Ocular Dominance Column Width and Contrast Sensitivity in Monkeys Reared with Strabismus Or Anisometropia

Ocular Dominance Column Width and Contrast Sensitivity in Monkeys Reared with Strabismus or Anisometropia

Morris L. J. Crawford,1 and Ronald S. Harwerth2

PURPOSE. To study the relationship between the width of ocular both eyes. An infant eye that is disadvantaged by having a poor dominance columns in primary visual cortex and spatial con- quality of retinal image (e.g., cataract, strabismus, or anisome- trast sensitivity functions in monkeys with strabismus or aniso- tropia) cannot sustain these synaptic connections and loses metropia during infancy. command over a critical share of cortical neurons, a condition METHODS. Adult monkeys having had monocular visual abnor- sufficient to explain the associated amblyopia. A metabolic and malities induced in infancy were tested behaviorally for spatial anatomic correlate of this synaptic failure is the relative shrinkage of the ocular dominance domain, or column (ODC), asso- contrast sensitivity and then subjected to functional enucle- 7 ation of one eye to reveal the ocular dominance columns ciated with the defective eye. The current report describes a (ODCs) of the primary visual cortex by cytochrome oxidase study to determine whether changes in ODC width are related (CO) staining. The relative widths of the left and right eyes’ to the psychophysical loss of contrast sensitivity (CS) in adult ODCs were measured and related to the contrast sensitivity monkeys that had been subjected to early abnormal binocular functions. vision from experimental strabismus or anisometropia during infancy. RESULTS. The relative widths of the ODCs having input from eyes with strabismic or anisometropic amblyopia were reduced in proportion to the age of onset and the duration of the METHODS early visual abnormality. The relative losses in contrast sensitivity were in ordinal agreement with the losses in relative Six infant macaque monkeys (Macaca mulatta) were used, according width of the ODCs. to the Public Health Service (PHS) Policy on Humane Care and Use of CONCLUSIONS. Amblyopia induced by the early monocular ab- Laboratory Animals (revised, 1986) and the ARVO Statement for the normalities of strabismus or anisometropia is proportional to Use of Animals in Ophthalmic and Vision Research. The research the loss in cortical afference as reflected in the reduction in protocols were reviewed and approved by the University of Houston width of the respective ODCs in the primary visual cortex. and The University of Texas—Houston, Institutional Animal Care and (Invest Ophthalmol Vis Sci. 2004;45:3036–3042) DOI: Use Committees. Monocular anisometropia was induced by having the 10.1167/iovs.04-0029 monkeys wear contact lenses of different power in one eye, thereby inducing a monocular image blur.8 Monkey HIG wore a Ϫ3-D lens ver the past 50 years, visual neuroscience has provided a continuously on the right eye beginning at 14 days of age and lasting Oneural basis for the clinical consequences of visual defects for 60 days. Monkey HT1 wore the same power (Ϫ3-D) lens from 21 to that occur during infancy, a phenomenon well known to the 133 days of age. Monkey ELL wore a Ϫ10-D lens from 14 to 134 days ophthalmic clinician throughout medical history.1 The site of of age. Monkey MIK wore a Ϫ10-D lens from 28 to 88 days of age. the initial neural defect of amblyopia has been shown to be in Surgical esotropia was initiated in the remaining two monkeys. the primary visual cortex2–4 with functional aspects of visual Under ketamine HCl anesthesia, the lateral rectus muscle of the right processing being relatively normal in more peripheral sites.5 eye was extirpated, and a tuck in the medial rectus shortened the The nature of the neural defect is in the failure of the afflicted muscle by ϳ1 mm, inducing esotropia.9 Monkey DAR underwent eye to maintain and strengthen synaptic control over cortical surgery at 1 month of age, whereas monkey TRX was operated on at a neurons. The afferent ocular pathway connections to the input later age (6 months). A caveat regarding this methodology is that it is layer 4C of visual cortex, present at birth,6,7 are normally not known what duration, and to what degree, the esotropia was balanced to divide the dominant control over primary recipient sustained over the next 2 years, as the infants remained with their cortical neurons (and higher order neurons) rather equally mothers in an outdoor colony. However, on delivery to the laboratory between the two eyes, with the majority of cortical neurons at 24 months of age, the degree of residual esotropia was undetectable. being binocular by having synaptic afferent connections from Monkey PT served as a normally reared control (Table 1). After the experimental treatment during infancy, several of the monkeys were trained and tested to measure their spatial CS functions

