with orientation preferences coincident with the early environment (Hirsch and Spinelli, 1970; Blakemore and COOper, 1970). Although not rigorously tested, Blakemore and Cooper reported that kittens reared, for example, in a vertical environment did not respond behaviorally to objects oriented horizontally. 15 Pettigrew and Olson (Pettigrew, Olson and Hirsch, 1973) found atypical orientation preferences in a three year old cat. Unknown to them at the time, the animal had originally been a part of the original Hirsch and Spinelli study. The left eye had viewed three horizontal stripes, while the right eye had viewed three vertical stripes during the sensitive period. In this animal optimum orientation responses obtained through stimulation of the left eye were within :200 Of horizontal and preferences Obtained through stimulation of the right eye were within 120° if vertical. All binocular interactions had been eliminated. Leventhal and Hirsch (1975) have since reared kittens wearing goggles where the left eye was exposed to a 45° diagonal and the right eye was exposed to a 1350 diagonal. Although some cortical cells were found to have orientation preferences similar to the diagonal viewed during the early experience (29%), the remaining cells were not binocular. These data suggest that some cells may maintain or develop intrinsic specificities without visual stimulation since horizontal and vertical preferences were Observed. During a recording session, Pettigrew and Garey (1974) exposed single units to vertical square wave gratings for 5 to 22 hours in visually inexperienc- ed animals. Immediately following the "exposure period", cells were found to be broadly tuned, with few cells exhibiting any orientation preferences. However, when the animals were returned to total darkness for durations of 1,18 and 20 days and then retested, large clusters of cells were found with orientation preferences coincident with the grating orientation seen during the "exposure period". Five cells (of 52) were found to respond optimally to horizontal orientations. When the initial exposure period was done using a monocularly split field, cortical cells from the unexposed hemisphere were found to be broadly tuned, characteristic l6 of the visually inexperienced animal. On the stimulated side, small dense terminals with high vesicle densities were found when compared to the unstimulated side (Garey and Pettigrew, 1974). Coleman and Riesen (1968) had similar data showing smaller dendritic lengths and fewer dendrites following dark rearing. The data showing early stimulus modification of orientation specificities have become somwehat controversial in light of a recent study. Maffei and Fiorentini (1974) raised kittens in vertical square wave grating environments. These kittens were found to have cortical units with normal orientation preferences. Secondly, reduced sensitivities were observed for gratings similar in spatial frequency to the early experience gratings. Similarly, Maffei, Fiorentini and Biste (1973) have shown that spatial frequency contrast thresholds can be elevated for Single simple cortical cells after repeated exposure to high contrast gratings. However, Muir and Mitchell (1973) have replicated the results of the Blakemore and Cooper study while varying spatial frequencies for both the horizontal and vertical environments. At the end of the deprivation period the kittens were taught to discriminate a black square wave grating from a blank field equated for mean luminance. Normal animals performed equally well regardless of grating orientation. Deprived animals make superior discriminations for orientations coincident with the orientation of the early experience. In conflict with the critical period data, it is interesting to note that plasticity has also been observed in the adult cat. The results, however, are not directly predictable from the kitten data. For example, two weeks after monocular paralysis has been surgically induced, reduced binocularityis found in simple cortical cells: (Brown and Salinger, 1975) and reduced prooortions of X-Type cells are Observed in the LGN l7 Creutzfeldt and Heggelund, 1975). Also adult cats exposed to vertical bars for only 14 days have been found to have fgwg5_cortical cells with vertical orientation specificities (Fiorentini and Maffei, 1974). Although somewhat controversial, the data do indicate that some functional aspects of cortical receptive field prooerties can be modified as a result of selective stimulus exposure during the critical period. The adaptability of the kitten visual system seems to be limited by a few inherent mechanisms and may pertain only to the fine tuning of stimulus Specificities. SPECIFIC BACKGROUND Cats like humans, appear to be highly binocular animals. They report- edly have 60° of binocular visual field (Sherman,1973). Hughes (1972) has shown that cats are capable of making vergence eye movements and can binocularly fixate a target at 10cm. His behavioral data