The visual pathway begins with photoreceptors in the retina, and it ends in the visual cortex of the occipital lobe. The rods and cones are the two types of photoreceptor cell. Rods have a special role in peripheral vision and vision under conditions of low illumination, and cones, which function in bright light, are responsible for central discriminative vision and for the detection of colors. The responses of the photoreceptors are transmitted by bipolar cells to ganglion cells within the retina, and axons of ganglion cells reach the lateral geniculate body of the thalamus through the optic nerve and optic tract. The final relay is from the lateral geniculate body to the visual cortex by way of the geniculocalcarine tract. In addition, some fibers from the retina terminate in various parts of the midbrain, in the pulvinar of the thalamus, and in the hypothalamus.
The following account of the visual system is restricted to a discussion of the neural elements and presupposes a general understanding of the structure of the eye and the optical mechanism that projects a focused, inverted image onto the retina.
Optic vesicles evaginate from the diencephalon at an early stage of embryonic development. Each optic vesicle “caves in” to form the optic cup, which consists of two layers and is connected to the developing brain by the optic stalk. The optic cup becomes the retina, and the optic stalk becomes the optic nerve. The cornea, lens, and other parts of the eye develop from nearby ectoderm and mesoderm. The retina contains neurons and neuroglial cells and resembles the gray matter of the brain. Similarly, the optic nerve is composed of white matter and is not a peripheral nerve.
Certain specialized regions serve as landmarks that need to be identified before the cellular components of the retina are described.
The cell layers of the retina, listed from the choroid to the vitreous body, are the pigment epithelium, rods and cones, bipolar cells, and ganglion cells (Fig. 20-1). Axons of ganglion cells run toward the posterior pole of the eye and enter the optic nerve at the optic papilla or optic disk. The papilla is slightly medial to the posterior pole, about 1.5 mm in diameter and is pale pink. The axons are heaped up as they converge at the margin of the optic papilla and then pass through the fibrous tunic (sclera) of the eyeball into the optic nerve. The optic papilla is a blind spot because it contains no photoreceptors.
The macula lutea, the central area of the retina in line with the visual axis, is a specialized region about 5 mm in diameter that abuts on the lateral edge of the optic papilla. The name macula lutea (yellow spot) is derived from the presence of a diffuse yellow pigment, which is apparent only when the retina is examined with red-free light. Consequently, the macula is not ordinarily visible with an ophthalmoscope, but its position is revealed by the absence of large blood vessels. The macula is specialized for acuity of vision. The fovea is a depression in the center of the macula, 1.5 mm in diameter and about 2.0 mm from the edge of the optic disk. Visual acuity is greatest at the fovea, the center of which (thefoveola) contains only cone receptors. The capillary network present elsewhere in the retina is absent from the center of the fovea. When the retina is viewed with an ophthalmoscope, the fovea appears darker red than the surrounding parts of the retina because the black melanin pigment in the choroid and the pigment epithelium is not screened by
capillary blood. (The visible fovea is commonly referred to as the “macula” in ophthalmoscopic descriptions of the retina.)
FIGURE 20-1 Schematic representation of the neurons of the retina. The numbers on the left are for the 10 histological layers. (Compare with Figure 20-5.) The outer and inner limiting membranes (layers 3 and 10) are formed from horizontally extending cytoplasmic processes of neuroglial cells (Müller cells), which are not otherwise included in the diagram.
The functional retina terminates anteriorly along an irregular border, the ora serrata. Forward of this line, the ciliary portion of the retina consists of a double layer of columnar epithelium, with the outer layer being pigmented.
The pigment epithelium is a single layer of cells that reinforces the light-absorbing property of the choroid in reducing the scattering of light within the eye (see Fig. 20-1). The basal part of each cell contains the nucleus and a few pigment granules. Processes extending from the free surface of the cell interdigitate with the outer photosensitive regions of rods and cones. The processes, which are filled with granules of melanin pigment, isolate individual photoreceptors and enhance visual acuity. A second function of the pigment epithelium is the removal, by phagocytosis, of membranous disks that are shed from the outer ends of the rods and cones.