1 between 2 and 4 years of age, according to our previously described From the Department of Ophthalmology and Visual Science, 10–13 University of Texas Health Science Center at Houston, Houston, Texas; methods. At approximately 5 years of age, the monkeys were and the 2College of Optometry, University of Houston, Houston, Texas. assigned to an experimental glaucoma project where intraocular pres- Supported by Grants R01 EY10608, R01 EY01139, P30 EY11545, sure was raised in one eye, leading (over a period of 6–25 months) to and P30 EY07751, Research to Prevent Blindness, and the Vale-Asche the death of ganglion cells. This functional enucleation permitted Foundation. subsequent marking of the ODCs in the primary visual cortex, using Submitted for publication January 12, 2004; revised May 14 and the histochemical staining technique for cytochrome oxidase (CO) May 18, 2004; accepted May 28, 2004. described by Wong-Riley.14 Disclosure: , None; , None M.L.J. Crawford R.S. Harwerth In preparation for the histochemistry, the brains were ﬂushed with The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertise- saline and ﬁxed in a 4% paraformaldehyde solution, in situ, then ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact. removed and cryoprotected through a sucrose series (10%, 20%, and Corresponding author: Morris L. J. Crawford, Department of Oph- 30%), frozen, and 30-␮m-thick tangential sections of the brain’s visual thalmology and Visual Science, University of Texas Health Science Center cortices were stained for CO, mounted on gelatinized slides, dehy- at Houston, Houston, TX 77030; [email protected]. drated, and coverslipped. The images of the brain tissue were captured

Downloaded from iovs.arvojournals.org on 09/30/2021 IOVS, September 2004, Vol. 45, No. 9 Pathologic Column Width and Contrast Sensitivity 3037

TABLE 1. Summary of the Individual Animal Treatments rons of the normal eye, whereas the interdigitated CO-pale stripes represented input from the experimental eye. Beginning Age of Ending Age of Monkey Treatment Treatment (d) Treatment (d) RESULTS DAR Surgical esotropia 30 — TRX Surgical esotropia 182 — Figure 1A shows the ODCs from layer 4C of the normal mon- ELL Ϫ10-D lens 14 134 key (PT) to be of approximately the same average width, as has MIK Ϫ10-D lens 28 88 Ϫ been reported before for six normal adult cynomolgus mon- HT1 3-D lens 21 133 keys.7 The inset indicates the approximate area from which the HIG Ϫ3-D lens 14 74 PT Normal control sample was taken. The white rectangles illustrate a typical scan path for quantifying the cytochrome oxidase reactivity (COR) in the ODCs of all monkeys. Multiple optical density scans were made of 10 pairs of ODCs in layer 4C of the V1 cortex, through a microscope (Stemi SV11; Carl Zeiss Meditec, Dublin, CA) representing central and paracentral visual space.15–17 Scans microscope with a digital camera (DMC-1; Polaroid, Cambridge, MA) (10 pixels in width) were made orthogonal to the long axis of digital camera with image-analysis software (PhotoShop; Adobe, Moun- the ODC. The scans began in the center of one ODC, traversed tain View, CA). From these images, the ODCs of the V1 layer 4C were the adjacent ODC to end in the center of the next ODC. The easily identiﬁed by the CO-dark stripes representing input from neu- comparison scans were taken at the same location but shifted

FIGURE 1. An illustration of the method of scanning the optical density of the COR over the width of the ODC. Top: the scan path (rectangles) was 10 pixels wide and orthogonal to the long axis of the columns and se- lected to minimize inclusion of the many blood vessels. Bottom: the average proﬁle of 10 scans each over the CO-dark and CO-pale ODC of the normal monkey cortex (PT) aligned for comparison of the relative widths. Note that the columns are of the same average width estimated by the perpendicular line drawn from the half-amplitude density transition from the dark to the pale columns.