Show a change in interpupillary distance (2.6mm) when the animal converges on a fixation target at 10cm from fixating a target at 200cm. Stryker and Blakemore (1972) measured eye movements during vergence in cats and report similar data. Blake and Hirsch (1975) present behavioral evidence suggest- ing that cats are able to make stereoscopic judgements. The normal adult cat was able to detect disparities of 4 minutes Of arc. Binocular performance of this task was an order of magnitude better than monocular performance levels. The main neurophysiological features presumed to underly stereopsis have recently been reviewed by Bishop (1973,1974) and will be briefly reviewed here. Because the two eyes are horizontally separated each eye sees the external world from a slightly different view point. This 18 difference information or retinal disparity has been thought to provide the basic cue for binocular depth perception (Gregory, 1966). The essential anatomical requirement for binocular depth perception is provided by the partial decussation of retinal ganglion cell axons at the Optic chiasm. This enables axons from corresponding or nearly correspond- ing loci of the two retinae to come together anatomically and to synapse on one cell in the visual cortex. The neural discrimination of retinal disparity is made possible by the existence of receptive field disparities. This concept is best illustrated by example. As a result of the topographical projections of the visual field to the visual cortex two sets of receptive fields can be mapped for each cell, one for the left eye and one for the right eye, when recording from any series of cortical cells in a localized area of the visual cortex. The receptive field maps for each eye are artificially superimposed (using a variable prism) and similar adjustments are made for each receptive field pair of the other eye, the other eye's receptive fields will not be superimposed but in slightly disparate positions exhibiting a horizontal distribution of disparaties. For example, within the cortical projections of the area centralis, the mean disparity was found to be 0.00 with a range of 11.20 and a standard deviation Of .50 (Nakara, Bishop and Pettigrew, 1968; Joshua and BishOp, 1970). Maximal stimulation will occur when the stimulus feature, for which that cell is specific, is located in the plane where the two receptive fields are superimposed as both receptive fields have the same stimulus specificities. Therefore a distribution of receptive field disparities within an orientation column would insure a range of depth responses for that particular stimulus configuration. 19 In conjunction with this presumed depth detecting mechanism, the two eyes are highly co-ordinated and in the normal animal a cell's receptive fields are never too far from being superimposed. Also, for example, the excitatory region of a receptive field is for some cells on the average 10 and is banded by inhibitory regions (BishOp, Henry and Smith, 1972). Thus eye misalignment or a lack of coordinated movement might place the receptive field pairs in an out Of register position resulting in mutual inhibition. Binocular and particular monocular occlusion during the critical period are known to disrupt cortical organization. In young kittens, surgically induced strabismus, interocular misalignment causing a failure of the visual axes to intersect on the Object of regard, also reduced binocular cortical interaction. Cells are found to be dominated completely by either the contralateral or ipsilateral eye. If simultaneous binocular stimulation is prevented during the sensitive period by alternately occluding one eye and then the other, 91% of the cortical cells are found to be monocular, responding only to one eye or to the other (Hubel and Wiesel, 1965). Using a conditioned suppression technique, Blake and Hirsch (1975) taught cats to make stereoscopic judgments. Alternately occluded kittens were deficient on the stereoscopic task, performing equally well under monocular and binocular viewing conditions. These data are the first behavioral evidence to suggest that alternate occlusion disrupts not only gross binocular interactions but also stereopsis and presumably the mechanism underlying functional binocular depth perception. Blake, Crawford and Hirsch (1974) noted that alternately occluded kittens are also strabismic. They have proposed that the Observed strabismus is a secondary consequence of the cortical deficits induced by alternate Occlusion. 