Online Issues
<< All Back-issues
<< This Issue's Table of Contents
ILAR Journal V40(2) 1999
Animal Models of Human Vision
Thomas T. Norton
| Thomas T. Norton, Ph.D., is Professor in the Department of Physiological Optics, School of Optometry, University of Alabama at Birmingham, Birmingham, Alabama. |
Introduction
As they grow up, approximately 25% of children in the United States become myopic (nearsighted) (Sperduto and others 1983). A much smaller percentage become significantly hyperopic (farsighted), and the majority develop little or no refractive error and thus are "emmetropic" (Sorsby and others 1957; Stenstrom 1948; Strömberg 1936). The causes of refractive error, especially myopia, have been the subject of debate for more than a century. Some have held that myopia is primarily an inherited disorder (Sorsby and others 1962; Steiger 1913; Zadnik 1997) and others, that myopia is caused by protracted near work (Donders 1864), especially by accommodation during protracted near work (Sato 1957; Young 1965). It has not been possible, based solely on clinical observations, to resolve the relative roles of heredity versus environment in the development of refractive error.
In the mid-1970s, several animal models were developed to study the mechanisms underlying refractive error. Using animal models, it was found that the visual environment exerts a powerful influence on refractive state by controlling the axial length of the eye during the postnatal developmental period. Although several species have been examined, three have emerged as primary models and have played complementary roles: chick (Wallman and others 1978b), tree shrew (Sherman and others 1977), and monkey (Wiesel and Raviola 1977). Each model has advantages and disadvantages. Collectively, research on animal models has provided evidence on three specific questions: (1) Can the visual environment produce refractive error? (2) Is there an "emmetropization" mechanism that normally guides eyes to low refractive error? (3) Does excessive accommodation cause myopia? In addition, the two decades of research on animal models have provided criteria that may be used to evaluate the usefulness of particular species as models for emmetropization.
Human Data
In developing and assessing animal models, it is important to describe the human condition that the animals are intended to emulate. A review of the extensive and growing literature concerned with the normal postnatal development of the human eye and the development of refractive error is beyond the scope of this paper. Readers are referred to other reviews (Curtin 1985; Fong 1992; Gordon and Donzis 1985; Goss and others 1988; Jensen 1991; McBrien and Barnes 1984; McCarty and others 1997; Morgan 1967; Saunders 1995a; Saw and others 1996; Sorsby and Leary 1970; Sorsby and others 1957). Two key features that are outlined herein are normal ocular and refractive development and features associated with myopia development in humans.
Normal Human Refractive Development
In Figure 1 are illustrated the optical and axial components of the eye, displaying hyperopic (A), emmetropic (B), and myopic (C) eyes. At birth, the cornea and lens are steeply curved, compared with their adult values, so the focal plane is short (close to the cornea) (Figure 1A). During postnatal development, the cornea and lens mature, causing the focal plane to move away from the cornea (Figure 1, B and C). The axial length of the eye, measured from the front of the cornea to the photoreceptors of the retina, also is short in young eyes such that the retina is typically located in front of the focal plane (Figure 1A), so the eye is hyperopic, although there is considerable variability at birth (Figure 2). By changing the curvature of the crystalline lens through accommodation, the focal plane of a hyperopic eye can be moved toward the retina, reducing the amount of defocus.
As the eye matures, the axial length increases, rapidly at first during the "infantile" high-growth period and then more slowly during the "juvenile" slow-elongation period (Sorsby and others 1961). This moves the retina away from the cornea and toward the focal plane so that, eventually, the axial length matches the focal plane, producing an emmetropic eye that focuses distant objects without accommodation (Figure 1B). Usually the retinal photoreceptors remain at, or slightly in front of, the focal plane, so the eye remains emmetropic or 0.5 to 1.0 diopter (D) hyperopic (Kempf and others 1928; Stenstrom 1948; Strömberg 1936). Often, in humans, the axial length gradually becomes longer than the focal plane (Figure 1C) so that the image of distant objects is focused in front of the retina (myopia). Unlike the situation in an eye with small amounts of hyperopia, myopia cannot be corrected by changing the level of accommodation. An optical correction (glasses with concave [minus-power] lenses, contact lenses, or surgery to flatten the corneal curvature) is needed to move the focal plane posteriorly to the retina.
Studies of refractive error in infants, children, and adults have revealed that the distribution in infants is broad and normally distributed (Cook and Glasscock 1951) with a mean value of approximately 2 D of hyperopia (Figure 2). By 6 to 8 yr of age, when the eye has grown from approximately 17 to 23 mm in axial length (Fledelius and Christensen 1996), the distribution of refractive errors is much narrower (Figure 2), with very few myopes and a peak at 0.5 D of hyperopia (Hirsch and Weymouth 1947; Kempf and others 1928; Sorsby and others 1957). In adults, especially in rural populations, refractive error shows a similar distribution (Lin and others 1988b; Stenstrom 1948; Strömberg 1936; Tron 1929). The distribution is leptokurtic, with many more eyes that are emmetropic or slightly hyperopic than could be accounted for by randomly combining the optical and axial components of the eye. Some of the change toward a narrow distribution, from infancy to adulthood, can be accounted for simply by the "passive emmetropization" (Troilo 1992) that occurs as the eyes enlarge (Wallman and Adams 1987). However, the narrowness of the emmetropic peak, coupled with the fact that the peak contains eyes with long, short, and average axial lengths as well as the high correlation of the axial length with corneal power, led to the suggestion (Sorsby and others 1957) that emmetropia is achieved during development by a coordination between the axial components of the eye and the optical components. Sorsby did not specify which component might be controlled to produce the correlation; based on the animal studies that are reviewed below, it is now apparent that the axial length is controlled to correlate with the optical power of the eye. It has been reported that the rate of decrease of refractive error in infants is related to the magnitude of the initial refractive error (Saunders and others 1995), suggesting that the change toward emmetropia may be visually guided. In an adult eye, a mismatch of axial length by 1 mm would produce approximately a 3 D refractive error. Thus, given the distribution of refractive errors, for the majority of adult human eyes, the axial length matches the focal plane within 0.1 to 0.2 mm.
Human Myopia Development
Excluding congenital myopia, myopia can be divided into two broad categories, the first of which is "physiological," "simple," or "school" myopia. This category includes low to moderate myopic refractive errors up to approximately 6 D (Curtin 1985; Ward and Thompson 1990). In the United States, this type of myopia typically develops after age 6, when the eye has completed the infantile, rapid-growth stage and has reached approximately 95 % of its normal adult axial length. Physiological myopia appears to be primarily a refractive error, without associated ocular pathology. For a myopia of 6 D, an eye is approximately 2 mm longer than if it were emmetropic. However, because of variability in the combined powers of the cornea and lens, the axial length of eyes with up to approximately 6 D of myopia is within the range of axial lengths of emmetropic eyes (Curtin 1985). Thus, an eye with physiological myopia is an eye with an axial length too long for that eye's optical power. The difference that accounts for most of the increased axial length is that the vitreous chamber is longer (deeper) than if that eye were emmetropic. In addition, the anterior chamber is slightly deeper (Fledelius 1982; Zadnik and others 1994) and the lens somewhat thinner than in an emmetropic eye. Animal models appear to emulate physiological myopia.
The second broad category of myopia is "pathological," "progressive," or "degenerative" myopia, which involves an "abnormal lengthening of the eye" (Curtin 1985). This "high" myopia is in excess of approximately 6 D and often progresses to values greater than 10 D. There is evidence in humans that pathological myopia may have a genetic basis (Naiglin and others 1998; Young and others 1998a,b) (see also Curtin 1985). The vitreous chamber of eyes with pathological myopia can become extremely long, the sclera (the posterior covering of the eye) can become very thin, and a bulging of the sclera at the posterior pole (posterior staphyloma) often develops. The elongation of the eye can lead to retinal tears and retinal detachment and, perhaps because of the enlarged globe, is a risk factor for open-angle glaucoma. Thus, myopia, which generally is regarded as an inconvenience, is in fact a major cause of blindness and visual impairment (Curtin 1985), especially among young adults of working age.
Even when myopia does not threaten sight, it imposes limits on career choices (such as entrance into military service academies) and is expensive. In the United States, the cost of glasses, contact lenses, and refractive surgery (which flattens the cornea, moving the focal plane to the retina, but does not affect the elongated length of the eye) has been estimated to be $13 billion annually (Sheedy 1996).
Many studies have revealed that the prevalence of physiological myopia in humans is correlated with factors related to "near work" (Curtin 1985; van Alphen 1961). These factors include education level, intelligence quotient, occupation/socioeconomic level, and amount of time spent reading (Angle and Wissmann 1980; Hirsch 1959; Tay and others 1992; Zylbermann and others 1993). In addition, a higher prevalence of myopia has been reported in East Asia than in the United States and Europe (Lam and Goh 1991; Lin and others 1988a), which may be related to ethnicity and/or intensive schooling associated with preparation for national examinations. The prevalence of myopia appears to increase with the level of educational achievement or amount of study (Dunphy and others 1968; Hirsch 1959; Lin and others 1988b; Septon 1984; Zylbermann and others 1993). These studies have suggested that something involved with near work could induce myopia, at least in some of the population. From the time of Donders (1864), one suggestion has dominated literature on the near-work hypothesis: that near work is a stimulus for accommodation and that "excessive" (that is, prolonged high levels of) accommodation causes myopia (Young 1977). The suggested mechanism has been contraction of the ciliary muscle during accommodation, which increases the power of the crystalline lens, produces increased intraocular pressure (Young 1975), and may cause stretching of the sclera (van Alphen 1961). When placed in near-work situations in which protracted accommodation occurs (as in reading, writing, or studying), this hypothesis posits that the sclera is mechanically stretched, producing a longer, myopic eye.
In addition to data that suggest an environmental cause of myopia, data also show that myopic parents tend to have myopic children. If both parents are myopic, a child has approximately a 30 to 40% probability of becoming myopic. If one parent is myopic, the probability decreases to 20 to 25%; and if neither parent is myopic, the probability is less than 10% (Mutti and others 1996). Studies of twins (Lin and Chen 1987; Sorsby and others 1962) and a recent longitudinal study (Zadnik and others 1994), among many other studies (see Curtin 1985), also found a significant relation between heredity and myopia development. Because the children of myopic parents, including twins, often grow up in an environment similar to that of their parents, and because identical twins might inherit a similar susceptibility to the effects of near work, it has been difficult to distinguish environmental from hereditary factors in the etiology of myopia.
In summary, human eyes are characterized by an infantile, rapid-growth stage until about age 3, and then a juvenile, slow-elongation stage extends through puberty. Refractive error is broadly and normally distributed at birth. By about age 3, the distribution narrows and has a peak at low hyperopia. Simple myopia typically develops during the juvenile stage and is characterized primarily by a vitreous chamber that is longer than the value needed to place the retina at the focal plane. Myopia development is associated with near work and with a family history of myopia.
The practical and ethical necessity to use traditional therapeutic strategies (such as corrective lenses) has precluded most studies in humans aimed at determining the causes of human myopia, although natural history studies aimed at discovering risk factors (Mutti and Zadnik 1995) and studies assessing traditional treatments are being conducted (Ong and others 1998). The development of animal models has provided important new data on the factors in the visual environment that regulate axial length and thereby the refractive state of the eye.
Development of Animal Models
Environmental Induction of Refractive Error
Although Young (1961) had reported the development of myopia in monkeys raised in a visual environment restricted to nearby visual stimuli, and Lauber and colleagues (1965) had discovered in the mid-1960s that chicks could develop an environmentally induced myopia, it was not until the mid-1970s that monkeys (Wiesel and Raviola 1977), tree shrews (mammals closely related to primates) (Sherman and others 1977), chicks (Wallman and others 1978b), and cats (Sommers and others 1978; Wilson and Sherman 1977) were developed as animal models of myopia. As is often the case, the experiments that led to these discoveries of induced myopia in monkeys, cats, and tree shrews were not intended to produce myopia. Rather, they were designed to assess the effects of preventing visual images from reaching the retina (visual form deprivation) on the connections and receptive-field properties of cells in the central visual pathway (Hubel and others 1977; Norton and others 1977; Wilson and Sherman 1977). In each case, the animals were reared from near birth with the eyelids of one eye surgically closed to prevent visual images from being focused on the retina. When the eyelids were opened, it was found that the visual form-deprived eyes were myopic and that the vitreous chamber was elongated relative to the open fellow eye. Wiesel and Raviola (1977) were the first to report this effect as an animal model for human myopia. Because the induced myopia occurs only in the form-deprived eye and not in the fellow control eye, the possibility of a genetic cause is excluded. Potential artifacts associated with the visual deprivation procedure (such as elevated temperature in the deprived eye [Hodos and others 1987]) or the possibility of corneal changes due to the deprivation procedure also have been ruled out or bypassed (Norton 1990; Raviola and Wiesel 1978). Visual form deprivation consistently produces myopia in chicks and tree shrews; however, the amount of myopia (for a given deprivation regimen) varies among individual animals. Macaque monkeys appear more variable (Raviola and Wiesel 1990; Smith 1991; Smith and others 1994; Tigges and others 1990). The within-species variability suggests that individual animals may have a differing (perhaps inherited) susceptibility to developing environmentally induced myopia.
