Consequences of alternating monocular deprivation on eye alignment and convergence in cats Randolph Blake, M. L. ]. Crawford, and Helmut V. B. Hirsch Four kittens were raised with an opaque contact lens covering one eye for one day and the other eye for the next day, alternating eyes day by day from shortly after birth until six months of age. This rearing procedure severely reduces the proportion of binocularly innervated neurones in visual cortex. Eye alignment measurements, obtained using a corneal reflex technique, disclosed that this neurophysiologic abnormality is accompanied by at least a mild strabismus. Videotape records of eye movements of the restrained cats showed consistent vergence movements upon presentation of an interesting object. These results indicate that: (1) deficits in the neural mechanisms subserving binocular vision can lead to strabismus; and (2) binocular neurones are unnecessary for the initiation of horizontal vergence eye movements. Key words: binocular vision, strabismus, vergence, visual cortex, early visual experience. I n normal cats almost all neurones in visual cortex receive excitatory inputs from both eyes, 1 and by virtue of their unique receptive-field geometry many of these binocular neurones seem particularly From the Department of Ophthalmology, Baylor College of Medicine, and University of Texas Graduate School of Biomedical Sciences, Houston, Texas 77025 (Drs. Blake and Crawford), and the Department of Biological Sciences, State University of New York, Albany, N. Y. 12222 (Dr. Hirsch). This work was supported by a Sigma Xi research grant and a National Institute of Mental Health postdoctoral fellowship to R. B. Manuscript submitted for publication Aug. 2, 1973; manuscript accepted for publication Sept. 6, 1973. Reprint requests: Dr. Randolph Blake, Sensory Sciences Center, University of Texas Graduate School of Biomedical Sciences, 6420 Lamar Fleming Blvd., Houston, Texas 77025. suited for the detection of retinal disparity. 2-3 Evidently early visual experience can crucially modify these binocular connections, for when the normal interrelationship between the eyes is disturbed during the first few postnatal months, as for example by a surgically induced squint, 4 the normal array of binocular neurones is not found. Instead, the majority of cortical neurones are only monocularly excited. At the behavioral level, there are several indications that the eyes of the cat are employed in the interest of binocular vision. For example, we know that the cat's eyes are accurately aligned to provide binocular fixation 5 and that they move disjunctively to maintain this fixation over a range of depths. 6 ' 7 Since in humans proper eye alignment and binocular single vision go hand in hand, it is natural to assume that cats, like most of us, enjoy true binoc- 121
122 Blake, Crawford, and Hirsch Investigative Ophthalmology February 1974 ular vision in all respects. Moreover, it is easy to imagine that binocular neurones in the cat's cortex play a significant role in the promotion and maintenance of stable fusion, a role which has been proposed by Burns and Pritchard 8 as well as others. We have been studying the visual behavior of ordinary, Siamese, and specially reared cats in an attempt to understand the role of binocular striate neurones in vision. In this paper we present results from several orthoptic tests performed on cats reared from birth with each eye alternately occluded on a day by day basis. Hubel and Wiesel, 4 who devised this ingenious but tiresome technique, found that the cortex of cats reared in this manner was normal in every respect except for a marked absence of binocularly innervated neurones. We have found that without exception these specially reared cats manifest a squint while still retaining the ability to verge their eyes when presented with an interesting object at different distances. Method Rearing procedure. Four kittens born in a laboratory breeding colony were reared under conditions of alternating monocular deprivation. When the animal's eyes first opened, one eye was covered with an opaque, scleral contact lens. This lens was left in place for 24 hours, at which time the second eye was also covered by a contact lens, and then the initially covered eye was uncovered. Before inserting a contact lens, ophthalmic ointment was applied to prevent infection and to lubricate the eye. This procedure was repeated at 24 hour intervals until the animals were six months old. During this period the kittens were housed in a continuously illuminated environment, and great care was taken to ensure that one eye always remained covered. As the eyes grew larger the size of the occluder lens was increased appropriately. When the animals reached six months of age, the rearing procedure was terminated, and the cats subsequently were permitted unrestricted visual experience with both eyes open. The integrity of the optics and retina of the eyes was confirmed by ophthalmoscopic and keratometric examinations performed while the cats were lightly anesthetized. Measurements of eye alignment and vergence movements were made within two months following termination of the rearing period. Similar measurements were also taken from a number of normally reared, ordinary cats and from several pure-bred Siamese cats. Corneal reflex technique. Interocular alignment was assessed using the corneal reflex technique described by Sherman. 