Downloaded from iovs.arvojournals.org on 09/30/2021 3038 Crawford and Harwerth IOVS, September 2004, Vol. 45, No. 9

FIGURE 2. Samples of layer 4C ODC patterns from the six experimental monkeys. Insets: the cortical location from which the ODC example was drawn, as well as where the scan profiles of Figure 3 were made. Ex- ample of ODCs for monkeys (A) HIG, (B) HT1, (C) MIK, (D) ELL, (E) DAR, and (F) TRX. The magnification of these images is unspecified and different between panels. The panel order in Figures 2 and 3 are the same.

to start in the adjacent column. Each scan profile was subjected Figure 2 presents sample ODC patterns for the six experi- to a 3-point rolling averaging procedure to reduce pixel-to- mental monkeys. The CO-dark ODC represents input from the pixel variation (the bright holes in the image are empty blood normal eye, whereas the CO-pale ODC represents the input vessels and represent a source of noise). The 10 scans were from the experimental eye. The inset indicates the approxi- normalized to the center of the column, and the average ODC mate cortical area where the samples were taken. In each of density profile (Ϯ1 SD) was then computed. The bottom of the illustrations, the ODC innervated by the defocused or Figure 1 is a graph of the average COR for the left and right deviated experimental eye (pale ODC) was significantly nar- (functional enucleate) eyes of the control monkey. The half- rower than the companion ODC with connections with the amplitude of the slope of the mean density change between normal eye, suggesting a significant shift in the cortical neuro- the CO-dark and the CO-pale ODC (as judged by eye) was used nal control exercised by the two eyes. Because normal ODCs in to define the ODC border and subsequently to determine the macaques are generally of average equal width, the effect of relative ODC widths. Note that the ratio of the averaged right- the experimental manipulation can be quantified by the degree eye ODC width to the averaged left-eye ODC width for the of change in the relative ODC widths. The normalized average normal monkey PT is approximately 1.0. Normally, the area COR density profiles from which the average width of the ODC and relative widths of macaque ODCs in layer 4C representing was estimated are shown in Figure 3. central and paracentral vision are approximately the same in each eye. For example, Horton and Hocking7 found an average Experimental Anisometropia difference between ODCs of 8% Ϯ 4%. In our example, the The lens-reared monkeys were treated for different periods measured average difference in width was less than 2%. starting at different ages, and all the treatments reduced the

Downloaded from iovs.arvojournals.org on 09/30/2021 IOVS, September 2004, Vol. 45, No. 9 Pathologic Column Width and Contrast Sensitivity 3039

FIGURE 3. Average COR profiles from which the relative ODC widths were estimated for the six experimental monkeys. Each graph shows the average density profile for 10 CO- dark (normal eye) and companion CO-pale (experimental eye) columns. The profiles have been normalized in the x-axis for the column centers, with no adjustment in the y-axis dimension. The ratio of widths, estimated from the half-amplitude transition between the CO- dark and CO-pale columns shows the ODC associated with the experimental eye to be narrower than that of the ODC for the normal companion eye. Monkeys (A) HIG, Ϫ3-D blur, 27% loss in relative ODC width; (B) HT1, Ϫ3-D blur, 32% loss in relative ODC width; (C) MIK, Ϫ10-D blur, 53% loss in relative ODC width; (D) ELL, Ϫ10-D blur, 62% reduction in relative ODC width; (E) DAR, surgical esotropia at 3 weeks, 58% reduction in relative ODC width; and (F) TRX, surgical esotropia at 6 months, 40% reduction in relative ODC width.

