The extraocular muscles: directing the eyes

The retina: initial processing of light patterns

The optic tract: visual transmission to the brain and further processing

The brain: high level processing, perception and cognition

Six muscles wrap around each eye. These 12 muscles work together to move both eyes in all possible directions. They are responsible for coordinated, accurate ocular skills. These oculomotor skills, when operating normally, allow eyes to fix on an object, near or far, and to hold fixation as the object moves (tracking or pursuing). The eyes also start and stop fixations (scan) as they swing from one object (like a word on a page) to the next. Eye muscles smoothly coordinate as they look from far to near or near to far. As they shift distances, the eyes turn inward together for near viewing, or turn outward at the same time for distance viewing.

Besides being a living biological organ, the eye is a sophisticated optical instrument. The front surface of the optical system is the cornea. The cornea has to be perfectly spherical and of a specific curvature for the optical system to function accurately. The cornea is a very powerful component. Behind the cornea is a fluid called the aqueous. It has a minimal affect on optical resolution. Behind the aqueous is the pupil, a hole in the iris through which the aqueous connects to a jelly like substance called the vitreous. Between the two fluids is an amazing living lens that can change it's shape to allow the optical system to automatically focus. Like the cornea, the lens is very powerful. The cornea, lens, aqueous, and vitreous (called the optical media) must be perfectly clear for the optical system to work normally. Light has to go through the cornea, through the aqueous, the lens and the vitreous unimpeded for correct visual information to be received at the retina. The aqueous and the vitreous are refreshed (replaced by fresh fluid) on a continual basis. There is a normal pressure that the fluids of the eye must maintain to keep the eye bathed in nutrients and cleansed of debris. When eye doctors examine vision to determine whether or not glasses should be prescribed, they measure the refractive status of the optical system.

The retina is part of the brain. It is the level at which sensation occurs. Visual sensation is that moment at which light energy begins to be processed by the brain. It is the first moment of awareness that a visual pattern is impinging on the system. Sensation is differentiated from higher level processing like perception. Sensation is what happens anatomically at the end organ, while perception involves the seemingly instantaneous recognition of patterns (an object, a face, movements, remembered experience). Perception occurs at high levels in the brain.

Early visual processing at the level of the retina is highly complex. The image on the retina contains initial information (a pattern) containing stable or fluctuating brightness contrast, color contrast, speed and direction of movement, field position, and form.

Under a microscope, the retina can be seen to have three layers. These are the choroid, where blood vessels nourish retinal tissue; the ganglion cell layer, where nerve fibers collect and move electrical impulses toward the optic nerve; and the retinal cell layer containing the photoreceptors.

There are two types of retinal receptors, the rods and the cones. Cones are clustered in an area of the retina called the macula (or fovea), an area about the size of a pin prick. Rods are thick in the peripheral parts of the retina. There is a gradual decrease of rods close to the macula, and a sharp decrease in cones as the distance from the macula increases.

Each photoreceptor has a different job. A principle job of the cones is to resolve detail in bright light. A routine job for the rod cells is to detect motion. Rods and cones have many sophisticated differences; they are, indeed, the beginnings of two separate information processing systems.

The human vision system is exceptionally complex. One way to look at this complexity is to understand that there are two separate visual information processing systems that work together at the level of the retina. These roughly correspond to the rod system and the cone system. It is important to understand the roles of these retinal cell processes.

The rod system evolved early in evolution. It is sensitive to motion and to sensation in low light. Laymen call it "side vision" and doctors call it "peripheral vision" (look straight ahead while wiggling a finger in the edge of the visual field to use your rod system). This system is responsible for detecting objects coming from the side. Interpreting visual signals from the rod system is called peripheral processing.

Peripheral processing (at higher brain levels) is an ambient system specializing in spatial orientation and peripheral monitoring. Information from peripheral processing is matched with motor centers in the brain to provide for steady movement, coordination, and balance. Another way to look at this is to think of the rod system as the "where is it" processing center. It is a navigational system, and it is the reason why most people who have low vision (ie. loss of central, or cone, vision) still have very good mobility skills. The "where" system is color blind and has poor resolving power for stationary patterns. It is very light sensitive and is used for seeing at night.