The light-sensitive part of the photoreceptor is its outer part, adjacent to the pigment epithelium. The incident light, therefore, has to pass through almost all the retina before being detected. These layers do not present a significant barrier to light because the retina is transparent and is no more than 0.4 mm thick at any point.
The human retina contains about 130 million rods, which is 20 times the number of cones. Rods are absent from the central part of the fovea and become progressively more numerous from that point to the ora serrata. The distribution is such that rods are important for peripheral vision. There is a high density of cones along the edge of the ora serrata, possibly to provide for the recognition of objects entering the periphery of the visual field. Rods are more sensitive to dim light than cones, and the rodfree foveola is night blind. A faint point of light such as a dim star is best detected by looking slightly away from it. Each rod has three parts: the outer segment, inner segment, and rod fiber. The outer and inner segments are about 2 µm wide, and their combined lengths vary from 60 µm near the fovea to 40 µm at the periphery of the retina. The rod fiber is a slender filament that includes the nucleus in an expanded region and terminates as a synaptic terminal in contact with bipolar and association neurons.
The pigment epithelium is fixed to the choroid but is not as firmly attached to the inner layers of the retina. Detachment of the retina, which may result from a blow to the eye or occur spontaneously, consists of separation of the neural layers from the pigment epithelium. Fluid accumulates in the space thus created between the parts of the retina derived from the two layers of the optic cup. Retinal detachment can lead to blindness if left untreated.
In electron micrographs, most of the light-sensitive outer segment is seen to be occupied by about 700 double-layered membranous disks or flattened saccules (Figs. 20-2 and 20-3). These disks are continuously renewed from the inner segment of the rod and shed at the outer end of the outer segment. (Similar renewal occurs in the cones.) The disks contain the pigment rhodopsin (visual purple), which gives the retina a purplish-red color when removed from the eye and viewed under dim light. Rhodopsin consists of a protein, opsin, in loose chemical combination with retinal, a derivative of vitamin A. Absorption of a quantum of light changes the configuration of a rhodopsin molecule. A subsequent series of reactions results in hyperpolarization of the surface membrane of the inner segment and rod fiber, with consequent inhibition of the release of the neurotransmitter (believed to be glutamate), which is secreted continuously in darkness. It is a curious property of
photoreceptors that they are inhibited by their specific stimulus.
FIGURE 20-2 Ultrastructural components of rods and cones and their component parts. Named structures are described in the text. (Modified and reprinted with permission from Enoch JM, Tobey FL, eds. Springer Series in Optical Sciences, vol 23. Heidelberg: Springer-Verlag, 1981. Courtesy of Dr. B. Borwein.)
The inner segment of a rod contains the organelles found in all types of cells: mitochondria, neurofilaments, vesicles, and granular endoplasmic reticulum. A cilium joins the inner to the outer segment (see Fig. 20-3).
The cone photoreceptors are especially important because of their role in visual acuity and color vision. Cones, similar to rods, consist of outer and inner segments and a cone fiber.
FIGURE 20-3 Electron micrograph of a rod from the human retina, showing part of the outer segment and the adjoining region of the inner segment. (A) Membranous disks in outer segment; (B) mitochondria; (C) centrioles; (D) cilium (magnification ×30,000). (Courtesy of Dr. M. Hogan.)
The tapering outer segment of a cone consists principally of double-layered pigmentbearing disks (see Fig. 20-2). There are three types of cone, each containing a different pigment. Each cone pigment resembles rhodopsin in consisting of retinal combined with a protein. Three proteins (cone opsins) are recognized, each combining with retinal in such a way as to provide maximum absorption of red, green, or blue light. The three types of cone provide for trichromatic vision.
The inner segment of a cone is similar to the inner segment of a rod, but larger.
The proportion of cones to rods is high in the macular area, but it steadily decreases from the macula to the periphery of the retina. The foveola, at the center of the fovea, contains only cones. The cone fibers and bipolar cells diverge from the center of the fovea, producing a slight concavity and reducing any slight impediment to light passing through the retina. The absence of retinal capillaries at the center of the fovea
eliminates scattering of light by flowing blood. Figure 20-4 shows cone photoreceptors as they appear in a scanning electron micrograph.