width of the ODC that had input from the lens-treated eye. visual input to one eye during the first month of life resulted in Experimental anisometropia induced with a Ϫ3-D lens in mon- a permanent large shift in the relative widths of the ODCs in keys HIG and HT1 (Figs. 2A–B) was started at 2 and 3 weeks of favor of the normal eye, with the greater shift in relative width age, respectively. Figure 3A shows that a shorter 60-day period associated with a greater degree of blur. The results of these of relatively mild Ϫ3-D monocular blur resulted in an average four experimental anisometropia experiments suggest that the reduction of 27% of the ODC width for input from the lens- greater the optical blur and the longer the duration of the blur, treated eye. In comparison, Figure 3B shows that the same the greater the shift in the relative widths of ODCs to the magnitude of defocus started 1 week later, but lasted for a detriment of the treated eye—results that are consistent with longer duration (112 days), resulting in a slightly larger differ- our previous behavioral findings.18 ence (32%) in the average ODC width with input from the treated eye. Experimental Surgical Esotropia A more severe monocular defocus, induced by a Ϫ10-D lens in monkeys MIK and ELL (Figs. 3C, 3D), resulted in a signifi- Similar effects were found for two infant monkeys that were cantly larger reduction in width of the ODCs associated with subjected to surgical esotropia at different ages. DAR (Figs 2E, the treated eye. Monkey MIK wore the contact lens from 30 to 3E), made esotropic at 30 days of age, showed a substantial 88 days of age, producing a 53% reduction in ODC width reduction of the ODC width, with a 0.42 ratio for the deviated (shown in Fig. 3C). Monkey ELL (Fig. 3D) who wore the Ϫ10-D to fixating eyes, or a 58% loss in relative width (Fig. 3E). The lens from 14 to 134 days of age had a larger average shrinkage second monkey, TRX, in which surgical esotropia was started of 62% in the associated ODCs. Therefore, defocusing the at a later age (6 months), showed somewhat less effect on the

Downloaded from iovs.arvojournals.org on 09/30/2021 3040 Crawford and Harwerth IOVS, September 2004, Vol. 45, No. 9

FIGURE 4. CS functions (CSFs) for the normal monkey (PT; A) and for four of the experimental monkeys. (A) The monocular CSFs of the left and right eyes of the normal monkey are indicated by the square and circle data points, whereas the diamond data points (A) represent the binocular CSF, showing the superior sensitivity of binocular summation. (B) CSFs for monkey HT1, subjected to Ϫ3-D blur that induced only a modest reduction in peak sensitivity. (C) Monocular blur of Ϫ10 D in MIK induced a severe reduction in CSF at all spatial frequencies. (D) Monocular blur of Ϫ10 D in monkey ELL had the same effect as in (C). (E) CSFs are normal and virtually identical for the two eyes in strabismus monkey TRX.