Eyelid closure is a rather primitive technique for producing visual form deprivation because it also sometimes produces other changes, such as alterations in corneal radius (McBrien and Norton 1992; Wildsoet and Pettigrew 1988). An improvement over eyelid closure was the development of systems that allowed translucent diffusers to be held in place with a mask (Hung and others 1994, 1995; Schaeffel and others 1988), glue (Hodos and Kuenzel 1984; Wallman and others 1978a) (later with Velcro) or by a goggle frame clipped to a semipermanent pedestal attached to the skull (Siegwart and Norton 1994). These noncontact methods reduced the likelihood of corneal changes (a reason to avoid using hard contact lenses) and could readily be removed to study the effects of reexposure to form vision (discussed below in Recovery from Induced Myopia).
Importantly, the normal refractive development in most animals appears to parallel human refractive development, and the changes in the eyes of animals with an induced myopia are generally very similar to the changes seen in myopic human eyes. The distribution of refractive values is quite broad at hatching in chicks, at eye opening in tree shrews, and at birth in monkeys. Generally the eyes are hyperopic. With normal development, the eyes progress toward emmetropia, and the variability of refractive values decreases (Norton and McBrien t992; Pickett-Seltner and others 1988; Wallman and others 1981). As illustrated in Figures 1 and 2, eyes generally approach emmetropia from hyperopia. In eyes with an induced myopia, measurements with A-scan ultrasound have shown that vitreous chamber elongation is the predominant change in all species examined. In addition, thinning of the choroid has been observed consistently (McBrien 1998; Norton and Kang 1996; Wallman and others 1995), as in myopic humans (Curtin 1985), along with thinning of the fibrous sclera (Gottlieb and others 1990; Norton and Kang 1996). Other changes, such as in corneal curvature (chicks) (Troilo and others 1995), reduced lens thickness (tree shrews) (McKanna and Casagrande 1978; Norton and Rada 1995), and increased variability in lens power (chicks) (Priolo and others 1998), may be species specific.
The fundamental discovery from visual form deprivation experiments in animals is that the visual environment plays a critically important role in the development of ocular refraction and axial length. Exposure to a "normal" visual environment during postnatal development provides critical information needed for eyes to achieve a normal emmetropic refractive state. When deprived of this information, most eyes elongate past the "proper" axial length where they would have stopped, as judged from their nondeprived fellow control eye; and they thus become myopic.
Evidence of an Emmetropization Mechanism
The concept of an "emmetropization mechanism" implies either that the optical components of the eye (cornea and/or lens) are regulated during postnatal development to place the focal plane on the retina or that the axial elongation of the eye is regulated to place the retina at the focal plane. Studies in animals have provided substantial evidence that axial length (primarily vitreous chamber depth) is actively regulated by a visual feedback mechanism. The precise features of the visual environment used by this mechanism are still undefined, but a common denominator appears to be the amount of defocus on the retina. Data from chicks suggest that high retinal image contrast and midrange spatial frequencies are important (Bartmann and Schaeffel 1994; Schmid and Wildsoet 1997). Several reviews of animal models of emmetropization and myopia have appeared recently (Edwards 1996; Goss and Wickham 1995; Norton and Siegwart 1995; Troilo 1992; Wallman 1995; Wildsoet 1997; Zadnik and Mutti 1995).
Compensation for Minus Lenses. Perhaps the strongest evidence for the active matching of the axial length to the focal plane has come from experiments in which the location of the focal plane was altered by the use of concave or convex lenses that were held in front of one or both eyes with a goggle frame or with another method by which the lens was attached to the head. As shown in Figure 3, a concave (minus-power) lens shifts the focal plane posteriorly, away from the cornea. In an emmetropic eye, this shift produces hyperopic defocus unless the eye accommodates to clear the image. In monkeys, tree shrews, and chicks, minus lens wear (with the lens held in front of the eye rather than with a contact lens that can affect the cornea) produces a compensatory increase in the axial elongation of the eye. Over a period of days, the axial length of the treated eye increases until the retinal location has shifted by an amount that approximately matches the shift of the focal plane. Compensation may be quite accurate (Hung and others 1995; Irving and others 1991, 1995; Siegwart and Norton 1999) and quite rapid (less than 2 wk in chicks and tree shrews), suggesting that there is active, precise regulation of the axial length by the visual environment.
Plus Lenses. When a low-power convex lens (+3 to +5 D) is worn as a spectacle lens by a normal developing eye, the axial elongation rate has been found to decrease in chicks (Irving and others 1995; Schaeffel and others 1988), tree shrews (Siegwart and Norton 1993), and macaque monkeys (Hung and others 1995). As lens wear continues, the eye gradually becomes shorter than it normally would be (although longer than it was at the start of lens wear) and is hyperopic when the lens is removed. Because monkeys and tree shrews appear to be somewhat hyperopic at the age when plus lenses have an effect, it is possible that the effect of the lens is to reduce the hyperopia. To the extent that the normal defocus experienced by a hyperopic eye is a stimulus for elongation, reducing the hyperopia with a lens could reduce or remove the stimulus for the eye to elongate.
When large (+10 to +15 D) plus lenses are used, species differences have been found. As is the case with low-power plus lenses, chicks slow their axial elongation rate and develop large hyperopia, measured when the lens is removed (Irving and others 1995; Schaeffel and others 1988). When exposed to high-power plus lenses, tree shrews develop an elongated myopic eye (Siegwart and Norton 1993). Monkeys respond to binocular treatment with large plus lenses with essentially no response, maintaining the slightly hyperopic refractive state that is present at the start of lens treatment (Smith and others 1999). However, if a series of low-power plus lenses is used, sequentially increasing the power of the lens and allowing time with each lens for the eye to slow its elongation rate and become increasingly hyperopic, a substantial hyperopia can be produced (Smith and others 1999).
The reason for this apparent species difference is a topic that is being actively investigated in several laboratories (Wildsoet 1997). Optically, a plus lens would be expected to produce "myopic defocus" (images in focus in front of the retina, as in Figure 3D). It may be that myopic defocus produces different neural responses (a "stop" signal) in the retina than does hyperopic defocus (which might be a "go" signal) (Rohrer and Stell 1994), allowing the retina to signal that it has moved past the focal plane. This signal, in turn, could produce slowed axial elongation. Species differences may occur because high-power plus lenses produce defocus outside the range that can be distinguished as myopic by the tree shrew and monkey retina (Flitcroft 1999; Smith and others 1999). Because most eyes of most species approach emmetropia from hyperopia and would never experience significant myopic defocus, animals may not have developed a consistent mechanism to distinguish myopic from hyperopic defocus.
An additional factor is that if an eye views nearby objects through the plus lens, the images will be focused on the retina by the lens. Because chicks spend a great deal of time pecking at nearby objects, it may be that very little myopic defocus normally occurs in the eyes of chicks wearing plus lenses. Tree shrews may spend less time examining nearby objects and thus may experience considerable myopic defocus. Recently, Nevin and others (1998) found that chick eyes exposed to high-power plus lenses (producing myopic defocus) became myopic, rather than hyperopic, unless the retina was exposed to clear images for some portion of each day. In contrast, however, Diether and Schaeffel (1998) exposed chicks to consistent myopic defocus and found hyperopic changes. Clearly, this issue remains to be resolved. Although there may be species differences in the response to high-power plus lenses, both the minus lens and low-power plus lens data suggest that the visual environment controls the elongation of the eye so as to place the retina approximately at the focal plane.
Recovery from Induced Myopia. "Recovery" from an induced myopia has been observed in chicks (Wallman and Adams 1987), tree shrews (McBrien and others 1996; Norton 1990; Siegwart and Norton 1998), monkeys (Hung and others 1995), marmosets (Graham and Judge 1999b; Judge and Graham 1995), and kestrels (Andison and others 1992). When an eye has become myopic as a result of form deprivation or minus lens wear, removing the lens while the animal is still maturing produces "recovery." As illustrated in Figure 4, axial elongation essentially stops, but the optics of the eye continue to mature, moving the focal plane toward the retina. This reduces, and even eliminates, the induced myopia. Two lines of evidence suggest that recovery from an induced myopia is visually guided. As noted above, the amount of myopia produced by form deprivation is quite variable across animals within a species, so a broad distribution of myopic refractive errors develops in the deprived eyes. However, after recovery, the variability in refractive error is quite low--all eyes progress toward emmetropia (Siegwart and Norton 1998; Wallman and others 1981). In addition, after myopia has been induced in an eye, if a lens that optically corrects the myopia is placed over the eye, it remains elongated and myopic (measured with the lens removed) (Diether and Schaeffel 1997; McBrien and others 1996). Thus, the recovery data, like the lens data, suggest that a mechanism uses information from the visual environment to actively regulate axial elongation and place the retina at the focal plane.
Communication from Retina to Sclera
Retinal Signal(s). Although it is known that procedures that prevent or degrade the formation of images on the retina, and images with hyperopic defocus, serve as stimuli for increased axial elongation, the specific neural response(s) and even the retinal neurons that are critically involved have not yet been identified. Some aspect of retinal activity appears to be involved. When stroboscopic illumination, or flickering light, is presented to form-deprived eyes, the amount of induced myopia can be reduced compared with non-stroboscopic conditions, depending on the flicker conditions (Gottlieb and Wallman 1987; Schwahn and Schaeffel 1997). The suggestion has been made (Gottlieb and Wallman 1987) that the flickering light stimulates neurons to respond that otherwise would have reduced activity in the absence of form vision. According to this view, retinal neurons in a normal visual environment exhibit some "normal" level or type of activity that is translated into a signal for slow axial elongation. Reduced retinal activity then might be a signal for increased axial elongation (Schwahn and Schaeffel 1997; Wallman 1991).
"Direct" Pathway. Most investigations of retinal transduction, retinal circuitry, and receptive-field organization have examined the visual signals that leave the eye via ganglion cell axons and proceed to central visual structures to produce vision. A rather surprising result from studies in animal models of emmetropization is that in addition, "direct" communication occurs between the retina and the sclera. After surgical, or functional, interruption of ganglion cell output to central structures, the eyes of chicks (Troilo and others 1987; Wildsoet and Pettigrew 1998), monkeys (Macaca mulatta) (Raviola and Wiesel 1990), and tree shrews (Norton and others 1994) can detect the presence of visual form deprivation and elongate to become myopic. Emmetropization still occurs in chicks after optic nerve section, although it is not completely normal (Wildsoet and Wallman 1992). Thus, signals of retinal origin that normally affect axial elongation must reach the sclera without leaving the eye.
Given that, embryologically, the neural retina/retinal pigment epithelium (RPE1) induces condensation of mesenchymeal tissue to form the choroid and then the sclera (Foster and Maza 1994), the existence of this communication pathway may not be surprising. However, it is not known how this communication occurs. This so-called direct pathway presumably involves a series of steps by which signals related to the visual environment (such as the amount of defocus) pass from the neural retina into the RPE, then pass into the choroid, and eventually reach the sclera. The signaling is affected by circadian and/or light/dark cycles (Devadas and Morgan 1996; Gottlieb and others 1992; Nickla and others 1998; Weiss and Schaeffel 1993), dopaminergic systems appear to be involved (Bartmann and others 1994; Schaeffel and others 1994), and recently an involvement of retinoic acid has been suggested (Mertz and others 1998). That communication from the retina to the sclera occurs, however, in no way implies that communication via central pathways is unimportant. In particular, central connections are critical for producing accommodation that may play an important role in emmetropization and myopia development, as discussed below.