There are several types of bipolar cells according to structure and physiological properties. These neurons are interposed between photoreceptor cells and ganglion cells (see Fig. 20-1). One bipolar cell is contacted by numerous rods (ranging from 10 near the macula to 100 at the periphery). Although there is some convergence of cones on bipolar cells in the peripheral parts of the retina, there is none at the fovea, at which point visual acuity is greatest. There, each cone fiber synapses with the dendrites of several bipolar cells.
Ganglion cells are rather large neurons with clumps of Nissl material, forming the last retinal link in the visual pathway (see Fig. 20-1). The bipolar cells contact both dendrites and somata of ganglion cells. The axons of ganglion cells, which form a layer adjacent to the vitreous body, converge on the optic papilla. There, bundles of axons and processes of neuroglial cells pass through foramina in the sclera, which at this point is called the lamina cribrosa. Behind the sclera, they constitute the optic nerve. The axons acquire myelin sheaths only after traversing the sclera, although in a few people, bundles of myelinated axons are present in the retina, where they appear ophthalmoscopically as white streaks.
FIGURE 20-4 Scanning electron micrograph of foveal cones in a monkey. There is a constriction of each photoreceptor at the base of its cilium (Ci). The outer limiting membrane (OLM) appears as a thin line. The inner cone fibers (white arrow) turn sharply back at an angle to the photoreceptors and their nuclei. This is a feature of foveal cones. (Reprinted with permission from Enoch JM, Tobey FL, eds. Springer Series in Optical Sciences, vol 23. Heidelberg: Springer-Verlag, 1981. Courtesy of Dr. B. Borwein.)
A minority of retinal ganglion cells respond directly to light. These neurons contain melanopsin, a visual pigment that absorbs in the blue part of the spectrum. Their axons terminate in the pretectal area of the midbrain and in the suprachiasmatic nucleus of the hypothalamus. There is evidence that the former connection mediates sustained pupillary constriction in bright light, a function that is retained in retinitis pigmentosa, a disease in which the rods and cones degenerate. (See Chapter 8 for circuitry of the pupillary light reflex.) The retinohypothalamic
projection has been shown in laboratory animals to regulate physiological responses to ambient illumination (see Chapter 11).
Color Vision Deficiency
The three types of cones allow the visual association cortex to identify a complete range of colors on the basis of signals received from the retina. If one of the cone opsins is not produced (or is produced but has a shifted absorption spectrum), vision is dichromatic, and certain different colors cannot be distinguished from one another. The most common color vision deficiency is an inability to separate certain red and green hues; this is caused by a defective gene encoding the cone opsin that absorbs in the middle part of the visible spectrum. This occurs in about 8% of males and 0.5% of females because the recessive abnormal gene is located on the X chromosome in most, but not all, cases. A yellow-blue dichromatism occurs much less frequently (1% of males, 0.01% of females). Monochromatic vision, which is caused by defective genes encoding either two or all three of the cone opsins, is the only true color blindness, with black-and-white vision mediated either by one type of cone or by rods only. Both types of monochromatic vision occur, but they are extremely rare.
Excitation and inhibition of ganglion cells depend on special properties of photoreceptors and bipolar cells. The presynaptic part of a photoreceptor leaks its transmitter continuously in darkness. The release of transmitter is suppressed by illumination. Thus, the activity of the receptor cell is suppressed by light. Bipolar cells do not conduct action potentials. Their neurites (and those of other retinal interneurons) are all called dendrites. Some bipolar cells respond to the transmitter from the photoreceptors with hyperpolarization of the cell membrane. Others respond to the same transmitter with partial depolarization. The quantity of transmitter released by the presynaptic neurites of a bipolar neuron varies with the magnitude of the partial depolarization of the cell.
The neurotransmitters in the retina are not yet certainly identified. Several candidate substances have been detected immunohistochemically in the human retina. These include glutamate, which is present in photoreceptors, many bipolar cells, and ganglion cells. Glutamate is known to be the excitatory transmitter at synapses in most other parts of the central nervous system (CNS).