The cone system is delicate and of late evolutionary heritage. It operates only when light is bright (daytime, or when there is sufficient illumination). It locks onto objects, examines them, and tracks their movement in space. Called the central processing system (to differentiate it from peripheral processing), the cones are attention oriented (related to conscious mental activity), allowing a person to center (focus) on detail. The central processing system is called the "what is it" retina because it is the beginning of the perceptual system that identifies and categorizes. It sees stationary patterns and colors in high resolution.

Peripheral and central systems work together. For example, after the peripheral system detects objects coming from the side, a decision must be made to turn attention to the peripheral stimulus or ignore it. If the signal to attend to the peripheral is strong, then central fixation is moved in the new direction so that conscious study can take place.

From the rods and cones, signals are sent to the ganglion cell layer of the retina. Three types of cells have been identified at the ganglion level, each having a different processing task. These ganglion layer cells are the beginnings of three parallel visual pathways, called the X, Y, and W (or Z) pathways. The X pathway is for object recognition. It is part of the central foveal "what is it" system. It handles four types of color coding and two types of luminance selectivity. The Y pathway tells where changes occur in space, and handles visual attention and motion detection. It is related to the peripheral "where is it" vision system. The W pathway is poorly activated by visual stimuli and conducts slowly, but it projects to areas near the Y pathways. The X and Y processing streams of the retina are called the P and M pathways respectfully after they exit the retina and enter the optic tract (and through the brain).

It's easy to get confused with all the X, Y ,W, Z, P, M, visual processing streams swimming around. Understand it this way: it's all about side vision (navigation/mobility) or central vision (object/pattern recognition). The cone cells in the retina connect to the X cells in the ganglion region, which connect to cells in the optic tract called the P (pattern) pathway which connect to brain level processing cells. The job of this whole system of cell connections is to interpret patterns that allow for object recognition (the "what is it" system). The rods connect to Y cells in the ganglion region, which connect to optic tract M (mobility) pathway cells which connect with brain level centers responsible for navigation (the "where is it" system). Got it? Let's take a closer look at the ganglion cell layer, X cells first.

A constant response to stationary spots of light

Insensitivity to fast motion

Linear spatial summation

High pattern discrimination

They project specifically to a single layer of the dorsal lateral geniculate body

More dense near the fovea, color sensitive, slow conductance, sustained response

Connected to cone cells

The beginning of the P stream

Phasic response to stationary spots of light (ie. respond to the onset or offset of a stimulus)

Robust response to fast motion

No linear spatial summation

Spatial phase independent (stimulus placed anywhere in the receptive field elicits response)

Not good at pattern recognition (but do respond to patterns moving over the receptive field)

Project to: dorsal LGN, ventral LGN, pulvinar, superior colliculus, pretectum

Broad band response to color (not color sensitive)

The number of cells increase with eccentricity from the fovea

Connected to rod cells; and the beginning of the M stream

Information coming from the central and peripheral retinal systems flows from the retina, to the ganglion cell layer, to the optic nerve. The two optic nerves (one from each eye) go straight back from the eyes for a short distance before turning and joining in the center of the head behind the nose. The area where the nerves join is called the optic chiasm. From the optic chiasm nerve fibers split into two pathways again and travel a short way to the lateral geniculate bodies (one on each side) after which fibers fan out in a horizontal plane on their way to the occipital lobe of the brain. Important, additional processing of visual information occurs along the tracts on their way to visual centers in the occipital cortex of the brain.

Most of the visual processing, after the level of the retina, occurs at the lateral geniculate body(s) (LGB). It is important to realize that at all levels of the system there are feedback loops. Nerve fibers may go from the LGBs to the visual cortex, but nerve fibers also come from the occipital lobe to the LGBs (and/or the retina). These feedback loops come and go from all over the brain (cortex areas, brain stem areas, sensory nuclei, etc.). The LGN is the pathway for conscious vision. Other nerve fibers leaving the retina go to subcortical regions of the brain, where visual reflexes are controlled.