ODC widths, with a 40% reduction from normal (Fig. 3F). data in Figure 4A. For HT-1, the monkey subjected to mild Assuming that ODCs were of the same average width at the Ϫ3-D monocular blur during infancy (Fig. 4B), the monocular beginning (as shown in Fig. 1) monocular esotropia initiated CS of the treated eye (circles) is only slightly, and insigniﬁ- during the ﬁrst 6 months of life produced a dramatic, perma- cantly, less than the untreated eye (squares)—that is, the sen- nent shift in the relative eye dominance over neurons in pri- sitivities of the two eyes are nearly the same, even though the mary visual cortex. V1 ODC widths were reduced by 32% of normal (see Fig. 3B). Contrast Sensitivity In contrast, both animals reared with Ϫ10-D of defocus demonstrated highly compromised spatial contrast sensitivities Some of the monkeys had been tested for spatial CS of the two over the middle and high spatial frequencies. Figures 4C and eyes at approximately 2 years of age (see Refs. 10–13, 18 for 4D shows the impact of the greater blur by a –10-D lens for methodology). Figure 4 illustrates the CS functions from the MIK and ELL, respectively, in which the large loss in sensitivity, normal monkey (PT; Fig. 4A), in which the peak sensitivity is especially for the higher spatial frequencies, reduced the high approximately 2 cyc/deg and with a high spatial frequency limit of approximately 35 cyc/deg. In addition, the monocular spatial resolution limit from approximately 32 cyc/deg to ap- functions (open circles and squares) are virtually identical, proximately 8 cyc/deg. This change in sensitivity was associ- while the binocular function (diamonds) shows a higher over- ated with a 53% loss in V1 ODC width (Fig. 3C). The results for all sensitivity that is typical of normal binocular summation.11 ELL (Fig. 3D) were similar, for both spatial vision (a relative Therefore, in the normal monkey, equality in ODC width is sensitivity loss from 32–9 cyc/deg.) and in ODC width (a 62% associated with equality in monocular CS and an enhanced loss in ODC width). sensitivity with binocular, compared with monocular, viewing. Figure 3E shows the monocular CS functions from TRX to Figures 4B–E present CS measurements from three of the be virtually identical, despite the demonstrated 40% reduction anisometropic monkeys and one of the strabismic monkeys, of the width of the ODC associated with the surgically deviated TRX, for comparison with the sensitivity of the normal monkey esotropic eye (see the Discussion section).

Downloaded from iovs.arvojournals.org on 09/30/2021 IOVS, September 2004, Vol. 45, No. 9 Pathologic Column Width and Contrast Sensitivity 3041