Synaptic transmission in the retina is subject to modification by interneurons known as association neurons (see Fig. 20-1). Horizontal cells are located in the outer part of the zone occupied by the cell bodies of the bipolar cells. Their dendrites make contact with the synaptic terminals of the photoreceptors and with the dendrites of bipolar cells, which they inhibit. Amacrine cells are located in the inner part of the zone occupied by the cell bodies of the bipolar cells. The dendrites of an amacrine cell all emerge from the same side of the cell to ramify and then terminate in the synaptic complexes between bipolar and ganglion cells and on the interplexiform cells, which are described next. Amacrine cells contain many putative transmitters, and there are probably inhibitory and excitatory types. The interplexiform cells are interspersed among the cell bodies of the bipolar cells. They are postsynaptic to the amacrine cells and presynaptic to the horizontal and bipolar cells, thus providing a feedback loop through which neural information is passed back from the inner to the outer of the two layers of retinal synapses.
The retinal interneurons provide lateral inhibition, an arrangement that enhances central transmission from adjacent dark and illuminated regions of the retina. The signals sent to the brain are thus weighted in favor of the edges of images. (A simpler example of lateral inhibition is explained in Chapter 19.)
The innermost layers of the retina contain astrocytes similar to those in the gray matter of the brain. Large numbers of radial neuroglial
cells, called Müller cells, are also present. These cells extend from the innermost layer of the retina to the junction of the inner segments of rods and cones with the rod and cone fibers. They have lateral processes that intervene between the neuronal elements of the retina and provide support equivalent to that of astrocytes (see Chapter 2) elsewhere in the CNS.
In sections stained with hemalum and eosin (a commonly used dye combination that colors cell nuclei blue purple and everything else pink), the retina is seen to consist of 10 layers. These are shown in Figure 20-5, which can be compared with the diagram of cells that constitute the retina in Figure 20-1.
The retina receives nourishment from two sources. The central artery of the retina enters the eye through the optic disk, and its branches spread out over the inner surface of the retina. Thin branches penetrate the retina and form a capillary network that extends to the outer border of the inner nuclear layer. The capillary bed drains into retinal veins that converge on the optic papilla to form the central vein of the retina. The other source of blood is the capillary layer of the choroid. Soluble nutrients, oxygen, and metabolites of small molecular size diffuse from the choroid into the outer part of the retina. The layers containing the pigment epithelium, the photoreceptor, and the bipolar neurons are devoid of capillaries.
FIGURE 20-5 Section of human retina showing the layers seen in a section stained with hemalum and eosin. (Compare with Fig. 20-1.)
Pathway to the Visual Cortex
There is a point-to-point projection from the retina to the dorsal nucleus of the lateral geniculate body of the thalamus and from this nucleus to the primary visual cortex of the occipital lobe. There is, therefore, a spatial pattern of cortical excitation according to the retinal image of the visual field. Before discussing the components of the visual pathway, it will be useful to establish certain general rules concerning the projection from the retina to the cortex.
Retinal Artery Occlusion
A small embolus, detached from a thrombus in the left atrium or a plaque of atheroma in a carotid artery, can obstruct the central artery of the retina at the optic disc, where the vessel is narrowed as it passes through the sclera. The eye immediately becomes blind. An even smaller embolus can block a branch of the central artery, causing a small visual field defect in one eye. Microscopic larvae of Toxocara canisand T. cati (nematodes commonly present in the intestines of dogs and cats) can enter the circulation of young children who eat dirt contaminated with the feces of pet animals. Visual field defects are produced when the larvae lodge in branches of retinal arteries. The parasitic embolus evokes a slow inflammatory response, generating a granular lesion that is easily seen with an ophthalmoscope.
For the purpose of describing the retinal projection, each retina is divided into nasal and temporal halves by a vertical line that passes through the fovea. A horizontal line, also passing through the fovea, divides each half of the retina into upper and lower quadrants. The macular area for central vision is represented separately from the remainder of the retina. Figure 20-6 illustrates the following rules with respect to the central projection of retinal areas:
cortex of the right hemisphere. The converse holds true for the contralateral projection.