Spatially Local Control. Another important discovery arising from research on animal models is that control of axial elongation is spatially local so that one portion of the eye may elongate while another portion remains essentially normal. Chicks (Diether and Schaeffel 1997; Hodos and others 1985; Wallman and others 1987) and tree shrews (Kang and Norton 1993; Siegwart and Norton 1993) exposed to translucent diffusers or to minus lenses that cover only half of the visual field, leaving the other half exposed to a normal visual environment, become elongated and myopic only in the treated hemifield (Diether and Schaeffel 1997; Kang and Norton 1993; Norton and Siegwart 1991; Siegwart and Norton 1993). Treatment with low-power plus lenses produces slowed elongation and hyperopia in the treated visual field (Diether and Schaeffel 1997; Siegwart and Norton 1993). Interestingly, it appears to make little difference whether treatment is applied to the temporal retina (nasal visual field), which contains the area centralis (analogous to the primate fovea), or to the nasal retina (peripheral visual field). Indeed, raising chicks in cages with a low roof, which would be expected to produce images with hyperopic defocus on the inferior retina as the birds focus on the walls or floor while pecking for food, produces a myopic shift in the inferior retina (Miles and Wallman 1990).
It is not surprising that the retina can spatially localize the presence of visual stimuli from different regions in visual space. The localized response of the eye, however, shows that the communication through the RPE and choroid to the sclera must also be spatially restricted, at least to some extent, to the portion of the retina that is spatially adjacent (so that there is a radial flow of information outward from the retina).
The existence of direct, spatially local communication from the retina to the sclera has shown that accommodation is not an essential feature of myopia or hyperopia induced by visual deprivation or lenses. When the retina is disconnected from central visual structures, the stimulus for accommodation is interrupted. In addition, when tetrodotoxin is used to achieve the disconnection, if it reaches the entering ciliary nerve fibers, the neural effectors necessary for accommodation are also affected (McBrien and others 1995; Norton and others 1994). When myopic refractive changes develop in half of an eye, it appears impossible that differing levels of accommodation could occur simultaneously in the two regions to account for the effect. Although emmetropization (and myopia) may occur without accommodation, this in no way implies that accommodation is not involved normally in emmetropization or in the development of refractive errors.
Role of the Choroid. The choroid is a vascular region that provides for the metabolic needs of the RPE, photoreceptors, and sclera. In species with no vasculature on the inner surface of the retina, such as chicks, the choroid may provide for the metabolic needs of a larger fraction of the neural retina and is thus considerably thicker (Buttery and others 1991). Because the rate of blood flow through the choroid is high, it is puzzling how signals related to the visual environment can pass through it from the retina to the sclera. To some extent, the answer appears to lie in the fact that the choroid itself changes in response to the visual environment. This was first discovered in chicks (Wallman and others 1995) but now also has been reported in tree shrews (McBrien and Lawlor 1995; Norton and Kang 1996) and monkeys (Smith and others 1999). In these species, the choroid becomes thinner in response to stimuli that produce increased axial elongation and myopia. During recovery or plus lens wear, it becomes thicker. Because the retina is held against the choroid, choroidal thinning moves the photo-receptors away from the cornea, reducing the amount of hyperopia or producing myopia; thickening of the choroid moves the retina toward the cornea. Changes in the choroidal thickness can be detected very quickly after the onset or removal of visual treatment (Yi and others 1996). Because the choroid is relatively thin in mammals, changes in its thickness do not have a large optical effect, as is possible in chicks. However, the choroid changes are always consistent with the change in axial elongation rate: It becomes thinner during environmentally induced elongation and thicker during slowed elongation.
Scleral Remodeling. In vertebrate eyes, the sclera is a largely extracellular matrix that forms the outer coat of the eye. Generally, it is composed of an outer fibrous layer and an inner cartilaginous layer. The fibrous sclera is comprised primarily of type I collagen fibrils arranged in interwoven lamellae similar to the corneal lamellae, associated proteoglycans, and elastin fibrils. Unlike the cornea, the fibril diameters are quite variable. The cartilaginous sclera is composed of chondrocytes that produce type II collagen and associated cartilage proteoglycans. In eutherian mammals (including tree shrews and primates), the inner cartilaginous layer is absent (Walls 1942).
The fibrous sclera responds in a consistent manner across species in response to stimuli (deprivation or minus lenses) that produce an increase in axial elongation rate and refractive changes toward myopia; it becomes thinner, and there appear to be a loss of material and an upregulation of degradative processes (Guggenheim and McBrien 1996; Kang and Norton 1996; Norton and Rada 1995; Rada and others 1998). With long-term form deprivation, changes in the collagen fibril morphology have been noted that resemble the changes found in the sclera of myopic humans (Avetisov and others 1984; Cornell and McBrien 1994; Curtin and others 1979). However, these changes appear to develop after the induced myopia and may be sequelae rather than causal factors. In tree shrews, both during induced myopia and during recovery from induced myopia, there is no change in the modulus of elasticity of the sclera (a measure of the amount of immediate elongation of a spring-like object in response to an applied force) (Phillips and McBrien 1995; Siegwart and Norton 1999). Rather, the sclera becomes more extensible, reflecting a change in the viscoelastic property of the sclera as measured by the creep rate (elongation rate under constant tension) (Siegwart and Norton 1999). The change in creep rate, coupled with the apparent absence of growth of the fibrous sclera in juvenile tree shrews (Kang and Norton 1996; Norton and Kang 1996; Norton and Miller 1995), has led Norton (1990) and Siegwart and Norton (1999) to suggest that increased remodeling of the sclera, with an upregulation of degradative processes, results in a sclera that is more readily expanded by normal intraocular pressure. However, this hypothesis has not yet been examined in species other than tree shrew.
The inner cartilaginous sclera also shows increased remodeling during the development of an induced myopia. However, in the cartilage there is a net increase in synthesis resulting in growth (Christensen and Wallman 1991; Rada and others 1991) that may be sufficient, in and of itself, to produce axial elongation of the globe. It is not yet known whether the fibrous sclera overlying the cartilage is more extensible in chick. Recent data suggest that the fibrous layer may exert some control over the growth of the cartilaginous layer (Marzani and Wallman 1997). During recovery, the scleral alterations appear to reverse, at least in part, with the fibrous region becoming less extensible and the cartilaginous region showing a reduction in growth rate. Although the phrase "increased growth" has been used frequently to describe the ocular changes associated with an environmentally induced myopia, it appears that this is correctly applied only to the cartilaginous sclera. It also does not appear appropriate to refer to the changes in the fibrous sclera as "increased stretch," which implies a passive extension of the sclera. "Increased extensibility" is perhaps a better phrase. In both the fibrous and cartilaginous regions, the remodeling of the sclera is changed, both during the development of induced myopia and again during recovery, but with an opposite sign (synthesis versus degradation).
In summary, there are many missing pieces in our understanding of how retinal activity is translated into changes in the sclera that can control the size of the juvenile eye. It is a major accomplishment of animal research to have learned that direct, spatially local communication must occur. This route of communication appears likely to be independent of changes in accommodation; however, as discussed below, accommodation appears to play a major role in the overall process of emmetropization.
Role of Accommodation: "Good" or "Bad"
When nearby objects are viewed by an emmetropic eye without accommodation, their image plane is behind the retina. By increasing the optical power of the lens, accommodation moves the image plane toward the retina, clearing the image. Thus, accommodation serves the important role of allowing nearby objects to be seen clearly.
As described above, it has been suggested that the contraction of the ciliary muscle that is required to produce accommodation may also produce tension on the sclera and thus potentially result in sclera stretching, leading to axial elongation and myopia. Based on this "use-abuse" hypothesis, prolonged accommodation is "bad" because it stretches the sclera and causes myopia (Parsons 1906). However, no studies in humans or animals have directly shown increased tension on the sclera in eyes that are developing myopia. Furthermore, studies of animals that develop an induced myopia have clearly shown that accommodation is not needed to produce myopia. Thus, the "scleral stretch" hypothesis, although not disproven, is also not supported by data.
An alternative "defocus" hypothesis has arisen, based partly on data from animal studies and partly on data from children: Accommodation is "good" because it acts to decrease defocus, and something about defocused images stimulates axial elongation. As shown in Figure 5A, underaccommodation to a near target is optically similar to wearing a minus lens, and minus lenses uniformly induce increased axial elongation in animals. To the extent that the eye accommodates accurately to near targets, the image plane is moved to the retina (Figure 5A) and defocus is reduced, removing the stimulus for elongation. According to this reasoning, eyes that accurately accommodate to near targets should remain emmetropic (Figure 5B). As illustrated in Figure 5C, to the extent that the eye underaccommodates to near targets, the retina experiences defocus, and a stimulus for elongation is present. To the extent that the image plane does not move to the retina, the retina moves to the image plane, resulting eventually (Figure 5D) in an eye that is elongated and myopic when viewing distant objects.
It has been well established that human eyes, including the eyes of emmetropes, myopes, and hyperopes, typically underaccommodate (have a "lag of accommodation") in response to near targets. Thus, it must not be essential to completely clear the image of near targets to maintain emmetropia (in other words, if defocus stimulates elongation in human eyes, then some defocus can be tolerated without stimulating elongation). However, Gwiazda and colleagues (1993a,b, 1995) found that children with progressing myopia exhibit a somewhat greater lag of accommodation than do emmetropic children. As modeled quantitatively by Flitcroft (1998), the interaction of the closeness of targets, the amount of time spent looking at them, and the eye's accommodative lag could be important in determining whether an eye elongates and becomes myopic.
The notion that accommodation is good--because it clears the image of nearby objects, allowing them to be seen clearly and also removing a stimulus for axial elongation--is consonant with the age-old suggestion that to avoid myopia, one should read in bright light that decreases pupil diameter and increases depth of focus, thus reducing defocus. The notion is also consistent with the correlation of myopia with intensive schoolwork. Accommodation requires effort (contraction of the ciliary muscle), and the lag of accommodation may increase over many hours of near work. The association between extensive near work and myopia is merely an association between a condition (near work) that tends to produce accommodation. It does not imply that the act of accommodation causes myopia. The data from animal experiments and from children suggest the contrary--that accommodation is good because it prevents excessive defocus.
Both human and animal studies point toward the hypothesis that reducing defocus in children may slow myopia progression. The hypothesis that progressing myopes have greater defocus than nonmyopic children and increased defocus may produce axial elongation has led to two clinical trials. The Correction of Myopia Evaluation Trial ("COMET") is a randomized multicenter trial, funded by the National Eye Institute, that is examining whether children who wear progressive addition lenses, which may reduce the amount of defocus, will show less myopia progression compared with children who wear single-vision lenses (COMET Study Group 1997). Also under way is a randomized clinical trial, likewise funded by the National Eye Institute, examining whether there is an effect of bifocals on myopia progression in esopheric children (Fulk and others 1998). Enrollment is complete in both trials, and results are scheduled to become available in several years.
Emmetropization Model
Based on studies of animals, several models of emmetropization have been published by Schaeffel and colleagues (1988), Wallman (1991), and others (Norton and Siegwart 1995; Shih and others 1989). The model shown in Figure 6 builds on these. The major difference between previous models and the model shown here is that this model separates the effects on the fibrous sclera (increased extensibility, based on results from tree shrews that are congruent with results from humans [Avetisov and others 1984]) from effects on cartilaginous sclera. The starting place in this model, as in others, is the visual stimulus ('T'). When the axial length is shorter than the focal plane (Figure 1A), defocus occurs on the retina unless cleared by accommodation. As discussed above (in Minus Lenses and in Plus Lenses), the components of the visual stimulus that promote axial elongation are not yet well defined. The term "defocus?" is used because other aspects of the visual stimulus may be involved, such as retinal image contrast (Bartmann and Schaeffel 1994). As has been well established by single-unit studies of retinal ganglion cells (Rodieck 1973), defocused images produce weaker responses ("2") from cells with center-surround receptive fields than do sharply focused images. As described above (in Communication from Retina to Sclera), retinally derived signals are communicated ("3") from the retina, through RPE and the choroid, to the sclera via one or more neurohumoral pathways (Nickla and others 1997; Wallman 1991). Communication via central projections and efferent connections to the sclera is also possible. When signals related to visual stimuli that cause increased elongation reach the sclera, remodeling ("4") of the scleral extracellular matrix is produced. In the fibrous sclera, (based on data from tree shrews [Kang and Norton 1996; Norton and Kang 1996; Norton and Miller 1995; Norton and Rada 1995; Siegwart and Norton 1999]), this increased remodeling involves a net loss of extracellular matrix. The fibrous sclera thins and becomes more extensible (scleral "creep rate" increases) ("5"). In species with only a fibrous sclera, the increased extensibility may cause an increase in the axial elongation rate ("6"), based on data obtained solely in tree shrews at the time of this writing. In the cartilaginous sclera (when present), signals for increased elongation have been found to produce an increase in remodeling that has taken the form of active growth (Christensen and Wallman 1991; Rada and others 1991), which also can act to increase the axial elongation rate ("6") of the eye.