20 Using a modified version of the human clinical technique to assess interocular eye alignment, Sherman (1972) noted that at the time of eye opening kittens exhibit a large divergent strabismus and appear to be behaviorally blind. Between the fourth and seventh postnatal weeks the kittens began tO make responses to visual stimuli and also developed normal interocular eye alignment. By the end of the seventh postnatal week the eyes were found to stabilize in the orthophoric position. The time course of development for interocular eye alignment is also coincident with the sensitive or critical period. Prior to the fourth postnatal week the ocular media are observed to be clouded by the presence of a residual hyloid system that seems to be absorbed by the fourth postnatal week. The data of Pettigrew (1972) and Barlow and Pettigrew (1971) suggest that it is highly probable that the "fine tuning" of receptive field disparaties also occurs during this time period. In visually inex- perienced and young kittens the disparity tuning curves cover a wide range (approximately 6°). With increased visual experience the binocular response curves become more peaked approximating the adult response curves. Presumably then, the visual axes stop converging when the images of the visual world are correctly lined up so as to stimulate both retinae in correspondent locations, thus exacting maximum facilitation of cortical neural response properties. However, if alignment were not to completely develop then the fine tuning Of cortical disparity function would probably not develop. 0n the other hand Blakemore and Van Sluyters (1974) have reported that kittens reared in striped cylindrical environments develop normal ocular dominance distributions but also exhibit stribismus. They propose that the repetitive nature of the visual environment 21 provided enough stimulus information to maintain binocular cortical interactions regardless Of the final angle of convergence. However, the receptive field disparities were not examined and binocularly deprived animals also maintain fairly normal ocular dominance histograms. On the other hand behavioral evidence suggests that severe disruption of the binocular field of overlap occurs following binocular deprivation. Sherman (1972) also noted that following bilateral and unilateral lid suture during the critical period that kittens were observed to be strabismic. From these data it would seem that the development of interocular eye alignment may be at least one of the necessary conditions for the complete development or the “fine tuning" of receptive field disparities presumed to underly stereopsis. Secondly, these data suggest that the development of interocular eye alignment is dependent upon simultaneous stimulation of both eyes, as monocular eye lid suture and alternate occlusion are known to disrupt the eye alignment process. Two variables would seem to be related to this developmental process--contour and light-- as eye lid suture is known to severely reduce luminance levels and contours are also assumed to be eliminated. The purpose of this investigation was to assess the relative contribution of light and contour in the develOpment of interocular eye alignment. Kittens were reared in two levels of ambient illumination, and in total darkness to assess the effects of light deprivation on alignment. Within both the light level conditions, half of the kittens in each litter were deprived of contour experience. Each litter was exposed to its respective, restrictive environment for the first three months of life during the critical period. The animals were binocularly deprived so as to provide simultaneous stimulus deprivation. METHOD Subjects Siamese cats are known to be strabismic and to manifest atypical visual projections and cortical interactions (Guillery, 1969; Hubel and Wiesel, 1971; Cool and Crawford, 1970). Care was taken to attempt to eliminate all Siamese animals from the study. Adult females and males were considered to be of non-Siamese genealogy when neither the previous owner's reports nor the animal's physical appearance suggested oterwise. All adults and kittens appeared to be of the tabby variety. Twenty two kittens, from eight litters, were born and raised in the laboratory colony. These animals were deprived of normal visual experience for the first three months of life. During the three month deprivation period each litter was individually housed with the nother cat. Water and Purina Cat Chow (dry) were always available. Puss-n-Boots canned food was used to suppliment the diet. Seven, nondeprived, three month old kittens were vrought into the laboratory only for the assessment of eye alignment. These kittens were from two litters and served as a control group. 