FIGURE 20-6 Topography of the projections from the retinas to the lateral geniculate body and primary visual cortex.
Visual defects that result from interruption of the pathway at any point from the retina to the visual cortex are described in terms of the visual field rather than the retina. The retinal image of an object in the visual field is inverted and reversed from right to left, just as an image on the film in a camera is inverted and reversed. The following rules, therefore, apply to the nuclear and cortical representation of regions of the visual field.
OPTIC NERVE, OPTIC CHIASMA, AND OPTIC TRACT
Each optic nerve contains about 1 million axons, all myelinated; this large number indicates the importance of human vision. The optic nerve is surrounded by extensions of the meninges (see also Chapter 26). The pia adheres to the nerve and is separated from the arachnoid by an extension of the subarachnoid space. The dura forms an outer sheath, and the meningeal extensions around the nerve fuse with the fibrous scleral coat of the eyeball. The central artery and central vein of the retina pierce the meningeal sheaths and are included in the anterior part of the optic nerve.
The partial crossing of optic nerve fibers in the optic chiasma is a requirement for binocular vision. Fibers from the nasal or medial half of each retina decussate in the chiasma and join uncrossed fibers from the temporal or lateral half of the retina to form the optic tract. Therefore, whereas impulses conducted to the right cerebral hemisphere by the right optic tract represent the left half of the field of vision, the right visual field is represented in the left hemisphere. Immediately after crossing in the chiasma, fibers from the nasal half of the retina loop forward for a short distance in the optic nerve. A lesion transecting the optic nerve close to the chiasma may, therefore, cause a temporal field defect in the opposite eye in addition to blindness in the eye whose optic nerve has been interrupted. The optic tract curves around the rostral end of the midbrain and ends in the lateral geniculate body of the thalamus.
Some of the fibers from the retina leave the optic chiasma and tract to proceed to sites other than the lateral geniculate body. These are described after a discussion of the pathway for conscious visual sensation.
LATERAL GENICULATE BODY, GENICULOCALCARINE TRACT, AND VISUAL CORTEX
The lateral geniculate body is a small swelling beneath the posterior projection of the pulvinar of the thalamus. The dorsal nucleus of the lateral geniculate body, in which the majority of the fibers of the optic tract terminate, consists of six layers of neurons. Within the general pattern
shown in Figure 20-6 and described previously, crossed fibers of the optic tract end in layers 1, 4, and 6; uncrossed fibers end in layers 2, 3, and 5.
An increase in pressure of cerebrospinal fluid around the optic nerve impedes the return of venous blood. Edema or swelling of the optic disk (papilledema) results. This is visible with an ophthalmoscope and is a valuable indication of an increase in intracranial pressure. Part of the swelling is caused by enlargement of the axons in the disk, attributed to partial obstruction of anterograde axonal transport (see Chapter 2) within the optic nerve fibers.
The geniculocalcarine tract originating in the lateral geniculate body first traverses the sublentiform and retrolentiform parts of the internal capsule. Its fibers then pass around the lateral ventricle, curving posteriorly toward the visual cortex (Fig. 20-7). Some of the geniculocalcarine fibers travel far forward over the temporal horn of the lateral ventricle. These fibers, which constitute the temporal or Meyer's loop of the geniculocalcarine tract, go to the visual cortex below the calcarine sulcus. It is evident from the retinal projection shown in Figure 20-6 that a temporal lobe lesion involving Meyer's loop causes a defect in the upper visual field on the side opposite the lesion. A lesion in the parietal lobe, on the other hand, may involve geniculocalcarine fibers that proceed to the visual cortex above the calcarine sulcus; the result is then a defect in the lower visual field on the side opposite the lesion.
FIGURE 20-7 Geniculocalcarine projections.