In a hyperopic eye, an increased axial elongation rate moves the retina closer to the focal plane. This reduces the visual stimulus for elongation (defocus?), which in turn increases the neural retinal responses. Communication to the sclera is then altered, such that scleral remodeling is affected, decreasing the extensibility of fibrous sclera and decreasing the growth of scleral cartilage. These decreases in turn slow the axial elongation.
A feedback mechanism as described above would produce the gradual approach of the axial length to the focal plane that is characteristic of developing eyes. The rate of approach, the precision with which the retina approximates the focal plane, and whether the retina will stop when the eye remains slightly hyperopic depend on the gain of the feedback loop and other factors such as the depth of focus. By reducing the defocus on the retina, accommodation would reduce the closed-loop gain of the combined feedback system without altering the intrinsic, or open-loop gain, of the ocular elongation feedback system. This emmetropization mechanism would produce compensation for a minus power lens. It also would produce recovery from induced myopia if (1) the visual stimulus for elongation "defocus?" experienced by the myopic eye was less than it would normally experience or (2) myopic defocus on the retina is a stimulus for slowed elongation.
Quantitative models have been developed of the emmetropization mechanism (Flitcroft 1998, 1999; Schaeffel and Howland 1988) or of portions of the feedback system, such as the sclera (Bryant and McDonnell 1998; Grissom and others 1996). For these models, data from studies in animals and humans have been used to generate specific quantitative predictions that can be tested by additional experiments in animals and humans.
Contributions of Animal Models
The most important contribution of experiments using the animal models described in this review is that they have revealed the existence of an active emmetropization mechanism that matches the axial length of the eye to its optical power. This and other fundamental advances toward understanding the role played by the visual environment in the development of refractive error have been possible using animal models due to strict control of the environmental conditions, appropriate experimental controls (including treatment of one eye with the fellow as a control), and sufficient numbers to achieve statistical significance. Invasive, even terminal, experiments can be used to provide analysis of histological, biochemical, and gene expression changes. The ability to produce an environmental change and then measure the ocular changes as they develop is particularly important in separating causal relationships from changes that develop as a consequence of the induced myopia. In contrast, histological, biochemical, and biomechanical studies in human myopic eyes have been based on eye examinations of people who had already developed myopia, in some cases many years earlier, but not of people who have just begun to develop myopia. It is now apparent that changes occur in the sclera within days (Norton and Siegwart 1998) or even hours (Kee and others 1998) after the onset of an environmental change. To understand the sequence of steps by which the visual environment affects the size of the eye, it is very important to examine the eyes at known brief intervals after the onset of visual stimuli that produce axial elongation and to follow the time course of the changes.
Factors in Selecting an Animal Model. Numerous factors are important in selecting an animal model in which to study emmetropization: (1) phylogenetic closeness to humans, (2) maturation rate, (3) similarity of ocular structure to humans, (4) similarity of normal development to humans, (5) ability to study the environmental effects during the juvenile period, (6) ability to control the visual environment precisely, (7) reliable response by the eyes to the environment, (8) presence of myopic changes similar to those in humans, (9) availability of the animals, and (10) presence of accommodative mechanisms similar to humans, such as having a "near triad" response. None of the animal models is perfect. The three species that have been used most---chick, tree shrew, and macaque monkey--illustrate the trade-offs involved in selecting a particular animal model as well as the benefits derived from studying several disparate species.
Chicks. By far and away, chicks (Gallus gallus domesticus) have been used more than any other model, and results from chicks have paved the way in nearly all areas of research on emmetropization. Chicks are readily available and relatively inexpensive, and they mature rapidly. Experiments in chicks thus can use sufficient numbers of animals and produce results within a relatively brief time at a modest expense in animal costs and care. Their normal refractive and ocular development has been examined and found to be similar to, although much faster than, that in humans (Pickett-Seltner and others 1988; Wallman and others 1981). Most experiments have altered the visual environment immediately after hatching (when chicks already have good vision), which appears to be in the infantile growth period; however, it is possible to study the effects of the visual environment at a somewhat later period, more comparable with the period when humans develop physiological myopia (Wallman and Adams 1987). Excellent control of the visual environment is provided by plastic lenses that are attached to the feathers around the eye. Chicks show reliable responses to deprivation and to minus and plus lenses, both in developing induced myopia and in recovery from induced myopia. The primary ocular changes are in the vitreous chamber depth and choroid thickness, although some corneal and lenticular changes have been noted (Priolo and others 1998; Troilo and others 1995). Changes occur very rapidly (within hours) after the visual environment is altered (Kee and others 1998), allowing investigation of the flow of information from the retina, through RPE and the choroid, to the sclera.
The greatest concerns about the chick as an animal model are that they are phylogenetically distant from humans and have numerous differences in ocular structure and physiology. For instance (in common with all models except monkey), chicks have laterally placed eyes and do not have a fovea. They have scleral ossicles that serve as a fulcrum about which they change the shape of the cornea during accommodation (Glasser and others 1994). The ciliary muscle is striated and therefore is innervated via nicotinic, rather than muscarinic, cholinergic receptors (McBrien and others 1993b). In addition to the fibrous sclera that thins during induced myopia, as in humans, chicks have an inner cartilaginous region that actively grows during the development of an induced myopia (Christensen and Wallman 1991; Rada and others 1991). Perhaps because there are no blood vessels on the vitreous surface of the retina, the choroid is much thicker than in primates. In addition, chicks have a substantial, well-developed efferent pathway in the optic nerve projecting to the retina (Crossland and Hughes 1978) that, if present in mammals, is much less well developed. Chicks are capable of independent accommodation in the two eyes (Schaeffel and others 1986).
These differences might lead one to question whether results from the chick could provide useful information about human refractive error. In fact, they have proven to be extremely useful and to provide results that translate to other species. Moreover, some of the cross-species differences have led to discoveries that might not have been made otherwise. For instance, because the choroid is thick in the chick, changes in choroid thickness during myopia induction and recovery have been readily detectable (Wallman and others 1995). Similar changes have subsequently been found in tree shrews and monkeys (and occur in myopic humans), leading to the conclusion that choroid thickness changes are a general phenomenon associated with the communication of information from the retina to the sclera. As another example, because accommodation in chicks is unaffected by atropine sulfate, the finding that atropine administration reduces form deprivation-induced myopia in chicks (McBrien and others 1993b; Stone and others 1991) could not be attributed to blocking accommodation. This finding suggested that the similar effect in mammals might instead be due to effects on the retina.
Tree Shrews. Tree shrews (Tupaia glis belangeri) are small (150 to 200 g) mammals, native to southeast Asia, that are closely related to primates (Lekagul and McNeely 1977). There has been considerable debate about their classification (Luckett 1980). Tree shrews are scansorial, semiarboreal, and highly dependent on vision; they have a cone-dominated retina and are diurnal. They are easy to house, breed readily, and mature quickly (birth to sexual maturity in about 4 mo). With a gestation period of 43 days and a litter size of two to four, young tree shrews are not as readily available as chicks. Although they are altricial at birth and their eyelids do not open until nearly 3 wk after birth, the substantial body of information collected on their subsequent refractive and ocular development has found it to be similar to humans (Norton and McBrien 1992). Tree shrews have an infantile high-growth period and a juvenile slow-elongation period, and they are susceptible to the visual environment during approximately the last 7% of normal axial elongation, similar to the period when most humans develop myopia (Siegwart and Norton 1998). Their ocular structure is similar to humans, although the lens is much thicker and the thin (40 gm) choroid does not contain a choriocapillaris. The goggle frame devised for tree shrews (Siegwart and Norton 1994) provides excellent, precise control of the visual environment. They are an extremely reliable animal model that yields consistent responses to the visual environment, both developing and recovering from induced myopia. The induced myopia is primarily a vitreous chamber elongation, accompanied by thinning of the fibrous sclera and slight lens thinning (McBrien and Norton 1992; McKanna and Casagrande 1978; Norton and Kang 1996; Norton and Rada 1995). Alterations in refractive state, axial elongation rate, and scleral creep rate have been found to occur in tree shrews within 36 hr after the start of minus lens wear or recovery from induced myopia (Siegwart and Norton 1999).
The primary disadvantage of tree shrews is that they are not generally available. For access to juvenile tree shrews, it is necessary to establish a breeding colony. This is difficult because the government of Thailand has not allowed export
of tree shrews since 1980, although tree shrews are plentiful there and elsewhere in southeast Asia. Because they are not primates, tree shrews are not covered by regulations set forth by the Convention on International Trade in Endangered Species of Wild Flora and Fauna ("CITES"). Although breeding colonies exist at universities in North America, Europe, and Australia, obtaining unrelated animals to begin a new colony is difficult.
Tree shrew eyes are relatively small (1/3 the size of humans). Thus, measuring the corneal curvature (~3.5 mm radius) requires modifications to standard keratometers or the use of clinical corneal analyzers that will work with very steep corneas. Streak retinoscopy and clinical infrared autorefractors provide consistent measures of refractive state, and A-scan ultrasound systems have been devised that provide accurate estimates of axial length.
A more important issue in terms of determining the visual stimuli that induce increased axial elongation is the question of whether a "small eye correction" of refractive measures is needed. This issue is of general importance for any animal model with an eye substantially smaller than human eyes. Glickstein and Millodot (1970) proposed that the retinoscopic reflex originates at the inner limiting membrane (the vitreous/ retina interface), rather than at the photoreceptors. This difference would produce an error of approximately 150 to 250 gm in most eyes, inasmuch as the thickness of the retina is relatively stable across species (Buttery and others 1991). In large eyes, such as human, the error is small. In eyes with a relatively short axial length, such as tree shrew, the error is proportionately larger and could be significant. In tree shrews, observed values appear to match Glickstein and Millodot's hypothesis. Measured by streak retinoscopy and with a coincidence optometer while accommodation was blocked with atropine sulfate, tree shrew eyes typically manifest values of ~5 D hyperopia (McBrien and Norton 1992; Norton and McBrien 1992; Siegwart and Norton 1998). These refractive state values are consistent with the eyes being emmetropic and having a 5D "small eye correction" based on calculations made with a simplified mathematical representation of the tree shrew eye (Norton and McBrien 1992). However, in a recent study in rats, Mutti and others (1997) compared streak retinoscopy values with estimates of the refractive state made with cortical visual evoked potentials. The retinoscopy values were slightly, but not significantly (2 D), hyperopic compared with the visual evoked potential values, a much lower difference than the nearly 10 D predicted in rats if the retinoscopic reflex originated at the inner limiting membrane. On this evidence, Mutti and others (1997) have questioned whether a small eye correction is needed. This issue should be resolved for tree shrews to examine further the visual stimuli that are responsible for increased and decreased axial elongation.
Monkeys. Macaque monkeys (M. mulatta) comprise the animal model most closely related to humans. Their optics, ocular structure, and eye size are similar to humans, as are their visual acuity, binocular vision, eye movements, presence of a "near triad" response, color vision, and visual development. With the development of a hood that holds lenses in place (Hung and others 1995), it has become possible to control the visual environment during the infantile developmental period and to avoid the difficulties of eyelid closure and contact lenses (Hung and Smith 1996; Smith and others 1994). The normal pattern of ocular development (Bradley and others 1999) and the ocular changes in eyes with induced myopia are very similar to humans (Raviola and Wiesel 1990). Balanced against these positive factors are several negatives, one of which is that young monkeys are not readily available except through regional primate centers. Monkeys typically have single infants that mature slowly, relative to other animal models, so it is difficult to examine large numbers of animals, and developmental studies proceed slowly. Because monkeys are both strong and dexterous, the young monkeys are hand-raised to prevent the mothers from removing the mask, involving considerable effort on the part of the investigators. An alternative--eyelid closure and leaving the infant monkey with the mother--precludes studies that examine ocular changes within days or hours of the onset, or removal, of visual stimuli that affect axial elongation. !n studies in which visual form deprivation continues for a long period of time, it is difficult to distinguish which ocular changes are causally related to the signaling from retina to sclera and which are consequences of the induced myopia.