142922 Estrus was induced in adult female cats. The eight day injection series of Pregnant Mares Serum Gonadotropin recommended by Colby (1972) was tried but subsequently found to be unnecessary. Instead a single, 2cc, intramuscular injection of the Gonadatropin, was ussually sufficient to stimulate the appropriate mating behaviors. Immediately following the injection, the female was placed in the male's cage. 22 23 Pregnancies terminated within sixty to seventy days. Five of the eight litters were obtained using this mehtod. Three pregnant cats were brought into the colony prior to delivery. At the end of the deprivation period they were returned to their owners. Contact Lenses To examine the role Of contour or pattern experience in the development of interocular eye alignment kittens were binocularly deprived of contour. Half of the kittens from each litter wore diffusing contact lenses. These translucent contact lenses were made from white Teflon stock.1 Each lens was .87mm thick and reduced light transmission by 68.5% (.5 log units density). The teflon lenses were worn by the experimenter and found to be effective in eliminating pattern vision. Littermates of contour deprived kittens wore clear plastic contact lenses tinted to reduce transmission by 68.5% also, with standard contact lense marking dye. Within each litter half Of the kittens were binoc- ularly deprived of pattern experience. The remaining kittens experienced pattern stimuli in the environment while serving as a within litter control for genetics, illumination and the effects of wearing the contact lenses. Twp diameters of contact lenses were used, 10.5 and 12.0mm. The base curve of the contact lenses was 7.5mm (45 diopters) based upon our measurements of corneal curvature as a function of age in kittens (Smolen, Ball and lacks, 1976). The contact lenses were cleaned and stored in Barnes-Hind Cleaning and Soaking solution. Conjunctivitis occured in some of the animals and was treated witn Ophthalmic ointment. 24 Room Illumination To assess the effects of varying degrees of light deprivation on interocular eye alignment, kittens were raised in a) total darkness b) with liminted exposure under scotopic illumination and c) with limited exposure under photOpic illumination. Classification of the low room illumination and high room illumination as scotopic (rod mediated) and photopic (cone mediated) was based upon electrOphysiological data determining the mesopic range, where the transition occurs from rod to cone vision, for single units Of the cat visual system. Using a Stiles two color increment threshold technique, Daw and Pearlman (1969) determined the mesopic range to extend from -1.0 to +1.0 log cd/mz. The rod system saturated at background levels of +1.0 to 1.48 log cd/mz. In good agreement with these data, Hammond and James (1971) found the mesopic range to extend from -1.0 to +1.41 log cd/mz. In both investigat- ions the pupils were fully dilated, using atropine, while the measurements were made. Room illumination in the high illumination condition was 2.45 log cd/mz, measured with a hand held Photometer (Salford Electrical Instruments). with the contact lens transmission reduction, the illumination available to each kitten was reduced to 2.28 log cd/mz. Based upon the electrophsiological data, the level Of illumination was well above the rod saturation point and can be classified as photopic. In the low illumination condition, room illumination was reduced to -.68 log cd/mz. With the contact lenses in place the available illumination at the brightest point in the room was reduced to -.85 log cd/mz, the lower boundary of the mesopic range. Pupil size was not measured but the pupils were probably very close to being fully dilated, producing conditions similar to those of Hammond and James. Thus although the 25 rod system was clearly operable the cone system may have been’juStIat threshold. However, Barlow and Levick (1969) have reported the meSOpic range to extend from -.5 to +1.41 log cd/m2 using a 7mm2 artifical pupil. One explanation offered by Hammond and James to account for the discrepancy between the two sets of data was pupil size. Theoretically then, the mesopic range as defined by a 7mm2 artifical pupil ought to lie over one full log unit above the range determined for the fully dilated pupil (0.0 to 2.41 log cd/mz) instead of -.5 to +1.41 log cd/m2 (Hammond and James, 1971). Since this did not occur a satisfactory explanation is not available. Thus the low illumination can probably be classified, although not unequivocally, as scotopic and was most certainly low mesopic. REARING CONDITIONS PhotOpic Room Illumiantion Four kittens (from two litters) were raised in photopic illumination wearing contact lenses that allowed for contour perception. Two kittens, from two litters, were raised wearing diffusing contact lenses to eliminate contour experience. all contact lenses were worn fbr four hours each day from the third postnatal week until the end of the third month. During this exposure period the kittens remained in the home cage. At the end Of the four hour period the contact lenses were removed, all light was terminated and the kittens remained in total darkness until the next day's exposure period. Thus each kitten experienced only fOur hours of visual stimulation per day. 26 Scotopic Illumiantion Eight kittens (from two litters) were raised in scoptic illumination. Four were exposed to contours and fOur were contour deprived. All contact lenses were worn for four hours each day from the third post natal week until the end of the third month. For the duration of the exposure period, the litter and the mother were removed from the home cage and