In summary, these psychophysical results demonstrate that of binocular sensitivity, whereas eye alignment, motor fusion, monocular blur during infancy produces a loss in CS associated and oculomotor responses were within the range of normal with an ordinal reduction in ODC width, although the relation- monkeys. These results suggest that (1) the most important ship between sensitivity loss and neuronal control appears to loss in neuronal control occurred in the “border strips” of the be nonlinear. ODC containing the binocular neurons and (2) that monocular CS functions are largely dependent on functions of neurons in the central monocular “core zone” of the ODC-neuronal losses DISCUSSION sufficient to reduce stereoscopic, but not CS functions (see Horton and Hocking23 for a description of these two compo- There are few data in the literature relating the behavioral nents of the ODC). sensitivity of a mammalian sensory system to the number of Kiorpes et al.24 have compared single-unit recording from cortical neurons supporting that sensitivity. The present exper- V1 cortex with CS functions in strabismic and anisometropic iments show that it is feasible, although additional experiments amblyopic monkeys. Their results were consistent with the will be necessary to define the functional relationships fully. results reported herein. The severity of the amblyopia, as For example, the change in the relative widths of the ODCs can indicated from the CS functions, was related to the degree of be related to the numbers of V1 neurons that serve the two shift in cortical eye dominance (i.e., loss of control over corti- eyes. Hendry et al.19 calculated the average numbers of neu- cal neuronal space). rons beneath a 1-mm2 patch of V1 cortex in five normal adult On the other hand, Fenstemaker et al.25 failed to find a cynomolgus monkeys to be 126,120 Ϯ 5.3/mm2, twice the significant shift in the relative ODC widths in two esotropic neuronal density of tissue in other cortical areas. Assuming that Macaca nemestrina, although they did report changes reflect- Macaca fascicularis and M. mulatta have the comparable ing a lack of normal binocular activation at the boundaries neuronal densities, in their normal hypercolumns (a left and between ODCs. However, they stated that both animals alter- right ODC pair) of approximately 1 mm in width,3 each eye nated in fixation and showed no evidence of amblyopia— within a hypercolumn would have synaptic dominance over factors mitigating a shift in relative ODC widths. slightly more than 60,000 neurons/mm2 (mostly monocular The results of the present study show that ODC widths in neurons in layer 4C). Therefore, a 32% reduction in the width V1 undergo functional shifts, with the normal ODC retaining of an ODC would translate into a functional loss of some control over cortical neuronal space, as there is a complemen- ϳ19,000/mm2 neurons throughout the full cortical thickness. tary decrease in neuronal space influenced by the impaired It is interesting that whereas monkey HT-1 showed a 32% eye. The degree of ODC shift is related not only to the sort of reduction in cortical ODC width (and a presumed loss of visual abnormality, but also to the manner in which the animal control over as many as 19,000 neurons), there was only a responds to the abnormality. For example, the pattern of fix- minor change in CS. In comparison, the more severe blur ation is thought to determine the degree to which this shift in experienced by monkeys MIK and ELL induced nearly twice cortical eye dominance is affected. Adopting a constant unilat- the reduction in ODC space (53% and 62%, and correspond- eral fixation pattern (common in the case of unilateral esotro- ingly, an estimated loss of ϳ32,000 and ϳ37,000 neurons/mm2 pia), the deviating, nonfixating eye most frequently develops throughout the translaminar column) with that loss associated amblyopia. In contrast, monkeys or humans with exotropia with a significantly greater impact on the balance in CS be- and divergent visual axes often adopt an alternating fixation tween the two eyes. One might expect there to be a threshold pattern, thereby stimulating and keeping intact the monocular effect (i.e., that some critical number of neurons would be lost neural connections within the cortex.1,26–28 In the current before a psychophysical change in CS could be detected using experiment using surgical esotropia, the former condition is our threshold methodology). Moreover, the numbers of neu- speculated to have occurred—that is, the monkey (DAR) rons lost before a behavioral defect could be exposed may be adopted a unilateral fixation pattern, inducing a functional quite large. For example, we20 and others,21 have described binocular neuronal disconnect of the esotropic eye from the such a phenomenon for the relationship between the loss of brain, with a concomitant shift in the control of cortical space retinal ganglion cells and the appearance of a visual field defect (the ODC) in favor of the normal eye.23 However, the CSF was in glaucoma, where as many as half the ganglion cells have virtually identical in the two eyes of monkey TRX, consistent become dysfunctional and died before a defect appears in the with a pattern of alternating fixation sufficient to reduce bi- clinical visual field measurement. If a similar rationale is ap- nocular functions, but not sufficient to reduce the function of plied to these experiments on anisometropia, the threshold for the ODC monocular neuronal core. This interpretation empha- a detectable loss in CS must be a functional loss in excess of sizes the spatial specialization of the ODC, where the central ϳ20,000 V1 neurons. With a neuronal loss greater than monocular core supports monocular CS functions and retains ϳ30,000 such neurons, there is a dramatic loss in CS. Obvi- functional integrity in the face of a substantial (Ͼ40%) reduc- ously, additional experiments are required to determine the tion in cortical space, whereas, the binocular zones of the ODC full functional relationship between the numbers of V1 neu- support stereoscopic functions and are readily degraded with rons required for a criterion level of CS. However, these few the loss of cortical space. data point to the clear possibility of the quantification of the The foregoing analysis describes the relative spatial shift in functional relationship between visual sensitivity and the num- eye dominance over neurons of the primary visual cortex and ber of cortical neurons. relates that shift to the psychophysical sensitivity. This spatial The CS results from the esotropic monkey TRX may at first shift is manifest in the relative area of cortex (and conse- glance appear to be incompatible with the change in ODC quently, the numbers of cortical neurons) devoted to each of width (i.e., a large 40% loss in ODC width but no significant the two eyes and does not obligate a change in the spatial change in CS), whereas larger losses of 53% and 62% by frequency of the pattern itself. However, in the literature there monocular image blur showed a decided loss in CS. Therefore, is a continuing and frequent confusion of the shift in relative one might argue that the threshold for altering CS requires a ODC width with a change in ODC spacing. A host of studies neuronal loss of between 40% and 53% of V1 neurons. How- have sought some change in the ODC basic spatial frequency ever, as previously reported,22 TRX (SM184) had severe defects after abnormal ocular conditions during infancy. Some have in functions associated with binocular visual neurons. In that reported an increase in the inter-ODC spacing,29,30 whereas earlier paper, TRX was shown to have significant impairment others have failed to find any such change.31,32 It is important