A related consequence of the animals' dexterity is that the visual environmental manipulations in monkeys are conducted when the animals are very young and still in the infantile growth stage. The availability of data only from young monkeys has raised questions about the applicability of the results to human myopia (Zadnik and Mutti 1995). However, it has been found that humans with congenital cataracts or ptosis develop axial elongation and myopia rapidly after birth (Hoyt and others 1981; O'Leary and Millodot 1979; Rabin and others 1981), suggesting that human refractive development may be susceptible to the visual environment in the infantile period, as are monkeys (and chicks). Indeed, the relatively low incidence of myopia in children in the United States before age 6 may reflect the absence from the visual environment of "sufficient" near (such as school) work that might stimulate axial elongation, rather than a lack of susceptibility to environmental influences in the infantile period. Nonetheless, it would be useful if a method could be devised for holding lenses or diffusers in a frame on juvenile monkeys.
A final concern in studies that have used monkeys has been the variability of the responses by individual animals to visual form deprivation or exposure to plus or minus lenses. In response to apparently identical manipulations, most of the monkeys in a group respond in a consistent manner, but one or more does not (Hung and others 1995; Tigges and others 1990, 1996). Because it is difficult to study large numbers of monkeys, this variability has been a concern. It may reflect differences in the environmental manipulations that are detected by the monkeys, but not by the investigators, or other factors that are not yet well understood about emmetropization. It also may reflect genetic differences in susceptibility between the animals.
Utility of Examining Multiple Species. As investigators have selected animal species in which to investigate the emmetropization mechanism, most have selected chicks and a smaller number, tree shrews. These studies typically have been built on studies in chicks, revealing that key results from chicks apply to mammals and enabling the identity of differences. Experiments on monkeys have been relatively few and have recently (Hung and others 1995; Smith 1998) emphasized the following: (1) experiments needed to resolve key issues raised by studies on chicks and tree shrews, and (2) experiments involving binocular manipulations, which are particularly important in monkeys because of their frontally placed eyes.
Perhaps the most remarkable feature of the animal studies is the similarity of responses across species. Although it sometimes has taken time for investigators to discover the similarities, given some obvious differences in ocular structure, results from one species have generally translated to the others. For example, the presence of a cartilaginous region in the sclera of chicks, but not in tree shrews, monkeys, and humans, raised concern that the response of chicks might be very different from that of other species. The realization that the outer fibrous region in chicks responds in an analogous manner to the fibrous sclera in other species, along with evidence that the fibrous sclera in chicks may regulate the cartilaginous region (Marzani and Wallman 1997), have largely allayed those concerns. The cross-species similarity in choroidal response is discussed above (in Role of the Choroid). When appropriate experimental controls have been conducted, the apparent species differences in the response to high-power plus lenses may also be resolved.
The many cross-species similarities suggest that the emmetropization mechanism is fundamentally similar in vertebrate eyes. It is difficult to imagine situations in which refractive error would be a selective advantage to a species. It thus is likely that a mechanism to minimize refractive error developed early in vertebrate evolution. Although adaptations of this mechanism presumably have developed along with specializations of the eye, fundamental principles of the emmetropization mechanism very likely remain the same. Thus, investigations of the emmetropization mechanism, and applications to human clinical treatment, may be most successful if they look for common themes across species. As a cautionary note, however, it must be recognized that species differences do exist, and new discoveries and new results cannot be assumed to translate across species without verification.
Other Species. In addition to chick, tree shrew, and monkey, other species described below have been shown to become myopic in response to environmental manipulations: cat, marmoset, rabbit, kestrel, gray squirrel, and fish.
Cats. One of the earliest models was the kitten (Felis catus) (Kirby and others 1982; Ni and Smith 1989; Smith and others 1980; Sommers and others 1978; Wilson and Sherman 1977; Yinon and others 1984) in which visual form deprivation with eyelid closure was found to produce a small but significant myopia and axial elongation. Because kittens are mammals, are readily available, mature rapidly, have ocular characteristics similar to humans, and appear to have normal refractive and ocular development similar to humans, they initially appeared to be a useful animal model. It is thus interesting, and perhaps instructive, that no new study of induced myopia in cats has appeared since the 1980s.
It appears that kittens were abandoned as an animal model for several reasons. One was that only a small amount of myopia developed when eyes were treated with minus spectacle lenses or soft contact lenses and compensation for the shifted focal plane was incomplete. For instance, Ni and Smith (1989) found that treatment with a -10 D soft contact lens, or a -12.5 D spectacle lens, produced approximately 3 D of myopic compensation. Unlike chicks (but similar to tree shrews), plus lenses also produced myopia. However, difficulties maintaining the lenses in place for extended periods required that the kittens remain in the dark for 21 hr per day, which may have affected the amount of compensation. A study using hard contact lenses to shift the focal plane (Nathan and others 1984) revealed no refractive change; however, as subsequent studies in many species have suggested, hard contact lenses may flatten the cornea. After refractive surgery to flatten the cornea, Hendrickson and Rosenblum (1985) noted axial and refractive compensation. Perhaps undeservedly, kittens achieved the reputation of being unreliable animal models and have been abandoned in favor of other species. In addition, cats were the sole animal model that had nocturnal specializations, raising concerns that it was too dissimilar in humans. It is possible that if an improved noncontact goggle system were devised to control the visual environment, and if manipulations of the visual environment were applied later in development during the juvenile period, cats might be a more useful model than currently believed.
Marmosets. Marmosets ( Callithrix jacchus) are small primates in which visual deprivation produces myopia in the treated eye (Troilo and Judge 1993). They have the advantage of being a true primate, yet they are smaller than macaque monkeys and their eyes mature more rapidly (Graham and Judge 1999a). Marmosets have been found to respond to minus lenses with elongation (Graham and Judge 1999b; Judge and Graham 1995). As in other species, atropine sulfate has been reported to block the induced myopia (Judge and Graham 1996). The choroid thins in eyes with an induced myopia (Troilo and others 1998), and there are decreased proteoglycan levels in fibrous sclera of the treated, relative to the control, eye (Rada and Troilo 1998). A concern is that normal cage-reared marmosets were reported to progress toward a small degree of myopia as they matured (Troilo and Judge 1993). In addition, myopia and elongation appear to continue to progress after form deprivation is removed (Rada and Troilo 1998). As the process of establishing the normal development and the details of their responses to visual stimuli is completed, marmosets are likely to become an extremely useful model.
Rabbits. Vo and colleagues (1987) reported that Dutch-belted rabbits (Oryctolagus cuniculus) with induced cataracts developed an elongated vitreous chamber but, because of the cataracts, could not verify that the eyes were myopic. Attempts in my laboratory using eyelid closure (McBrien and Norton, unpublished) and a translucent diffuser held in a goggle frame (Siegwart and Norton, unpublished) have revealed no axial elongation or induced myopia. Recently, Bryant and McDonnell (1997) used flattening of the cornea with an eximer laser as a way to alter the focal plane, similar to placing a minus lens in a goggle frame. They reported an elongation of the vitreous chamber to compensate for the altered corneal power. However, because the cornea tends to regress to its original power, similar to removing a minus lens, it is uncertain whether the elongation is truly visually guided. An advantage of the rabbit is that it is readily available and has a fairly large eye. Because it has not yet been studied thoroughly, it is not known to what extent it will become a useful animal model.
Kestrels. The American kestrel (Falco sparverius), a raptor, is interesting because it is an avian species that is semialtricial at birth compared with the precocial chick. Kestrels are born with their eyes closed, and when they open, they were found by Andison and others (1992) to be myopic. Within a few days, the eyes become emmetropic or slightly hyperopic, apparently because the cornea flattens, perhaps analogous to the corneal flattening found after the removal of eyelid closure in chicks (Wildsoet and Pettigrew 1988). When raised with eyelid closure, kestrels develop an induced myopia (Andison and others 1992). When the eyelids are opened, they recover from the induced myopia.
Gray Squirrels. In their research with the gray squirrel (Sciurus carolinensis), McBrien and others (1993a) found that form deprivation with eyelid closure produced vitreous chamber elongation and myopia. This finding is of interest because the squirrel has no accommodation.
Fish. Kröger and Wagner (1996) found that the eyes of the cichlid fish (Aequidens pulcher) grow to compensate for chromatic aberration. Rearing the fish in monochromatic light and white light, they found that fish raised in short-wavelength light (485 nm) developed eyes significantly shorter than those raised in blue light (623.5 nm). Subsequent "recovery" occurred when the fish were returned to white light. The eye size in fish reared in white light was intermediate between the values expected if only blue-sensitive single or the red- and green-sensitive double cones contributed to the control of eye growth. This result is interesting both because it provides evidence that emmetropization occurs in a very wide range of species and because it suggests that all chromatic channels participate in emmetropization in the fish eye.
Considering the Need for New Animal Models
To develop a new species as a model for the study of ocular development and the emmetropization mechanism requires considerable effort. As described, the normal development and the responses of the eyes to various environmental stimuli must be examined to determine (as has been done in several species) the similarity to, and differences from, human refractive and ocular development. Nonetheless, if a species is identified that appears likely to be a better model--with fewer of the drawbacks than have been found in other species or with a particularly attractive feature not available in other animal models--its development may be worthwhile.
Future Directions
Research using animal models to learn more about the emmetropization mechanism and the causes of refractive error is in a very productive period that is likely to continue and perhaps accelerate. For this progress to occur, it is important that we understand how the visual environment controls axial elongation to produce emmetropia. We must characterize more fully the specific stimuli on the retina that comprise the signal for increased or decreased axial elongation rate. Timing issues (such as how long a stimulus for elongation must be present on the retina and whether it is more effective at various times during the day) must be resolved, and we must understand the temporal interaction between stimuli that increase elongation and stimuli that slow it (Schmid and Wildsoet 1996; Shaikh and others 1999).
Characterizing the visual stimulus is an important step in learning how retinal responses encode these stimuli that signal increased and decreased elongation rate. Continued progress will be made on how the signal moves through the RPE, through the choroid, and into the sclera. The nature of the neural signaling, its timing and duration, and the routes by which circadian and light/dark cycles affect it are also of great importance. To pursue this information, we must strive to answer several key questions: (1) Which neurons respond, and what is the nature of their response? (2) In the sclera, what physically causes the fibrous sclera to become more extensible? (3) When the scleral creep rate increases, what is "creeping"?
In resolving these issues, it will be helpful in most instances and where possible to use minus or plus power lenses, rather than form deprivation, because these lenses provide visual stimuli that merely perturb the emmetropization mechanism, although removing the visual feedback and making it "open loop" with form deprivation can provide a strong stimulus for elongation. In addition, the timing of measurements after the beginning of lens treatment, or its removal, will probably continue to decrease. Early studies used form deprivation that lasted weeks or months. It is increasingly common to find studies that use short-term treatments of days or hours because these allow investigation of the signals for increased or decreased elongation as the eye is actively changing, rather than in eyes that have reached a steady-state condition.
Finally, although animal studies have heretofore examined the effects of the visual environment on emmetropization, it should be noted that animal studies may be useful to study the genetics of myopia. If it is possible to identify animals that consistently produce high-responding or low-responding offspring when exposed to form deprivation or to short-term minus lens wear, it might also be instructive to examine the genetics of these animals.
1Abbreviation used in this paper: RPE, retinal pigment epithelium.
Acknowledgments
The ideas presented in this review result, in part, from 22 yr of involvement with animal models of emmetropization and myopia. However, they have evolved as much from discussions with former and current postdoctoral students in the laboratory and with colleagues at other universities as from specific experiments in this or other laboratories. Myopia research has been, and continues to be, a group effort. Especially important has been the continuing dialogue with Drs. John T. Siegwart, Jr., and Wendy Marsh-Tootle. Very helpful suggestions were provided by two anonymous reviewers. Much of the research on animal models reviewed herein has been supported by grants from the National Eye Institute, a National Institutes of Health (NIH) branch of the US Public Health Service. The tree shrew work reported here was supported by NIH RO1 EY-05922.
References
Andison ME, Sivak JG, Bird DM. 1992. The refractive development of the eye of the American kestrel (Falco sparverius): A new avian model. J Comp Physiol 170:565-574.
Angle J, Wissmann DA. 1980. The epidemiology of myopia. Am J Epidemiol 111:220-228.
Avetisov ES, Savitskaya NF, Vinetskaya MI, Iomdina EN. 1984. A study of biochemical and biomechanical qualities of normal and myopic eye sclera in humans of different age groups. Metab Pediatr Syst Ophthalmol 7:183-188.
Bartmann M, Schaeffel F. 1994. A simple mechanism for emmetropization without cues from accommodation or colour. (Letter). Vision Res 34:873-876.