placed in an adjoining room. The adjacent room contained a caged area in which the animals were free to move about. At the end of the fbur hour period the contact lenses were removed, all light was terminated and the kittens were returned to the home cage. Thus each kitten received only four hours Of visual stimulation per day. Dark Reared Six kittens from two litters were reared in total darkness for three months, from birth. These litters were housed in light tight cages. All feeding and cleaning was done in total darkness. ASSESSMENT OF INTEROCULAR EYE ALIGNMENT A modified version of a clinical technique utilizing the corneal reflex was used to assess interocular eye alignment. This method was similar to the method used by Sherman (1972). At the end of the three month deprivation period each kitten was photographed facing a 200W light source. The light source was used to form the corneal light reflex. (The corneal light reflex is the first Purkinje image and is formed by the reflection of light at the anterior surface of the cornea.) The light source was placed 144cm from the animal. The light. camera and fixation target were all located in the animal's midsagital plane. The fixation target was introduced with a wiggling movement and 27 was most often a feather or a key. The target was located 12cm from the animal and varied slightly in position but was always near to the animal's midline. A ruler was attached to a restraining framework, at eye level, to be used as a magnification standard. Each kitten was suspended in a sling which minimized gross body movements. Head movements were not restricted. The kittens showed no Signs of distress or avoidance of the light Source once they were adapted to the room illumination. Several photographs were taken of each animal for five consecutive days. With the exception of the photographic sessions, the animals were kept in darkness during this period. At the end of the measurement period the kittens were moved to a normal colony room and were kept on a 12-12 hour light-dark cycle for further study. Some of the kittens were retested after one month, 3 months and six months Of normal visual experience to assess the stability of the measurements of interocular eye alignment at the end of the deprivation period. The Dependent Measure From projections of all in-focus photographic negatives the distance between the corneal light reflex and the center of the pupil was measured for each eye and corrected to unit magnification (Figure 1). TO eliminate comparisions between adult and kitten ocular anatomies the distance in mm between the center of the pupil and the corneal reflex in each eye was used as an index fOr interocular eye alignment instead of the degree of deviation for each eye. This measure (the sum of the distance between the center of the pupil and the corneal light reflex in each eye) is more accurately described as the interpupilary distance minus the inter-reflex distance. When the interpupillary distance was 28 greater than the inter-reflex distance the cat would appear to be exotropic ( the eyes would turn slightly outward). Cats normally exhibit a slight divergent (exotropic) strabismus. When the Inter- pupillary distance was less than the reflex, negative in Sign, then the kitten appeared to be esotropic. The eyes were turned inward, indicative Of a convergent strabismus. Because several photographs were taken of each kitten during a test session a mean difference for each animal was calculated. Seven three month Old normal cats were tested in this manner to extablish a range of "normal" interocular eye alignments from which to compare a suspected strabismus. A normal range was established by grouping all of the photographic measurements for all seven normal cats. From these pooled data, a group mean and standard deviation were calculated. Thus an animal was classified as strabismic when that kitten's mean difference (between the interpupillary distance and the inter-reflex difference was significantly greater than or less than the derived normal range. 29 Figure l. The distance (in mm) between the corneal light reflex and the center of the pupil in each eye was measured and summed. This measure described as the interpupilary distance (IPD) minus the inter- reflex distance (IRD) is positive when the animal is divergent and negative when the animal is convergent. Pupil 30 "I COFDGOI Reflex IEQ \ F igure I RESULTS Differences in interocular eye alignment were found in animals that had been deprived of contoured visual experience regardless of the level of room illumination. In Figure 2 the data are shown from the first day of measurement following the end of the deprivation period. The abscissa represents the conditions of deprivation. Proceeding by group from left to right (from normal to dark reared) each panel (A to F) signifies an additional deprivation condition. The ordinate represents the mean difference (in mm) between the interpupillary distance and the inter-reflex