The primary visual cortex occupies the upper and lower lips of the calcarine sulcus on the medial surface of the cerebral hemisphere. The
area is much larger than suggested by cortical maps because of the depth of the calcarine sulcus. The primary visual cortex (area 17 of Brodmann) is marked by the line of Gennari (see Fig. 14-3) and is known alternatively as the striate area. There is a detailed point-to-point projection of the retina on the lateral geniculate body and on the visual cortex. The size of the retinal point is reduced to the diameter of a single cone for most acute vision in the central part of the fovea. Precise coordination of movements of the eyes ensures that the retinal patterns of activation correspond with one another, as required for binocular vision. The human visual association cortex is extensive, including the whole of the occipital lobe, the adjacent posterior part of the parietal lobe, the posterior part of the lateral surface of the temporal lobe, and much of the inferior surface of the temporal lobe. This cortex is involved in recognition of objects and perception of color, depth, motion, and other aspects of vision that increase in complexity with distance from the calcarine sulcus. In general, the occipital and posterior parietal cortex examines the positions of objects in the visual fields, and the temporal cortex is concerned with their identification. Color recognition takes place in the cortex of the medial part of the inferior surfaces of the occipital and temporal lobes. The inferolateral surface of the temporal lobe, discussed
also in Chapters 15 and 18, is involved in interpreting, remembering, and recalling formed images. The organization of the visual cortex into columns of cells is briefly reviewed inChapter 14. For disorders of the visual association cortex, see Chapter 15.
Visual Defects Caused by Interruption of the Pathway
Certain general rules governing defects in the visual field as a result of lesions in the visual pathway are indicated by examples in Figure 20-8. Example 1 is an obvious one: severe degenerative disease or injury involving an optic nerve results in blindness in the corresponding eye. Multiple sclerosis, in which central axons lose their myelin sheaths, can produce this effect. Example 2 refers to interruption of decussating fibers in the optic chiasma, which causes bitemporal hemianopia if the full thickness of the chiasma is interrupted. (This name implies blindness in the lateral halves of the visual field, but each lateral half is still visible, with the unaffected half of the contralateral retina.) The medial halves of the visual fields have normal binocular vision, but there is only monocular vision in the lateral halves. The lesion that most commonly affects the optic chiasma is a pituitary tumor pressing on it from below. This first interrupts fibers from the inferior nasal quadrants of both retinas. The visual defect begins as a scotoma in each upper temporal quadrant of the visual field and spreads throughout the temporal fields as the chiasma is increasingly affected. Pressure on the lateral edge of the optic chiasma (example 3) happens rarely, but it may occur when there is an aneurysm of the internal carotid artery in this location. The field defect, in the case of pressure on the right edge of the chiasma, is nasal hemianopia for the right eye. Interruption of the right optic tract (example 4) causes left homonymous hemianopia.
Example 5 is a large lesion that damages the geniculocalcarine tract or the primary visual cortex. An extensive right-sided lesion results in left homonymous hemianopia, except that central vision may remain intact (macular sparing). The cortex of the occipital lobe controls the involuntary eye movements that maintain fixation of the gaze on a target in the visual field. It is likely that a slight shifting of the patient's fixation or gaze during examination of the visual fields is responsible for the phenomenon known as macular sparing in patients with occipital cortical lesions. Destruction of only a part of the geniculocalcarine tract or the primary visual cortex causes field defects of lesser proportions than hemianopia. An example is provided by the upper quadrantic defect in the opposite visual field after interruption of fibers comprising Meyer's loop in the white matter of the temporal lobe (see Fig. 20-7).
It is important to remember that defects in the visual field can result from lesions of the eye as well as of the central pathways or cortex. For example, senile degeneration of the macula is a common condition that results in an area of blindness in the center of the field, often bilaterally. In chronic glaucoma, caused by increased intraocular pressure, atrophy of the peripheral parts of the retina occurs.
FIGURE 20-8 Visual field defects caused by lesions at five different sites in the visual pathway.
A small bundle of axons from the optic tract bypasses the lateral geniculate body and enters the superior brachium (see Figs. 6-2 and 7-15). These fibers, which participate in the afferent limbs of reflex arcs, go to the superior colliculus and to the pretectal area, which is a group of small nuclei immediately rostral to the superior colliculus.