5 With the cat held in bright sunlight, photographs were made of the eyes with the corneal reflex from the sun centered on one of the constricted pupils. The distance from the film plane to the eyes was at least 50 cm. From enlargements of the photographs, two naive individuals used precision calipers to measure both the distance between the centers of the constricted pupils and the distance of the uncentered corneal reflex from the center of the pupil of that eye. Repeated measurements from the same photograph established a measurement error of less than ± 0.25 mm., and measurements by the two individuals were in excellent agreement. Vergence measurement. Using the techniques of Hughes 6 and Stryker and Blakemore, 7 we have taken changes in the distance between the centers of the two constricted entrance puprls as an index of vergence eye movements. (See Hughes 6 for a thorough discussion of the geometry of this technique.) Videotape records of eye movements were obtained while the cat's body was immobilized in a small box to which was attached a restraining collar which fit snugly around the cat's neck. In this way, we could restrict movement of the head without resorting to surgicauy implanted head restraints. The cats were adapted to the procedure by handling and feeding them in the box for several days prior to data collection. The restrained cat was placed at one end of a viewing tunnel which was brightly illuminated to produce maximum pupil constriction. Mounted 120 cm. from the cat's eyes was a video-camera used to transcribe eye-movement records onto a videotape recorder. Vergence eye movements were elicited by the sudden presentation of a small, live insect tethered and weighted on a thin cord. The insect was lowered at a distance of either 12 or 100 cm. on midline from the cat's eyes. The vigorous wing action of the insect proved to be an excellent stimulus for capturing the cat's attention. The recorded eye movements were displayed on a television monitor, and the horizontal separation between the centers of the constricted entrance pupils was measured directly from the screen using precision calipers. The videotape was always advanced at least 10 frames between successive readings and a minimum of 10 different measurements were taken at each of the two viewing distances. Measurements were taken only when the cat's head was in the frontoparallel plane and when it was noted on the videotape that the cat appeared to be staring at the insect. Electrophysiology. We confirmed the physio-
Volume 13 Number 2 Alternating monocular deprivation on alignment 123 Fig. 1. This series of photographs illustrates the corneal reflex technique {see Reference 5) for classifying eye alignment in cats. In all photographs the corneal reflex from the sun is centered on one of the constricted pupils. (A) Normal eye alignment is characterized by the slightly (1 to 2 mm.) medial position of the uncentered reflex. (B) For esotropia, or convergent strabismus, the reflex appears either centered in both eyes (not illustrated) or slightly lateral in the uncentered eye. (C) In the case of exotropia, or divergent strabismus, the uncentered reflex appears more than 2 mm. medial to the constricted pupil. logic effect of alternating monocular deprivation by recording action potentials from single units in area 17 of one cat reared in this fashion. The cat was anesthetized with ketamine hydrochloride, the eyes dilated with atropine and phenylephrine, the corneas covered with contact lenses, and the refractive state corrected with spectacle lenses. We used an ophthalmoscope to project bright slits or black-white grid patterns alternately onto each eye during direct visual control. A unit was classified as monocular if it fell into Hubel and Weisel's eye-dominance groups one or seven. Using varnished tungsten electrodes, we isolated 43 units in two penetrations into the medial edge of the post-lateral gyrus of the left hemisphere. Of these, 39 units were exclusively monocular (34 contralateral) while the remaining 4 could be driven to at least some degree by stimulation of either eye. Since we were interested only in eye dominance, no attempt was made to map receptive fields. Nonetheless, we are confident, based on directional sensitivity5' and action potential waveforms,10 that our recordings were from cortical units and not axons in the optic radiation. The cat was revived following the experiment and subsequently used in the alignment and vergence tests. Results Eye alignment. Based on observations of the cats in their cages, only one of the specially reared animals showed any