distances. The in- dividual data from seven normal, three month old kittens are shown within a derived normal range. The normal range depicted for across group comparisions is the mean interpupillary distance minus the inter- reflex distance for all seven cats :_lSD. These data serve to illustrate the normal variability found to exist as a result of the measurement technique. A divergent strabismus is defined as a mean difference for any animal that is greater than the mean for the normal three month Old group +lSD or greater than a l.3mm difference. A convergent strabismus is defined as a mean difference that is less than the normal mean -lSD or less than a .5mm difference. In panel B the data from four kittens are shown. These animals were reared in high illumination with contoured visual experience (HCl,HC2,HC3,HC4). These kittens were Observed to have mean differences clearly well with in the normal range. The 31 32 Figure 2. The mean interocular eye alignment (interpupillary distance minus inter-reflex distance) is depicted for individual animals :ISE. Each panel (A to F) represents an additional deprivation condition from Normal (N), High illumination with Contour experience (HC), Low illumination with Contour experience (LC), High illumination Contour Deprived (HCD), Low illumination Contour Deprived (LCD) to Dark Reared (OR). The normal range depicted for across group comparisions is the group mean 1150. 33 MEAN IPD-IRD (mm) ab: Im Poi is. ._ W we as No I 823:.» 828:.» W m a a s.aaa a maW a a sass a a a O > m n D m fl i_.o+ \QQSVNN 34 data for this group indicate that the contact lenses did not interfere with the normal development Of interocular eye alignment. In panel C, the data are shown for kittens reared in low illumination with contoured visual experience (LCl,LC2,L03,LC4). These animals show more individual variability in eye alignment. Two kittens LCl and LC4 exhibit mean differences that are within the normal range while LC2 and LC3 seem to be slightly above the range, exhibiting a small divergent strabismus for the target distance of ]2cm. In the fourth panel (0) the data from two (HCDl,HC02) kittens, from two different litters are shown. These animals have mean differences that are clearly outside the normal range, exhibiting a rather large divergent strabismus. Similarly, the kittens that were raised in low illumination and deprived of contoured visual experience and the dark reared kittens are also Observed to have mean differences that are outside the normal range in the direction characteristic of a divergent strabismus (See panels E and F). On the other hand it is interesting to note that data representing animal LC02 in panel E. At first glance this animal appears to be within the normal range. However, two data points are shown for this animal. The data were not averaged into one point as the distribution of eye alignments for this animal was bimodal. The second data point is indicative Of a large convergent strabismus, clearly atypical when compared to both the normal data and the data from the other groups. A two by two analysis of variance (high illumination versus low illumination and contour versus contour deprived ) suggests that contour was the significant factor (p .Ol). Neither light nor an interaction between light and contour was found to be significant (Table I). Comparing the means Of the five experimental groups with the normal group mean Table l summarizes the 2 x 2 anova. contour. (Weiner, 1962) 35 TABLE I The significant factor is Source of variability SS df MS F SSA (light) 0 l 0 55B (contour) 2.496 1 2.496 l3.7l** SSAxB (interaction) -.196 1 -.l96 -l.06 SS within cell 2.16 12 .18 F.0I (I, I?) = 4.75 36 and simultaneously with each other (using Duncan's Multiple Range test) indicates that groups N (normal, HC (high contour) and LC (low contour), are not significantly different from each other (p .01). On the other hand, groups HCD (high contour deprived), LCD (low contour deprived) and DR (darkreared) were found to be a) significantly different from groups N,HC and LC and b) not significantly different from each other (Table II). A qualification is necessary here. The mean difference between groups LCD and LC only approached significance at the .01 level but were found to be significant at the .05 level. A cursory inspection of the data necessitated a check for the homogeneity of variance. The F Max statistic proved to be non-significant, supporting the assumption of homogeneity Of variance. Some genetic comparisions can be made from the data shown in Figure 3. In each panel (A-F) the data are shown for an entire litter. For example, in panel A the data are shown for kittens that were contour experienced and for a kitten that was contour deprived. All three were from the same litter. It is interesting to note that the contour deprived animal appears to be very different from the contour experienced animals. Within each litter this occured. The contour deprived kittens exhibited eye