Bartmann M, Schaeffel F, Hagel G, Zrenner E. 1994. Constant light affects retinal dopamine levels and blocks deprivation myopia but not lens-induced refractive errors in chickens. Vis Neurosci 11:199-208.
Bradley DV, Fernandes A, Lynn M, Tigges M, Boothe RG. 1999. Emmetropization in the rhesus monkey (Macaca mulatta): Birth to young adulthood. Invest Ophthalmol Vis Sci 40:214-229.
Bryant MR, McDonnell PJ. 1997. Axial elongation via optical feedback controlled scleral remodeling. (ARVO Abstract). Invest Ophthalmol Vis Sci 38:S681.
Bryant MR, McDonnell PJ. 1998. Optical feedback controlled scleral remodeling as a mechanism for myopic eye growth. J Theor Bio 1193:613-622.
Buttery RG, Hinrichsen FL, Weller WL, Haight JR. 1991. How thick should a retina be? A comparative study of mammalian species with and without intraretinal vasculature. Vision Res 31:169-187.
Christensen AM, Wallman J. 1991. Evidence that increased scleral growth underlies visual deprivation myopia in chicks. Invest Ophthalmol Vis Sci 32:2143-2150.
COMET Study Group, The. 1997. The design of the Correction of Myopia Evaluation Trial. (Abstract). Optom Vis Sci 74:130.
Cook RC, Glasscock RE. 1951. Refractive and ocular findings in the newborn. Am J Ophthalmol 34:1407-1413.
Cornell LM, McBrien NA. 1994. Alterations in collagen fibril diameter in the sclera of experimentally myopic tree shrew eyes. (ARVO Abstract). Invest Ophthalmol Vis Sci 35:S2068.
Crossland WJ, Hughes CP. 1978. Observations on the afferent and efferent connections of the avian isthmo-optic nucleus. Brain Res 145:239-256.
Curtin BJ. 1985. The Myopias: Basic Science and Clinical Management. Philadelphia PA: Harper & Row.
Curtin BJ, lwamoto T, Renaldo DP. 1979. Normal and staphylomatous sclera of high myopia. Arch Ophthalmol 97:912-915.
Devadas M, Morgan I. 1996. Light controls scleral precursor synthesis. NeuroReport 7:2010-2012.
Diether S, Schaeffel F. 1997. Local changes in eye growth induced by imposed local refractive error despite active accommodation. Vision Res 37:659-668.
Diether S, Schaeffel F. 1998. Severe constant "deprivation" from myopic defocus causes hyperopia and weak "deprivation" from frosted occluders causes myopia in chickens. (ARVO Abstract). Invest Ophthalmol Vis Sci 39:S714.
Donders FC. 1864. On the Anomalies of Accommodation and Refraction of the Eye. Translated by W. D. Moore. London: The New Syndenham Society.
Dunphy EB, Stoll MR, King SH. 1968. Myopia among American male graduate students. Am J Ophthalmol 65:518-521.
Edwards MH. 1996. Animal models of myopia. Acta Ophthalmol Scand 74:213-219.
Fledelius HC. 1982. Ophthalmic changes from age of 10 to 18 years: A longitudinal study of sequels to low birth weight. III. Ultrasound oculometry and keratometry of anterior eye segment. Acta Ophthalmol Suppl 60:393-402.
Fledelius HC, Christensen AC. 1996. Reappraisal of the human ocular growth curve in fetal life, infancy, and early childhood. Br J Ophthalmol 80:918-921.
Flitcroft DI. 1998. A model of the contribution of oculomotor and optical factors to emmetropization and myopia. Vision Res 38:2869-2879.
Flitcroft DI. 1999. The lens paradigm in experimental myopia: Oculomotor, optical and neurophysiological considerations. Ophthalmic Physiol Opt 19:103-111.
Fong DS. 1992. Postnatal ocular growth and its regulation. Int Ophthalmol Clin 32:25-33.
Foster CS, Maza MSdl. 1994. The Sclera. New York NY: Springer-Verlag.
Fulk GW, Cyert LA, Parker DE. 1998. Baseline characteristics in the Myopia Progression Study: A clinical trial of bifocals to slow myopia progression. Optom Vis Sci 75:485-492.
Glasser A, Troilo D, Howland HC. 1994. The mechanism of corneal accommodation in chicks. Vision Res 34:1549-1566.
Glickstein M, Millodot M. 1970. Retinoscopy and eye size. Science 168:605-606.
Gordon RA, Donzis PB. 1985. Refractive development of the human eye. Arch Ophthalmol 103:785-789.
Goss DA, Hampton M J, Wickham MG. 1988. Selected review on genetic factors in myopia. J Am Optom Assoc 59:875-884.
Goss DA, Wickham MG. 1995. Retinal-image mediated ocular growth as a mechanism for juvenile onset myopia and for emmetropization. Doc Ophthalmol 90:341-375.
Gottlieb MD, Joshi HB, Nickla DL. 1990. Scleral changes in chicks with form deprivation myopia. Curr Eye Res 9:1157- l 165.
Gottlieb MD, Nickla DL, Wallman J. 1992. The effects of abnormal light/ dark cycles in the development of form deprivation myopia. (ARVO Abstract). Invest Ophthalmol Vis Sci 33(Suppl):1052.
Gottlieb MD, Wallman J. 1987. Retinal activity modulates eye growth: Evidence from rearing in stroboscopic illumination. Soc Neurosci Abstr 13:1297-1297.
Graham B, Judge SJ. 1999a. Normal development of refractive state and ocular component dimensions in the marmoset (Callithrix jacchus). Vision Res 39:177-188.
Graham B, Judge SJ. 1999b. The effects of spectacle wear in infancy on eye growth and refractive error in the marmoset (Callithrix jacchus). Vision Res 39:189-206.
Grissom JD, Eberhardt AW, Siegwart JT, Norton TT. 1996. Biomechanical modeling of scleral creep in tree shrews with deprivation-induced myopia. Dev Theor Appl Mechanics 18:304-309.
Guggenheim JA, McBrien NA. 1996. Form-deprivation myopia induces activation of scleral matrix metalloproteinase-2 in tree shrew. Invest Ophthalmol Vis Sci 37:1380-1395.
Gwiazda J, Bauer J, Thom F, Held R. 1995. A dynamic relationship between myopia and blur-driven accommodation in school-aged children. Vision Res 35:1299-1304.
Gwiazda J, Thorn F, Bauer J, Held R. 1993a. Emmetropization and the progression of manifest refraction in children followed from infancy to puberty. Clin Vision Sci 8:337-344.
Gwiazda J, Thom F, Bauer J, Held R. 1993b. Myopic children show insufficient accommodative response to blur. Invest Ophthalmol Vis Sci 34:690-694.
Hendrickson P, Rosenblum W. 1985. Accommodation demand and deprivation in kitten ocular development. Invest Ophthalmol Vis Sci 26:343-349.
Hirsch MJ. 1959. The relationship between refractive state of the eye and intelligence test scores. Am J Optom Arch Am Acad Optom 36:12-21.
Hirsch M, Weymouth F. 1947. Notes on ametropia: A further analysis of Stenstrom's data. Am J Optom Arch Am Acad Optom 24:601-603.
Hodos W, Fitzke FW, Hayes BP, Holden AL. 1985. Experimental myopia in chicks: Ocular refraction by electroretinography. Invest Ophthalmol Vis Sci 26:1423-1430.
Hodos W, Kuenzel WJ. 1984. Retinal-image degradation produces ocular enlargement in chicks. Invest Ophthalmol Vis Sci 25:652-659.
Hodos W, Revzin AM, Kuenzel WJ. 1987. Thermal gradients in the chick eye: A contributing factor in experimental myopia. Invest Ophthalmol Vis Sci 28:1859-1866.
Hoyt CS, Stone RD, Frommer C, Billson FA. 1981. Monocular axial myopia associated with neonatal eyelid closure in human infants. Am J Ophthalmol 91:197-200.
Hubel DH, Wiesel TN, LeVay S. 1977. Plasticity of ocular dominance columns in monkey striate cortex. Philos Trans R Soc Lond 278:377-409.
Hung L-F, Crawford MLJ, Smith EL. 1995. Spectacle lenses alter eye growth and the refractive status of young monkeys. Nat Med 1:761-765.
Hung L-F, Smith EL. 1996. Extended-wear, soft, contact lenses produce hyperopia in young monkeys. Optom Vis Sci 73:579-584.
Hung L-F, Smith EL, Crawford MLJ. 1994. Infant monkey eyes can grow to compensate for an optically induced anisometropia. (ARVO Abstract). Invest Ophthalmol Vis Sci 37:S1805.
Irving EL, Callender MG, Sivak JG. 1991. Inducing myopia, hyperopia, and astigmatism in chicks. Optom Vis Sci 68:364-368.
Irving EL, Callender MG, Sivak JG. 1995. Inducing ametropias in hatchling chicks by defocus--Aperture effects and cylindrical lenses. Vision Res 35:1165-1174.
Jensen H. 1991. Myopia progression in young school children--A prospective study of myopia progression and the effect of a trial with bifocal lenses and beta blocker eye drops. Acta Ophthalmol Suppl 200:1-79.
Judge S J, Graham B. 1995. Differential ocular growth of infant marmoset (Callithrix jacchus) induced by optical anisometropia combined with alternating occlusion. (Abstract). J Physiol 485:27P.
Judge SJ, Graham B. 1996. Atropine prevents spectacle-lens-induced refractive changes in the infant common marmoset. (ARVO Abstract). Invest Ophthalmol Vis Sci 37:S326.
Kang RN, Norton TT. 1993. Alteration of scleral morphology in tree shrews with induced myopia. (Abstract). Invest Ophthalmol Vis Sci 34(Suppl): 1209.
Kang RN, Norton TT. 1996. Electronmicroscopic examination of tree shrew sclera during normal development, induced myopia, and recovery. (ARVO Abstract). Invest Ophthalmol Vis Sci 37:S324.
Kee C-S, Marzani D, Wallman J. 1998. Lens-compensation myopia is not the same as form-deprivation myopia. (ARVO Abstract). Invest Ophthalmol Vis Sci 39:S715.
Kempf GA, Collins SD, Jarman BL. 1928. Refractive errors in the eyes of children as determined by retinoscopic examination with a cycloplegic--Results of eye examinations of 1,860 white school children in Washington, D.C. In: Treasury Department, editors. United States Public Health Service. Washington DC: GPO. p. 1-56.
Kirby AW, Sutton L, Weiss H. 1982. Elongation of cat eyes following neonatal lid suture. Invest Ophthalmol Vis Sci 22:274-277.
Kröger RHH, Wagner H-J. 1996. The eye of the blue acara (Aequidens pulcher, Cichlidae) grows to compensate for defocus due to chromatic aberration. J Comp Physiol 179:837-842.
Lam CSY, Goh WSH. 1991. The incidence of refractive errors among schoolchildren in Hong Kong and its relationship with the optical components. Clin Exp Optom 74:97-103.
Lauber JK, McGinnis J, Boyd J. 1965. Influence of mitotics, diamox and vision occluders on light-induced buphthalmos in domestic fowl. Proc Soc Exp Biol Med 120:572-575.
Lekagul B, McNeely JA. 1977. Mammals of Thailand. Bangkok: Kurusapha Ladprao Press.
Lin LLK, Chen CJ. 1987. Twin study on myopia. Acta Genet Med Gemellol 36:535-540.
Lin LLK, Chen C J, Hung PT, Ko LS. 1988a. Nation-wide survey of myopia among schoolchildren in Taiwan 1986. Acta Ophthalmol Suppl 185:29-33.
Lin LLK, Hung PT, Ko LS, Hou PK. 1988b. Study of myopia among aboriginal school children in Taiwan. Acta Ophthalmol Suppl 185:34-36.
Luckett WP. 1980. Comparative Biology and Evolutionary Relationships of Tree Shrews. New York NY: Plenum Press.
Marzani D, Wallman J. 1997. Growth of the two layers of the chick sclera is modulated reciprocally by visual conditions. Invest Ophthalmol Vis Sci 38:1726-1739.
McBrien NA. 1998. Regulation of scleral extracellular matrix metabolism in a mammalian model of axial myopia. (Abstract). Exp Eye Res 67(Suppl 1 ):S263.
McBrien NA, Barnes DA. 1984. A review and evaluation of theories of refractive error development. Ophthalmic Physiol Opt 4:201-213.