hint of strabismus the eye alignment of the other three cats appeared normal. The quantitative measurements of eye alignment, however, disclosed at least a mild strabismus in all four cats when compared with normally reared animals. This comparison can be made in Fig. 1, which il- lustrates the pupil-reflex relationship seen in normal (Fig. 1A), esotropic (Fig. IB), and exotropic (Fig. 1C) cats. Notice in Fig. 1A that normal eye alignment is characterized by the slightly medial placement of the uncentered corneal reflex. This results from the noncorrespondence of the optic axis and visual axis,11 and it is observed in humans as well.12 Because the angle between the visual and optic axes varies among cats, it was necessary to adopt some criterion by which to assess our specially reared cats for strabismus. We used the criterion of Sherman,5 namely to classify as strabismic any cat displaying a pupil-reflex relationship in excess of the range found in normal cats. (Accidental inclusion of an undetected strabismus among our normal cats would serve only to reduce the chance of classifying a specially reared cat as strabismic.) When compared in this way to our 7 normal cats, one specially reared cat was clearly exotropic while the remaining 3 cats were mildly esotropic. These results are based on measurements from at least 8 different photographs of each cat. Because of variations in distance during photography, we have no way of expressing these deviations in degrees. However, based on similar measurements obtained from photographs of two Siamese cats, we can state with confidence that the strabismus of these specially reared cats is some-
124 Blake, Crawford, and Hirsch Investigative Ophthalmology February 1974 o 10 o 15 ae. O o 20 0 _ OD Fig. 2. Angle of convergence estimated for alternaing monocularly deprived ( ), normal (O), and Siamese ( ) cats. These estimates are based upon differences in interpupillary distance during fixation of an interesting object located either 12 cm. or 100 cm. from the eyes of the restrained cat. The arrow at 17 indicates the convergence angle appropriate for near (12 cm.) fixation. what less severe than that observed in two Siamese cats, both of whom possessed a marked esotropia. In repeated measurements over several months we have noticed no change in this pattern of results. Convergence. We obtained videotape records of eye movements from two normal and three specially reared cats, and from one Siamese cat. For all six cats, the distance between the centers of the constricted entrance pupils was consistently smaller during near fixation (12 cm.) than during far fixation (100 cm.). Lateral displacement of the cornea during eye rotation would contribute only a negligible amount to this difference in interpupillary distance (IPD), because the center of rotation of the eye coincides closely to the center of curvature of the cornea. 13 Consequently, the decrease in IPD during the near fixation must be attributed to the medial translation of one or both pupils, i.e., to convergence. Using values from the schematic eye of Vakkur and Bishop, 13 it is possible to estimate for each cat the convergence angle of the eyes during near fixation. We begin by assuming that the eyes rotate about the posterior nodal point, since this point coincides with the center of curvature of the retina. In the schematic eye this point is shown 4.68 mm. behind the anterior surface of the lens, against which the margin of the real pupil is applied. Now as the eyes rotate inward the pupils will be displaced medially; the magnitude of the sum of this displacement for the two eyes is given by the decrease in distance between the entrance pupils, corrected for corneal magnification. For small pupil widths the magnification factor is 1.18 (p. 369, Vakkur and Bishop 13 ). Using the corrected values for the IPD decrease, the estimate of the convergence angle follows from trigonometry. Fig. 2 shows the estimated convergence angle for each cat tested. These estimates represent angular deviations relative to the vergence angle measured during fixation of the object at 100 cm., and they are based on the assumption of symmetrical convergence. This assumption is not critical, however, since the angle estimates, assuming asymmetrical convergence, differ by less than a degree from those in Fig. 2. All cats were approximately the same age and weight when tested. Consequently, any deviations from the schematic eye values should be relatively constant across animals. It is interesting to compare the observed convergence with that which would be required in order for the cat to fixate an object 12 cm. from its eyes. Based on a