alignments that were clearly different from their littermates who had experienced contour. The second salient point to be made from the data is the relationship between kittens LCl,LC2,LCDl,LC02 (panel 0) and kittens HC3,HC4,HC02 (panel C). These two litters had identical genealogies. The primary feature to be noted in this comparision is that these data have provided, essentially, a genetic control across the light conditions as kittens in panel C were raised in scotopic illumination and kittens in panel 0 were raised in photopic illumination 37 TABLE II Table II summarizes the multiple comparisons between group means. (Duncans Multiple Range, Winer 1969) The groups are abbreviated as follows: N - Normal HC - High Illumination with Contour LC - Low Illumination with Contour LCD- Low Illumination Contour Deprived HCD- High Illumination Contour Deprived DR - Dark Reared N HC LC LCD HCD DR Group Means .94 1.01 1.1 1.88 1.97 2.24 N .94 -——- .07 .16 .94** 1.03** 1.3 ** HC 1.01 -———- .09 .87** .96** 1.23** _ *‘k *‘A‘ ** LC 1.1 (.5g) .87 1.14 LCD 1.88 -——- .09 .36 HCD 1.97 -—-—- .27 OR 2.24 -———— R=2 R=3 R=4 R=5 R=6 critical » qr(2];-—-I 4. 007 4.177 4.292 4.375 4.439 .8014 .8354 .8584 .8750 .8878 T, 2, 3 4, 52 6 38 Figure 3. Within each panel the data are shown for a litter. In panels A and C data are shown for kittens who were raised in high illumination, in panels 8,0 and E the data are shown for kittens reared in low illumination and in panel F the data are depicted for dark reared kittens. Within each panel (A-D) it can be seen that littermates who were contour deprived had eye alignments that were clearly different from animals who were contour experienced. The data presented in panels C and 0 provide a genetic control across light conditions depicting contour deprivation as the significant factor. 39 MEAN IPD-lRD(mm)s ISE — .... N 'o 8 b 'o ‘6' 1 l l l D U D l-O—i r— _ ——n— —— —— — — _ :I D a [—0—] to O — — fi— — —— — — -— I—o-i . O I-O—I _ — q— — — — —— — U I - U I _ _ —b _ — — _ _ I. m —— _ —1— — — — — — I-I—i E '11 F/GUREJ 40 and the same parentage. Again contour deprivation appears to be the factor most related to eye misalignment. Recall that animal LCDZ as seem in Figure 2 was observed to exhibit both a divergent and a convergent strabismus. These comparisions suggest that the observed differences in eye alignment were not a result of any genetic tendencies such as partial Siamese geneology or a tendency towards poor vergence but were related to the experimental conditions and to contour deprivation in particular. The data for all groups on the first day of measurement suggest that a) the contact lenses themselves did not affect the development of interocular eye alignment, b) that even in low illumination exposure to patterned stimuli was sufficient for the development of eye alignment, c) contour deprivation disrupts the eye alignment process and d) the observed differences in eye alignment were not a result of genetic tendencies. Eye Alignment: Test Days 2,3,4 and 5 Subsequent photographic measurements were made to assess eye alignment on days 2,3,4 and 5 following deprivation. With the exception Of testing, the kittens were not exposed to normal visual stimuli during this five day photographic period. In Figure 4 the relationship between each group of kittens by condition is shown for each test day. The mean for each group was obtained by pooling the individual data recorded for each animal on each test day. The same trends seem to exist over the five day test period. Although somewhat variable the high contour kittens as a group and low contour kittens as a group are found to have similar means. The high contour deprived animals, low contour deprived animals and the dark reared kittens as a group exhibit mean differences that 41 Figure 4. Mean interocular eye alignment is depicted as the mean difference betweeen interpupillary distance and inter-reflex distance for each group for days 2,3,4 and 5 of testing. The same trends that existed on day 1 appear to hold over the five day period. The High Cpntour Deprived, pr Qpntour Deprived and Dark Reared kittens show eye alignments that are not normal. 42 MEAN lPD-IRD (mm) D>< N 0:1 . I,— a»: \QQSQNI A 023 J 1. z z r o n r 288m 0»; _ m _J_ o z _. n o LLLL 43 are clearly not normal. The dark reared kittens exhibit a group mean that generally seems to be the largest mean difference as compared to the other groups. It is important to note here that the normal group was only measured during one photographic session, thus its representation on days 2 through 5 is for comparision purposes only. To examine the stability Of eye alignment over the five day period comparisions were made within each group, treating each animal's daily mean difference as a treatment variable. Although impossible to prove, the hypothesis that stability of eye alignment existed over the five day period was generally supported (see Table III). This analysis suggests that individual variability was