McBrien NA, Cottriall CL, Crisp M. 1996. Optical correction of induced axial myopia prevents emmetropization in tree shrews. (ARVO Abstract). Invest Ophthalmol Vis Sci 37:S1000.
McBrien NA, Lawlor P. 1995. Increased proteoglycan synthesis in the sclera of tree shrew eyes recovering from form deprivation myopia. (ARVO Abstract). Invest Ophthalmol Vis Sci 36:S760.
McBrien NA, Moghaddam HO, Cottriall CL, Leech EM, Cornell LM. 1995. The effects of blockade of retinal cell action potentials on ocular growth, emmetropization and form deprivation myopia in young chicks. Vision Res 35:1141-1152.
McBrien NA, Moghaddam HO, New R, Williams LR. 1993a. Experimental myopia in a diurnal mammal (Sciurus carolinensis) with no accommodative ability. J Physiol 469:427-441.
McBrien NA, Moghaddam HO, Reeder AP. 1993b. Atropine reduces experimental myopia and eye enlargement via a nonaccommodative mechanism. Invest Ophthalmol Vis Sci 34:205-215.
McBrien NA, Norton TT. 1992. The development of experimental myopia and ocular component dimensions in monocularly lid-sutured tree shrews (Tupaia belangeri). Vision Res 32:843-852.
McCarty C, Livingston PM, Taylor HR. 1997. Prevalence of myopia in adults--Implications for refractive surgeons. J Refract Surg 13:229-234.
McKanna JA, Casagrande VA. 1978. Reduced lens development in lid-suture myopia. Exp Eye Res 26:715-723.
Mertz JR, Nickla DL, Marzani D, Wallman J. 1998. Visual conditions alter choroidal retinoic acid synthesis potentially altering eye growth. (ARVO Abstract). Invest Ophthalmol Vis Sci 39:S868.
Miles FA, Wallman J. 1990. Local ocular compensation for imposed local refractive error. Vision Res 30:339-349.
Morgan MW. 1967. A review of the major theories for the genesis of refractive state. In: Hirsch M J, editor. Synopsis of the Refractive State of the Eye. Minneapolis MN: Burgess Publishing Company. p 8-12.
Moses RA, Hart WM. 1987. Adler's Physiology of the Eye: Clinical Application. 8th ed. St. Louis MO: Mosby.
Mutti DO, Ver Hoeve JN, Murphy CJ. 1997. The artifact of retinoscopy revisited: Comparison of refractive error measured by retinoscopy and visual evoked potential in the rat. Optom Vis Sci 74:483-488.
Mutti DO, Zadnik K. 1995. The utility of three predictors of childhood myopia: A Bayesian analysis. Vision Res 35:1345-1352.
Mutti DO, Zadnik K, Adams AJ. 1996. Myopia: The nature versus nurture debate goes on. Invest Ophthalmol Vis Sci 37:952-957.
Naiglin L, Clayton J, Gazagne C, Dallongeville F, Malecaze F, Calvas P. 1998. Familial severe myopia a segregation analysis. (ARVO Abstract). Invest Ophthalmol Vis Sci 39:S280.
Nathan J, Crewther SG, Crewther DP, Kiely PM. 1984. Effects of retinal image degradation on ocular growth in cats. Invest Ophthalmol Vis Sci 25:1300-1306.
Nevin ST, Schmid KL, Wildsoet CF. 1998. Sharp vision: A prerequisite for compensation to myopic defocus in the chick? Curr Eye Res 17:322-331.
Ni J, Smith EL. 1989. Effects of chronic optical defocus on the kitten's refractive status. Vision Res 29:929-938.
Nickla DL, Wildsoet C, Wallman J. 1997. Compensation for spectacle lenses involves changes in proteoglycan synthesis in both the sclera and choroid. Curr Eye Res 16:320-326.
Nickla DL, Wildsoet C, Wallman J. 1998. Visual influences on diurnal rhythms in ocular length and choroidal thickness in chick eyes. Exp Eye Res 66:163-181.
Norton TT. 1990. Experimental myopia in tree shrews. In: Bock G, Widdows K, editors. Myopia and the Control of Eye Growth. Chichester: Wiley. p 178-194.
Norton TT, Casagrande VA, Sherman SM. 1977. Loss of Y-cells in the lateral geniculate nucleus of monocularly deprived tree shrews. Science 197:784-786.
Norton TT, Essinger JA, McBrien NA. 1994. Lid-suture myopia in tree shrews with retinal ganglion cell blockade. Vis Neurosci 11:143-153.
Norton TT, Kang RN. 1996. Morphology of tree shrew sclera and choroid during normal development, induced myopia, and recovery. (ARVO Abstract). Invest Ophthalmol Vis Sci 37:S324.
Norton TT, McBrien NA. 1992. Normal development of refractive state and ocular component dimensions in the tree shrew (Tupaia belangeri). Vision Res 32:833-842.
Norton TT, Miller EJ. 1995. Collagen and protein levels in sclera during normal development, induced myopia, and recovery in tree shrews. (ARVO Abstract). Invest Ophthalmol Vis Sci 36:S760.
Norton TT, Rada JA. 1995. Reduced extracellular matrix accumulation in mammalian sclera with induced myopia. Vision Res 35:1271 - 1281.
Norton TT, Siegwart JT. 1991. Local myopia produced by partial visual-field deprivation in tree shrew. (Abstract). Soc Neurosci Abstr 17:558.
Norton TT, Siegwart JT. 1995. Animal models of emmetropization: Matching axial length to the focal plane. J Am Optom Assoc 66:405-414.
Norton TT, Siegwart JT. 1998. Visual regulation of matrix remodeling in mammalian fibrous sclera. (Abstract). Exp Eye Res 67:S263.
O'Leary DJ, Millodot M. 1979. Eyelid closure causes myopia in humans. Experientia 35:1478-1479.
Ong E, Grice K, Held R, Thorn F, Gwiazda J. 1998. Effects of spectacle intervention on the progression of myopia in children. (Abstract). Optom Vis Sci 75:230.
Parsons JH. 1906. The Pathology of the Eye. Vol. 3. New York NY: G. P. Putnam' s Sons.
Phillips JR, McBrien NA. 1995. Form deprivation myopia: Elastic properties of sclera. Ophthalmic Physiol Opt 15:357-362.
Pickett-Seltner RL, Sivak JG, Paternak JJ. 1988. Experimentally induced myopia in chicks: Morphometric and biochemical analysis during the first 14 days after hatching. Vision Res 28:323-328.
Priolo S, Sivak JG, Irving EL, Callender MG, Moore SE. 1998. Effect of experimentally-induced ametropia on the optics of the young chicken lens. (ARVO Abstract). Invest Ophthalmol Vis Sci 39:S505.
Rabin J, VanSluyters RC, Malach R. 1981. Emmetropization: A vision-dependent phenomenon. Invest Ophthalmol Vis Sci 20:561-564.
Rada JA, Slover M, Perry CA. 1998. Expression of gelatinase A (MMP-2) in the sclera of chick eyes during the induction and recovery from form deprivation myopia. (Abstract). Exp Eye Res 67(Suppl 1):S264.
Rada JA, Thoft RA, Hassell JR. 1991. Increased aggrecan (cartilage proteoglycan) production in the sclera of myopic chicks. Dev Biol 147:303-312.
Rada JA, Troilo D. 1998. Proteoglycans in the marmoset sclera are affected by form deprivation. (ARVO Abstract). Invest Ophthalmol Vis Sci 39:S505.
Raviola E, Wiesel TN. 1978. Effect of dark-rearing on experimental myopia in monkeys. Invest Ophthalmol Vis Sci 17:485-488.
Raviola E, Wiesel TN. 1990. Neural control of eye growth and experimental myopia in primates. In: Bock G, Widdows K, editors. Myopia and the Control of Eye Growth. Chichester: Wiley. p 22-39.
Rodieck RW. 1973. The Vertebrate Retina. Principles of Structure and Function. San Francisco: W. H. Freeman and Company.
Rohrer B, Stell WK. 1994. Basic fibroblast growth factor (bFGF) and transforming growth factor beta (TGF-b) act as stop and go signals to modulate postnatal ocular growth in the chick. Exp Eye Res 58:553-562.
Sato T. 1957. The Causes and Prevention of Acquired Myopia. Tokyo: Kanehara Shuppan.
Saunders KJ. 1995. Early refractive development in humans. Surv Ophthalmol 40:207-216.
Saunders KJ, Woodhouse JM, Westall CA. 1995. Emmetropisation in human infancy: Rate of change is related to initial refractive error. Vision Res 35:1325-1328.
Saw SM, Katz J, Schein OD, Chew SJ, Chart TK. 1996. Epidemiology of myopia. Epidemiol Rev 18:175-187.
Schaeffel F, Glasser A, Howland HC. 1988. Accommodation, refractive error and eye growth in chickens. Vision Res 28:639-657.
Schaeffel F, Hagel G, Bartmann M, Kohler K, Zrenner E. 1994.6-Hydroxy dopamine does not affect lens-induced refractive errors but suppresses deprivation myopia. Vision Res 34:143-149.
Schaeffel F, Howland HC. 1988. Mathematical model of emmetropization in the chicken. J Opt Soc Am 5:2080-2086.
Schaeffel F, Howland HC, Farkas L. 1986. Natural accommodation in the growing chicken. Vision Res 26:1977-1993.
Schmid KL, Wildsoet CF. 1996. Effects on the compensatory responses to positive and negative lenses of intermittent lens wear and ciliary nerve section in chicks. Vision Res 36:1023-1036.
Schmid KL, Wildsoet CF. 1997. Contrast and spatial-frequency requirements for emmetropization in chicks. Vision Res 37:2011-2021.
Schwahn HN, Schaeffel F. 1997. Flicker parameters are different for suppression of myopia and hyperopia. Vision Res 37:2661-2673.
Septon RD. 1984. Myopia among optometry students. Am J Optom Physiol Opt 61:745-751.
Shaikh AW, Siegwart JT, Norton TT. 1999. The effect of interrupted lens wear on compensation for a minus lens in tree shrews. Optom Vis Sci (Forthcoming).
Sheedy JE. 1996. What is the role of glasses in optometry? Optom Educ 21:111-113.
Sherman SM, Norton TT, Casagrande VA. 1977. Myopia in the lid-sutured tree shrew (Tupaia glis). Brain Res 124:154-157.
Shih YF, Lin LL, Peng Y, Ko LS. 1989. Clinical observations on occlusion myopia. J Formos Med Assoc 88:164-168.
Siegwart JT, Norton TT. 1993. Refractive and ocular changes in tree shrews raised with plus or minus lenses. (ARVO Abstract). Invest Ophthalmol Vis Sci 34:S1208.
Siegwart JT, Norton TT. 1994. Goggles for controlling the visual environment of small animals. Lab Anim Sci 44:292-294.
Siegwart JT, Norton TT. 1998. The susceptible period for deprivation-induced myopia in tree shrew. Vision Res 38:3505-3515.
Siegwart JT, Norton TT. 1999. Regulation of the mechanical properties of tree shrew sclera by the visual environment. Vision Res 39:387-407.
Smith EL. 1991. Experimentally induced refractive anomalies in mammals. In: Grosvenor T, Flom MC, editors. Refractive Anomalies. Boston MA: Butterworth-Heinemann. p 246-267. Smith EL. 1998. Performance characteristics of the emmetropization process in macaque monkeys. (ARVO Abstract). Invest Ophthalmol Vis Sci 39:S460.
Smith EL, Hung L-F, Harwerth RS. 1994. Effects of optically induced blur on the refractive status of young monkeys. Vision Res 34:293-301.
Smith EL, Hung L-F, Harwerth RS. 1999. Developmental visual system anomalies and the limits of emmetropization. Ophthalmic Physiol Opt (Forthcoming).
Smith EL, Maguire GW, Watson JT. 1980. Axial lengths and refractive errors in kittens reared with an optically induced anisometropia. Invest Ophthalmol Vis Sci 19:1250-1255.
Sommers D, Kaiser-Kupfer MI, Kupfer C. 1978. Increased axial length of the eye following neonatal lid suture as measured with A-scan ultrasonography. (Abstract). Invest Ophthalmol Vis Sci 17(Suppl):295.
Sorsby A, Benjamin B, Davey JB, Sheridan M, Tanner JM. 1957. Emmetropia and its aberrations. Med Res Counc Spec Rep Ser 293:1-69.
Sorsby A, Benjamin B, Sheridan M, Stone J, Leary GA. 1961. Refraction and its components during the growth of the eye from the age of three. Med Res Counc Spec Rep Set 301:1-67.