The pupillary light reflex is tested in the routine neurological examination; the response consists of constriction of the pupil when light, as from a pen flashlight, is directed into the eye. Impulses from the retina stimulate neurons in the olivary pretectal nucleus, which is one of the nuclei of the pretectal area. Neurons in the pretectal area project to the Edinger-Westphal
nucleus of the oculomotor complex, which, in turn, sends fibers to the ciliary ganglion in the orbit. This ganglion innervates the sphincter pupillae muscle of the iris (see Chapter 8and Fig. 8-6). Both pupils constrict in response to light entering one eye because (1) each retina sends fibers into the optic tracts of both sides and (2) the pretectal area sends some fibers across the midline in the posterior commissure to the contralateral Edinger-Westphal nucleus.
Visual signals from the retina that reach the superior colliculus collaborate with input from the parietal and occipital cortex, frontal eye field, pallidum, and spinal cord, which are all sources of afferent fibers to the colliculus. The layered cytoarchitecture of the superior colliculus together with its diverse sources of afferent fibers indicate that considerable integrative activity occurs in the region. Efferent fibers go to the accessory oculomotor nuclei, paramedian pontine reticular formation, and pretectal area, and a few descend to the cervical segments of the spinal cord. This last pathway is known as the tectospinal tract.
The functions of the retinal afferents of the superior colliculus cannot be easily separated from the functions of the other afferents. The efferent fibers to the accessory oculomotor nuclei and to the paramedian pontine reticular formation are part of the pathway for control of both voluntary and involuntary movement of the eyes, as described in Chapter 8. An indirect connection to the Edinger-Westphal nucleus by way of the pretectal area controls the contractions of the ciliary and sphincter pupillae muscles in accommodation (see later). The small tectospinal tract is thought to influence movements of the head required for fixation of gaze.
When attention is directed to a near object, the accommodation-convergence reaction consists of three events: ocular convergence, pupillary constriction, and thickening of the lens. The reflex is tested by asking the subject to examine an object held about 30 cm in front of the eyes after looking into the distance and noting whether or not there is convergence and pupillary constriction. When attention is directed to a near object, the medial rectus muscles contract for convergence of the eyes. At the same time, contraction of the ciliary muscle allows the lens to thicken, increasing its refractive power, and pupillary constriction sharpens the image on the retina.
Argyll Robertson Pupil
Many patients with syphilis of the CNS (now a rare disease) have a loss of pupillary constriction in response to light but not to accommodation: the Argyll Robertson pupil or light-near dissociation. The lesion that typically causes dissociation of the responses is in the pretectal area, but cases have been described in which there was no abnormality in this part of the midbrain. The small size and slight irregularity of the Argyll Robertson pupil are probably caused by local disease of the iris.
For accommodation to near objects, instructions from the visual association cortex reach the midbrain through fibers traversing the superior brachium and terminating in the superior colliculus. The subsequent connections to the nuclei of those cranial nerves that supply the extraocular muscles and to the Edinger-Westphal nucleus have already been described.The frontal eye field, which is necessary for voluntary conjugate movements of the eyes, is not involved in convergence. The pathways for constriction of the pupil in the light and accommodation reflexes are known to be different because they may be dissociated by disease.
Dilation of the pupils occurs in response to severe pain or strong emotional states. The pathway is presumed to begin with fibers from the amygdala and hypothalamus, which influence the intermediolateral cell column of the spinal cord. The pathway continues to the superior cervical ganglion of the sympathetic trunk, and it is completed by postganglionic fibers in the carotid plexus to the dilator pupillae muscle in the iris (see Chapter 24). At the same time, the parasympathetic supply to the sphincter pupillae muscle is inhibited.
The human condition known as blindsight is seen occasionally in patients with destructive lesions of the geniculostriate pathways. Despite the complete lack of conscious vision, behavioral tests can detect the perception of movements or of changes in illumination.
Other Optic Connections
Experimental investigations in animals have revealed that the axons of retinal ganglion cells end in several parts of the brain in addition to the lateral geniculate body, pretectal area, and superior colliculus.