Volume 13 Number 2 Alternating monocular deprivation on alignment 125 36 mm. separation of the nodal points of our young cats' eyes, the required angle would be 17, a value achieved by only one of the six cats. These results are similar to those reported by Stryker and Blakemore, 7 who found 14 to be the maximum convergence in their three cats. Discussion Allowing the very young kitten to see with only one eye at a time severely reduces the proportion of binocular neurones characteristic of ordinary cats. By comparing these specially reared cats to normal control cats, we have discovered that this physiologic deficit is accompanied by misalignment of the two eyes. It is interesting to note that similar abnormalities are also seen in the Siamese cat, although in this case the absence of binocular interaction within the cortex 14 ' 15 apparently arises from aberrant visual projections due to misrouting of the optic fibers at the chiasm. 16 These results are consistent with the hypothesis that deficits in the cortical mechanisms subserving binocular vision can produce strabismus, an idea which dates back to the noted British ophthalmologist Claud Worth. 17 We must emphasize, however, that the existence of a normal array of binocular neurones does not necessarily guarantee orthophoria, 18 nor is it entirely clear that the visual cortex alone is involved. 5 With respect to vergence eye movements, the results indicate that the specially reared cats converged their eyes upon presentation of a novel object close to the cat, and the same was true for the Siamese cat. Of course in humans vergence movements can be elicited by a variety of stimuli including binocular disparity, changes in either real or perceived image size, and shifts in accommodation (for a detailed discussion of these, see Alpern 19 ). From our procedure it is impossible to specify which of these cues were responsible for the eye movements observed in the cats. Nevertheless, We can at least state with confidence that a full compliment of binocular neurones is unnecessary for the initiation of horizontal vergence eye movements. Finally, there is another aspect of binocular vision which cannot be ignored and this is stereopsis, the uniquely binocular sense of depth based on retinal disparity. Because different cells have different optimal disparities, binocular neurones in the cat's cortex provide a plausible neural mechanism for stereoscopic vision, a capacity which the cat is known to possess. 20 If this is true, the specially reared cats should suffer a loss in stereopsis. We are currently in the process of testing this prediction. We are grateful to Anthony Wright and Thomas Wheeler for photographic assistance. REFERENCES 1. Hubel, D. H., and Wiesel, T. N.: Receptive fields, binocular interaction and functional architecture in the cat's visual cortex, J. Physiol. 160: 106, 1962. 2. Barlow, H. B., Blakemore, C, and Pettigrew, J. D.: The neural mechanism of binocular depth discrimination, J. Physiol. 193: 327, 1967. 3. Nikara, T., Bishop, P. O., and Pettigrew, J. D.: Analysis of retinal correspondence by studying receptive fields of binocular single units in cat striate cortex, Exp. Brain Res. 6: 353, 1968. 4. Hubel, D. H., and Wiesel, T. N.: Binocular interaction in striate cortex of kittens reared with artificial squint, J. Neurophysiol. 28: 1041, 1965. 5. Sherman, S. M.: Development of interocular alignment in cats, Brain Res. 37: 187, 1972. 6. Hughes, A.: Vergence in the cat, Vision Res. 12: 1961, 1972. 7. Stryker, M., and Blakemore, C: Saccadic and disjunctive eye movements in cats, Vision Res. 12: 2005, 1972. 8. Burns, B. D., and Pritchard, R. : Cortical conditions for fused binocular vision, J. Physiol. 197: 149, 1968. 9. Campbell, F. W;, Cleland, B. C, Cooper, G. F., et al.: The angular selectivity of visual cortical cells to moving gratings, J. Physiol. 198: 237, 1968. 10. Hubel, D. H.: Single unit activity in lateral geniculate body and optic tract of unrestrained cats, J. Physiol. 150: 91, 1960. 11. Bishop, P. O., Kozak, W., and Vakkur, G. J.: Some quantitative aspects of the cat's eye: axis and plane of reference, visual field co-
126 Blake, Crawford, and Hirsch Investigative Ophthalmology February 1974 ordinates, and optics, J. Physiol. 163: 466, 1962. 12. Krimsky, E.: The management of binocular imbalance, Philadelphia, 1948, Lea and Febiger. 13. Vakkur, G. J., and Bishop, P. O.: The schematic eye in the cat, Vision Res. 3:357, 1963. 14. Hubel, D. H., and Wiesel, T. N.: Aberrant visual projections in the Siamese cat, J. Physiol. 218: 33, 1971. 15. Cool, S. J., and Crawford, M. L. J.: Absence of binocular coding in striate cortex units of Siamese cats, Vision Res. 12: 1809, 1972. 16. Guillery, R. W.: An abnormal retinogeniculate projection in Siamese cats, Brain Res. 14: 739, 1969. 17. Worth, C: Squint: its causes, pathology, and treatment, Philadelphia, 1929, Blakiston's and Co. 18. Van Sluyters, R. C, and Blakemore, C: A procedure for the maintenance of cortical binocularity in kittens reared with divergent strabismus, Paper presented at meetings of the Association for Research in Vision and Ophthalmology, Sarasota, Fla., 1973. 19. Alpern, M.: Muscular mechanisms, In: The Eye, Ed. 2. Davson, H., Editor. New York, Academic Press, 1970, vol. 3, pp. 1-252. 20. Fox, R., and Blake, R.: Stereoscopic vision in the cat, Nature 233: 55, 1971.