probably a result Of random fluctuation (measurement error) rather than a consistent trend. Only in the case of the low illumination, contour deprived group was the hypothesis rejected. In Figure 5A each low contour deprived kitten's individual differences for each day are shown. As in Figure 2 animal LC02 is clearly atypical. The overall trend during the five day period at first glance seems to be one of improvement within the normal range. However, with these brief "normal stimulus“ exposure periods during the photographic sessions the occurances of convergent eye alignment were found to increase. If this animal's data had been deleted from the analysis, the hypothesis for stability would have been supported. As a Group, with the exception of LC02 only one kitten, LCDl on day 4 was observed with a mean difference within the normal range. In Figure 58 individual data are depicted for the low illumination contour EXperienced group. As a group a majority of the mean differences were within the normal range. Kittens LC2 and LC3, although borderline cases, were within the normal range for 2 out of the three test days and 4 out of 5 test days respectively. It is also of value in this Each table summarizes the anova for each group as indicated. 44 TABLE III Hi illumination contour Source of variability SS df MS F SS days .24 4 .06 3.00 SS residual .23 12 .02 F.05 (4.12) 3.26 Low illumination contour Source of variability SS df MS F SS days .17 4 .04 .91 SS residual .57 12 .05 F.05 (4.12) 3.26 Hi illumination—contour deprived Source of variability SS df MS F SS days .32 4 .08 .14 SS residual 2.16 4 .54 F.05 (4.4) 6.39 Low illumination-contour deprived (including aninal LC02) Source of variability SS df MS F SS days .97 4 .24 3.43** SS residual .86 12 .07 F.05 (4.12) 3.26 45 TABLE III (cont'd) Low illumination-contour de rived (excluding aninal LCD2 Source Of variability SS df MS F 55 days .93 4 .23 2.56 SS residual .72 8 .09 F.05 (4.8) 3.84 Dark Reared Source of variability SS df MS F SS days .3 3 .l .37 SS residual 4.12 15 .27 F.05 C3-IB)3L29 46 Figure 5. In figure 5 A and 8 individual means are shown over the five day period for kittens reared in Dew illumination. The data shown in A are from Dentour Deprived kittens, the data shown in B are littermates that experienced Dpntour. The normal range is illustrated for comparision. With the exception of C02, the Dpntour Deprived kittens are seen to have abnormal eye alignments compared to both the normal range and to littermates that were Dentour experienced. FMVO 47 MEAN lPD-IRD (mm) nae _ 8 m@ om Pros ___c3_:o=o: I _ s _ . .o. % _mmwwmm_ _ _ I J on + o _ m _ _ _ 38¢me .3 _ W _ _ a 8T s n _ m.ro£___c3.:o=o: No n. on nu _ as I no ououomum " .9 a m a mo o_ _ Joanna 0.0 (9 E 48 Figure to note within litter comparisions. Kittens LCDl,LC02,LCl and LC2 were littermates as were LC03,LCO4 and LC3. As pointed out before, theses genetic data indicate that contour was the significant factor. In Figure 6A and 8 individual data are depicted for the high illu- mination group. In 6A it is very interesting to note the data for HCDB. This kitten developed an eye infection at the end of the second postnatal month and was not able to wear the translucent contact lenses for a two week period. Thus, during those two weeks the kitten was exposed to contoured stimuli for four hours each day. The mean differences for this kitten were clearly within the normal range. In Figure 68 the data are shown for high illumination, contour experienced kittens. Animal H05 and HC03 were littermates. HC5 was also not able to wear its lenses due to an eye infection. Data for all kittens were within the normal range. Figure 7 shows the individual data for the dark reared kittens. (Dark reared 1 through 4 were only tested for four days because of a distemper outbreak in the colony.) Only one observation was within the normal range for all Six kittens. Analysis of variance on day 5 (high illumination versus low illumination and contour versus contour deprived) yielded results similar to the analysis done on day 1 data. Contour was found to be the significant factor (p .01) (Table IV). On the basis of the demonstrated stability over the testing period, all of the data for each animal on each test day was combined. Using Duncan's Multiple range test, comparisions were made between groups. Groups N, LC and HC were not found to be significantly different from each other and groups HC0 and LCD were not significantly different from each other 49 Figure 6. In figure 6 the individual means are shown for each test session for animals raised in ngh illumination. In A the data are shown for kittens that were Dentour Deprived. In B the data are shown for littermates that were exposed to Dentours. Kitten HCD3 exhibits eye alignments that are clearly within the normal range. (See text for details.) 50 Imla m a1. \l_.01 flIfMWWmm m 8. 8» now l\ 93 :. FESzpjoz m . o. am am no am e I a..- .. I N61 mmmw mWW W %Wmm mm m WW0 m 1.... Ia a... a3 a I D>