Sorsby A, Leary GA. 1970. A longitudinal study of refraction and its components during growth. London: Her Majesty's Stationery Office. p 1-41.
Sorsby A, Sheridan M, Leary G. 1962. Refraction and its components in twins. Med Res Counc Spec Rep Ser 303.
Sperduto RD, Seigel D, Roberts J, Rowland M. 1983. Prevalence of myopia in the United States. Arch Ophthalmol 101:405-407.
Steiger A. 1913. Die Entstehung der spharischen Refracktionen des menschlichen Auges. (In German). Berlin: Karger.
Stenstrom S. 1948. Investigation of the variation and the correlation of the optical elements of human eyes. Am J Optom Arch Am Acad Optom Monograph 58, parts I-VI.
Stone RA, Lin T, Laties AM. 1991. Muscarinic antagonist effects on experimental chick myopia. (Letter). Exp Eye Res 52:755-758.
Strömberg E. 1936. Uber refraktion und achsenl[inge des menschlichen auges. (In German). Acta Ophthalmol Suppi 14:281.
Tay MTH, Au Eong KG, Ng CY, Lim MK. 1992. Myopia and educational attainment in 421,116 young Singaporean males. Ann Acad Med Singapore 21:785-791.
Tigges M, Sugrue MF, Mallorga P, Stone RA, Laties AM, Femandes A, Iuvone PM. 1996. Effects of atropine (ATR) and pirenzepine (PIR) on ocular growth and muscarinic cholinergic receptors in young rhesus monkeys. (ARVO Abstract). Invest Ophthalmol Vis Sci 37:S326.
Tigges M, Tigges J, Fernandes A, Eggers HM, Gammon JA. 1990. Postnatal axial eye elongation in normal and visually deprived rhesus monkeys. Invest Ophthalmol Vis Sci 31:1035-1046,
Troilo D. 1992. Neonatal eye growth and emmetropisation--A literature review. Eye 6:154-160.
Troilo D, Gottlieb MD, Wallman J. 1987. Visual deprivation causes myopia in chicks with optic nerve section. Curr Eye Res 6:993-999.
Troilo D, Judge SJ. 1993. Ocular development and visual deprivation myopia in the common marmoset (Callithrix jacchus). Vision Res 33:1311-1324.
Troilo D, Li T, Glasser A, Howland HC. 1995. Differences in eye growth and the response to visual deprivation in different strains of chicken. Vision Res 35:1211-1216.
Troilo D, Nickla D, Wildsoet C. 1998. Changes in choroid thickness with induced refractive error in a primate The common marmoset. (ARVO Abstract). Invest Ophthalmol Vis Sci 39:S869.
Tron E. 1929. The optical elements of the refractive power of the eye. Graefes Arch Ophthalmol 122:1-33.
van Alphen GWHM. 1961. On emmetropia and ametropia. Ophthalmology Suppl 142:1-92.
Vo TD, Coleman DJ, Iwamoto T, Silverman RH, Rondeau MJ. 1987. An animal model for myopia: Increase in axial length of the rabbit eye by ultrasonically induced cataract. (Abstract). Invest Ophthalmol Vis Sci 28:217.
Wallman J. 1991. Retinal factors in myopia and emmetropization: Clues from research on chicks. In: Grosvenor T, Flom MC, editors. Refractive Anomalies. Boston MA: Butterworth-Heinemann. p 268-286.
Wallman J. 1995. How many myopias? In: Christen Y, Doly M, Droy-Lefaix MT, editors. Vision et Adaptation. Paris: Elsevier. p 1-8.
Wallman J, Adams JI. 1987. Developmental aspects of experimental myopia in chicks: Susceptibility, recovery and relation to emmetropization. Vision Res 27:1139-1163.
Wallman J, Adams JI, Trachtman J. 1981. The eyes of young chicks grow toward emmetropia. Invest Ophthalmol Vis Sci 20:557-561.
Wallman J, Gottlieb MD, Rajaram V, Fugate-Wentzek LA. 1987. Local retinal regions control local eye growth and myopia. Science 237:73-77.
Wallman J, Ledoux C, Friedman MB. 1978a. Simple devices for restricting the visual fields of birds. Behav Res Meth Instrumen 10:401-403.
Wallman J, Turkel J, Trachtman J. 1978b. Extreme myopia produced by modest change in early visual experience. Science 201:1249-1251.
Wallman J, Wildsoet C, Xu A, Gottlieb MD, Nickla DL, Marran L, Krebs W, Christensen AM. 1995. Moving the retina: Choroidal modulation of refractive state. Vision Res 35:37-50.
Walls G. 1942. The Vertebrate Eye and Its Adaptive Radiations. Bloomfield Hills NJ: The Cranbrook Press.
Ward B, Thompson FB. 1990. The Myopias: Pathogenesis and Pathophysiology, New York NY: Macmillan.
Warwick R. 1976. Eugene Wolffs Anatomy of the Eye and Orbit. 7th ed. Philadelphia PA: Saunders.
Weiss S, Schaeffel F. 1993. Diurnal growth rhythms in the chicken eye: Relation to myopia development and retinal dopamine levels. J Comp Physiol 172:263-270.
Wiesel TN, Raviola E. 1977. Myopia and eye enlargement after neonatal lid fusion in monkeys. Nature 266:66-68.
Wildsoet CF. 1997. Active emmetropization--Evidence for its existence and ramifications for clinical practice. Ophthal Physiol Opt 17:279-290.
Wildsoet CF, Pettigrew JD. 1988. Experimental myopia and anomalous eye growth patterns unaffected by optic nerve section in chickens: Evidence for local control of eye growth. Clin Vision Sci 3:99-107.
Wildsoet C, Wallman J. 1992. Optic nerve section affects ocular compensation for spectacle lenses. Invest Ophthalmol Vis Sci 33(Suppl):1053.
Wilson JR, Sherman SM. 1977. Differential effects of early monocular deprivation in monocular and binocular segments of cat striate cortex. J Neurophysiol 40:891-903.
Yi G, Marino M, Troilo D. 1996. Changes in choroid thickness and eye growth following experimentally-induced myopia. Optom Vis Sci 73:219.
Yinon U, Koslowe KC, Rassin MI. 1984. The optical effects of eyelid closure on the eyes of kittens reared in light and dark. Curt Eye Res 3:431-439.
Young FA. 1961. The effects of restricted visual space on the primate eye. Am J Ophthalmol 52:799-806.
Young FA. 1965. The effect of atropine on the development of myopia in monkeys. Am J Optom Arch Am Acad Optom 42:439-449.
Young FA. 1975. The development and control of myopia in human and subhuman primates. Contacto 19:16-31.
Young FA. 1977. The nature and control of myopia. J Am Optom Assoc 48:451-457.
Young TL, Ronan SM, Alvear AB, Wildenberg SC, Oetting WS, Atwood LD, Wilkin DJ, King RA. 1998a. A second locus for familial high myopia maps to chromosome 12q. Am J Hum Genet 63:1419-1424.
Young TL, Ronan SM, Drahozal LA, Wildenberg SC, Alvear AB, Oetting WS, Atwood LD, Wilkin DJ, King RA. 1998b. Evidence that a locus for familial high myopia maps to chromosome 18p. Am J Hum Genet 63:109-119.
Zadnik K. 1997. Myopia development in childhood. Optom Vis Sci 74:603-608.
Zadnik K, Mutti DO. 1995. How applicable are animal myopia models to human juvenile onset myopia? Vision Res 35:1283-1288.
Zadnik K, Satariano WA, Mutti DO, Sholtz RI, Adams AJ. 1994. The effect of parental history of myopia on children's eye size. JAMA 271:1323-1327.
Zylbermann R, Landau D, Berson D. 1993. The influence of study habits on myopia in Jewish teenagers. J Pediatr Ophthalmol Strabismus 30:319-322.
Figure 1 (A). Optical surfaces of the eye (cornea and crystalline lens ["lens"]) focus parallel light rays in a focal plane, illustrated by the solid lines converging at the dashed vertical line. In young humans and animals, the focal plane generally is closer to the front of the cornea (solid vertical line) than it is in adults. As the eye matures to the juvenile and adult stages, the focal plane moves posteriorly (away from the cornea) as illustrated by the dashed vertical lines in B and C. In most young eyes, the retina is anterior to the focal plane, so the eyes are hyperopic, as illustrated in A. As the vitreous chamber ("vitreous") enlarges during normal juvenile development, the retina moves toward the focal plane, producing emmetropia. Myopia (C) typically occurs when vitreous chamber elongation moves the retina past the focal plane so that light rays are in focus in front of the retina. Additional details on ocular anatomy may be found in textbooks such as Moses and Hart (1987).
Figure 2 Approximate change in eye size and refractive error distribution from infancy to 6 yr of age. Eyes were redrawn from Warwick (1976). Newborn refractive data, in 1 D steps, is plotted from Cook and Glasscock (1951). Refractive data from children at age 6 yr, in 0.5 D steps, is plotted from Kempf and others (1928). At 6 yr, nearly all children are slightly hyperopic.
Figure 3 Compensation for a minus (concave) lens. In an emmetropic eye (A), the focal plane for distant objects, without accommodation, is coincident with the retina as in Figure 1B. Placing a concave lens in front of one eye (B) displaces the focal plane posteriorly, assuming accommodation is set by the other, control, eye. In response to the lens, the vitreous chamber elongates (C), moving the retina toward the displaced focal plane. When the lens is removed (D), the focal plane reverts to its original location. The eye is myopic because the vitreous elongation has moved the retina so that it now is behind the focal plane. In A through D, the vertical dashed line demarcates the original axial location of the focal plane.
Figure 4 Schematic representation of normal development of the focal plane and axial length, deprivation-induced myopia, and "recovery'' in the tree shrew animal model. On the first day of visual experience (VE) the axial length (dark line) is much shorter (closer to the cornea) than the focal plane (thin solid line), so the eye is very hyperopic. During the first 2 wk of VE, the focal plane moves away from the cornea because of normal corneal flattening and lens growth. The axial length (mostly due to vitreous chamber enlargement) elongates rapidly, reducing the hyperopia, although the eye remains slightly hyperopic (focal plane longer than the axial length). A 12-day period of monocular deprivation starting at 21 days of VE (dark bar MD on abscissa) produces increased axial elongation in the deprived eye (dashed line), so that the axial length becomes greater than the location of the focal plane, producing a myopic eye at 33 days of VE. If the deprivation is then removed, the axial length remains constant, and the focal plane continues to mature, allowing "recovery" so that the eye no longer is myopic by 75 days of VE. Drawing based on data from Norton and McBrien (1992) and from Siegwart and Norton (1998).
Figure 5 Choice between accurate accommodation or axial elongation in response to near work. When an emmetropic eye (top) is presented with a nearby target, represented by E in A and in C, the image plane will fall behind the retina unless the eye accommodates. As shown by the arrows in A, accommodation moves the image plane (dashed vertical line) toward the retina. To the extent that this is accomplished, defocus is reduced, removing a possible stimulus for elongation so that (B) the eye remains the same length and is emmetropic. As illustrated in C, if the eye does not reduce the defocus of the image through accurate accommodation, the image plane remains behind the retina and may serve as a stimulus for elongation similar to the compensation that occurs with minus lens wear (Figure 3). Over time (D), continued underaccommodation might lead to an elongated myopic eye.
Figure 6 An emmetropization model. In 1, a visual stimulus that promotes axial elongation (defocus?) acts to reduce retinal responses (minus sign affecting 2). Retinal responses are communicated to the sclera (3), where responses related to (defocus?) cause an increase in scleral remodeling (4), indicated by the plus sign. Scleral remodeling causes increased creep rate in fibrous sclera and growth in the cartilaginous sclera (5), as indicated by the plus sign. High creep and/or growth in sclera increases the axial elongation rate (6), as indicated by the plus sign. In a hyperopic eye, an increased axial elongation rate reduces the visual stimulus (defocus?), as indicated by the minus sign impinging on "1." In the loop involving accommodation, the visual stimulus (defocus?) causes increased accommodation, which in turn reduces the visual stimulus (defocus?). Whether or not the eye elongates depends on the amount of "Defocus?" to which the eye is exposed and on whether or not accommodation sufficiently reduces the "Defocus?" as in Figure 5A. See text for additional information.
Copyright © 2008. National Academy of Sciences.
All rights reserved.
500 Fifth St. N.W., Washington, D.C. 20001.
Terms of Use and Privacy Statement