Downloaded from iovs.arvojournals.org on 09/30/2021 3042 Crawford and Harwerth IOVS, September 2004, Vol. 45, No. 9

to re-emphasize that (as originally described, empirically and 15. Daniel PM, Whitteridge D. The representation of the visual field on theoretically, in several studies2–4,33 [Jones DG, et al. IOVS the cerebral cortex in monkeys. J Physiol Lond. 1961;159:203– 1996;37:ARVO Abstract 1964] from microelectrode record- 221. ings) it is the relative numeric control over cortical visual 16. Rolls ET, Cowey A. Topography of the retina and striate cortex and neurons indicated by the relative cortical area (and ODC its relationship to visual acuity in the rhesus monkeys and squirrel width) that is altered by abnormal early visual experience, and monkeys. Exp Brain Res. 1970;10:298–310. it occurs without an obligate change in ODC spacing. Although 17. Tootell RHB, Switkes E, Silverman MS, Hamilton SLJ. Functional anatomy of macaque striate cortex. II. Retinotopic organization. differences in ODC spatial frequency occur within the cortices J Neurosci. 1988;8:1531–1568. of individual animals,3,34 between animals,3 and between mam- 35 18. Smith EL III, Harwerth RS, Crawford MLJ. Spatial contrast sensitiv- malian species, these variations in ODC spacing have yet to ity deficits in monkeys produced by optically induced anisometro- be incorporated within any theoretical functional framework pia. Invest Ophthalmol. 1985;26:330–342. of visual information processing. 19. Hendry SHC, Schwark HD, Jones EG, Yan J. Numbers and propor- tions of GABA-Immunoreactive neurons in different areas of mon- Acknowledgments key cerebral cortex. J Neurosci. 1987;7:1503–1519. 20. Harwerth RS, Carter-Dawson L, Shen F, Smith EL III, Crawford MLJ. The authors thank Gunter K. von Noorden, MD, and Earl Smith, OD, Ganglion cell losses underlying visual field defects from experi- PhD, for contributions in the treatment and testing of the animals. mental glaucoma. Invest Ophthalmol Vis Sci. 1999;40:2242–2250. 21. Quigley HA, Dunkelberger GR, Green WR. Retinal ganglion cell References atrophy correlated with automated perimetry in human eyes with glaucoma. Am J Ophthalmol. 1989;107:453–464. 1. Von Noorden GK, Campos EC. Binocular Vision and Ocular 22. Harwerth RS, Smith EL III, Crawford MLJ, von Noorden GK. Ste- Motility: Theory and Management of Strabismus. 6th ed. St. reopsis and disparity vergence in monkeys with subnormal binoc- Louis: Mosby; 2002:512–513. ular vision. Vision Res. 1997;37:483–493. 2. Wiesel TN, Hubel DH. Ordered arrangement of orientation col- 23. Horton JC, Hocking DR. Monocular core zones and binocular umns in monkeys lacking visual experience. J Comp Neurol. 1974; border strips in primate striate cortex revealed by the contrasting 158:307–318. effects of enucleation, eyelid suture, and retinal laser lesions on 3. Hubel DH, Wiesel TN, LeVay S. Plasticity of ocular dominance cytochrome oxidase activity. J Neurosci. 1998;18:5433–5455. columns in monkey striate cortex. Philos Trans R Soc Lond B Biol 24. Kiorpes L, Kiper DC, O’Keefe LP, Cavanaugh JR, Movshon JA. Sci. 1977;278:377–409. Neuronal correlates of amblyopia in the visual cortex of macaque 4. LeVay S, Wiesel TN, Hubel DH. The development of ocular dom- monkeys with experimental strabismus and anisometropia. J Neu- inance columns in normal and visually deprived monkeys. J Comp rosci. 