Some retinal ganglion cells have axons that enter the retinohypothalamic tract, a small population of fibers that leave the dorsal surface of the optic chiasma and synapse with neurons in the suprachiasmatic nucleus of the hypothalamus. The visual input synchronizes the intrinsic circadian rhythm of the firing pattern of the neurons of the suprachiasmatic nucleus with the changes in ambient illumination. This is responsible for the influence of different levels of illumination on the secretion of pituitary gonadotrophins and the pineal hormone melatonin (see Chapter 11) in response to longer days and shorter nights. Retinohypothalamic projections may also influence sleep (see Chapter 9).
The accessory optic tract consists of small fascicles that pass from the optic tract to various small nuclei in the tegmentum of the midbrain. These nuclei project, directly and through synaptic relays in the inferior olivary nuclei, to the flocculonodular lobe of the cerebellum. (The principal input to this part of the cerebellum is from the vestibular system.) These connections implicate the accessory optic tract in coordination of movements of the eyes and head. Other fibers of the accessory optic tract turn rostrally, to terminate in the anterior perforated substance; they may be involved in integrated responses to visual and olfactory stimuli.
Some optic axons end in thalamic nuclei other than the lateral geniculate body. The main area of termination of such fibers is the pulvinar, which projects to the cortex of the occipital and parietal lobes, which include much of the visual association cortex. The function of this alternative pathway from the retina to the cerebral cortex is not yet understood, but evidence from animal studies indicates that this pathway may permit some residue of conscious vision after destruction of the lateral geniculate body or the primary visual cortex.
Barton JJS, Simpson T, Kiriakopoulos E, et al. Functional MRI of lateral occipitotemporal cortex during pursuit and motion perception. Ann Neurol 1996;40:387-398.
Berson DM. Strange vision: ganglion cells as circadian photoreceptors. Trends Neurosci 2003;26:314-320.
Borwein B. The retinal receptor: a description. In: Enoch JM ed. Optics of Vertebrate Retinal Receptors. Berlin: Springer-Verlag, 1982.
Cowey A, Stoerig P. The neurobiology of blindsight. Trends Neurosci 1991;14:140-145.
Dente C, Gurwood A. The Argyll Robertson pupil. Optometry Today 1999:23-25.
Elkington AR, Inman C, Steart PV, et al. The structure of the lamina cribrosa of the human eye: an immunohistochemical and electron microscopical study. Eye 1990;4:42-57.
Grill-Spector K, Malach R. The human visual cortex. Ann Rev Neurosci 2003;27:649-677.
Hubel DH, Wiesel TN. Brain mechanisms of vision. Sci Am 1979;241:150-162.
Kawasaki A, Kardon AH. Intrinsically photosensitive retinal ganglion cells. J Neuro-Ophthamol 2007;27:195-204.
Kolb H, Fernandez E, Nelson R. Webvision. The organization of the retina and visual system. Available online at http://retina.umh.es/Webvision/. Accessed April 11, 2008.
McKeefry DJ, Zeki S. The position and topography of the human colour centre as revealed by functional magnetic resonance imaging. Brain 1997;120:2229-2242.
Mick G, Cooper H, Magnin M. Retinal projection to the olfactory tubercle and basal telencephalon in primates. J Comp Neurol 1993;327:205-219.
Milner AD, Goodale MA. The Visual Brain in Action, 2nd ed. Oxford: Oxford University Press, 2006.
Moore RY, Speh JC, Card JP. The retinohypothalamic tract originates from a distinct subset of retinal ganglion cells. J Comp Neurol 1995;352:351-366.
Sakai K, Watanabe E, Onodera Y, et al. Functional mapping of the human colour centre with echo-planar magnetic resonance imaging. Proc R Soc Lond Series B—Biological Sciences 1995;261:89-98.
Sowka JW, Gurwood AS, Kabat AG. Toxocariasis (Ocular larva migrans). In: Handbook of Ocular Disease Management. New York: Obson Publishing LLC, 2001.
Szel A, Rohlich P, Caffe AR, et al. Distribution of cone photoreceptors in the mammalian retina. Microsc Res Tech 1996;35:445-462.
Tootell RBH, Dale AM, Sereno MI, et al. New images from human visual cortex. Trends Neurosci 1996;19:481-489.
Williams RW. The human retina has a cone enriched rim. Vis Neurosci 1991;6:403-406.