1998;18:6411–6424. Neurol. 1980;191:1–51. 25. Fenstemaker SB, Kiorpes L, Movshon JA. Effects of experimental 5. Blakemore C, Vital-Durand F. Effects of visual deprivation on the strabismus on the architecture of macaque monkey striate cortex. development of the monkey’s lateral geniculate nucleus. J Physiol. J Comp Neurol. 2001;438:300–317. 1986;380:493–511. 26. Sireteanu R. Binocular vision in strabismic humans with alternating 6. Rakic P. Prenatal development of the visual system in rhesus fixation. Vision Res. 1982;22:889–896. monkey. Philos Trans R Soc Lond [Biol]. 1977;278:245–260. 27. Steinbach MJ. Alternating exotropia: temporal course of the switch 7. Horton JC, Hocking DR. Intrinsic variability of ocular dominance in suppression. Invest Ophthalmol Vis Sci. 1981;20:129–133. column periodicity in normal macaque monkeys. J Neurosci. 1996; 28. van Leeuwen AF, Collewijn H, de Faber JTHN, van der Steen J. 16:7228–7339. Saccadic binocular coordination in alternating exotropia. Vision 8. Hung LF, Smith EL III. Extended-wear, soft, contact lenses produce Res. 2001;41:3425–3435. hyperopia in young monkeys. Optom Vis Sci. 1996;73:579–584. 29. Lo¨wel S. Ocular dominance column development: strabismus 9. Von Noorden GK, Dowling JE. Experimental amblyopia in changes the spacing of adjacent columns in cat visual cortex. monkeys: II. Behavioral studies in strabismus amblyopia. Arch J Neurosci. 1994;14:7451–7468. Ophthalmol. 1970;84:215–220. 30. Tieman SB, Tumosa N. Alternating monocular exposure increases 10. Harwerth RS, Crawford MLJ, Smith EL III, Boltz RL. Behavioral spacing of ocularity domains in area 17 of cats. Vis Neurosci. studies of stimulus deprivation amblyopia in monkeys. Vision Res. 1997;14:929–938. 1981;21:779–789. 31. Rathjen S, Schmidt KE, Lo¨wel S. Two-dimensional analysis of the 11. Harwerth RS, Smith EL III, Boltz RL, Crawford MLJ, Von Noorden spacing of ocular dominance columns in normally raised and GK. Behavioral studies of the effects of abnormal early visual strabismic kittens. Exp Brain Res. 2002;145:158–165. experience in monkeys: spatial modulation sensitivity. Vision Res. 32. Crawford MLJ. Column spacing in normal and visually deprived 1983;23:1501–1510. monkeys. Exp Brain Res. 1998;123:282–288. 12. Harwerth RS, Smith RL III. The rhesus monkey as a model for 33. Hubel DH, Wiesel TN. Receptive fields of cells in striate cortex of normal vision of humans. Am J Optom Physiol Optics. 1985;62: very young, visually inexperienced kittens. J Neurophysiol. 1963; 633–641. 26:994–1002. 13. Harwerth RS, Smith EL III, Duncan GC, Crawford MLJ, Von Noor- 34. Kaschube M, Wolf F, Puhlmann, et al. The pattern of ocular den GK. Multiple sensitive periods in the development of the dominance columns in cat primary visual cortex: intra- and inter- primate visual system. Science. 1986;232:235–238. individual variability of column spacing and its dependence on 14. Wong-Riley M. Changes in the visual system of monocularly su- genetic background. Eur J Neurosci. 2003;18:3251–3266. tured or enucleated kittens demonstrable with cytochrome oxi- 35. Adams DL, Horton JC. Capricious expression of cortical columns dase histochemistry. Brain Res. 1979;171:11–28. in the primate brain. Nat Neurosci. 2003;6:113–114.

Downloaded from iovs.arvojournals.org on 09/30/2021