Each lateral geniculate body has six layers. The four top layers, with small cells, is called the parvicellular layer (P layer). The bottom two layers containing large cells is called the magnocellular layer (M-layer). These two divisions of the LGB are important because they carry nerve fibers that have specific information processing roles. Some fibers carry information about form, some about color, others about motion. The LGN is also subject to influence from a variety of ascending and descending non-visual neuron pathways; information coming and going from other sensory centers and from motor centers.

Nerve fibers passing through the parvicellular layer are called P pathway nerves, and those in the magnocellular layer are called M pathway nerves. The M and P pathways contain the signals that have been channeled from the cones to the X ganglion cells to the P pathway for central vision, and from the rods to the Y cells to the M pathway for peripheral processing. Keep the P and M pathways in mind because they are the key to understanding the high level processing of visual information discussed below (in the section on brain level processing).

Projections from the retina also go to subcortical centers in the superior colliculus, suprachiasmatic nucleus of the hypothalamus, to four pretectal nuclei, to three nuclei of the accessory optic system, and to the pulvinar complex. All these are reflex centers. Subcortical connections from the lateral geniculate body and the centers listed above also have feedback loops to the visual cortex, basal ganglia, pulvinar complex, thalamic reticular nucleus, parabigeminal nucleus, the pons, oculomotor nuclei, and the cerebellum. Some ganglion cells terminate in the suprachiasmatic nuclei of the hypothalamus (and therefore are implicated in circadian and ultradian (annual) rhythms). These connections help illustrate how the vision system is intricately interwoven with the entire human nervous system.

Another way to think about the many feedback neuron loops that connect vision to the entire brain, is to remember the evolution of the brain. The early, reptilian brain, the spinal column and the medulla, are concerned with automatic, reflexive functions. Vision is linked to this system for defensive purposes. Loud peripheral sounds, quickly approaching (looming) objects coming from the sides, sudden, unexpected touch, cause the body to go into emergency mode, to swing the head into position so that the eyes can examine the threat.

The midbrain (the limbic system, cerebellum, and thalamic regions) is responsible for emotional response, for coordinated movement, and for the integration of sensory input (linking the eyes, ears, skin sensation-touch and proprioception, and olfactory inputs). The non-verbal, unconscious visual communication system discussed in anothet chapter of this e-book is mediated at this level of the brain. So is the unconscious visual navigation system. The coordination of gait patterns, posture, and balance are also controlled by the midbrain. There are inner connections between the midbrain and the retina, the optic tracts, and the lower brain.

The most evolutionarily advanced part of the brain, the cortex, is responsible for conscious behavior; for high level thinking skills, future planning and projection, creativity, and concentration. It is a control center that can override (to some extent) the unconscious, automatic activities of the lower and mid brains. The cortex has feedback loops connecting to the lower and midbrain.

Anatomically, sensation occurs at the level of the end organ, while perception happens at pattern analysis levels of the brain. An upside down image (a pattern of reflected light) strikes the retina; sensation occurs. Visual processing takes place from the retina, through the optic tracts, to the occipital lobe. All the mind knows (at the level of the occipital lobe) is that a pattern is present. The next level of visual processing takes place in the parietal lobe and in the temporal lobe. The parietal areas are concerned with spatial perception ("He's facing to the right!"). The temporal processing areas identify and label visual objects ("It's a human face; Hey, Uncle Bill!").

Sensation is a pattern collecting process. Perception is the act of finding objects (figures) in the pattern (pulling the figure from the ground), including registering characteristic movement flows of objects. Perception is about finding constancies within the dynamic flux. At association levels of the brain, where cognition takes over from perception (temporal and parietal lobes), the figure is identified, named, and a determination is made regarding optical flow (whether or not there is movement in the pattern). At a higher processing level, within the frontal lobe cortex, further visual processing determines whether or not the object has emotional or utilitarian value (whether it has meaning for the person), and a plan is issued to respond (or not) to the visual message ("I remember that face! I love that guy. I think I'll give Uncle Bill a big hug!"). Also, the frontal lobes come into play when perception and association turn up a puzzling pattern (ie. when there is no perceptual match; "Hey! That's not Uncle Bill . . . so, who is this guy?"). The frontal lobe is (among many other things) a visual problem solving center. The frontal lobes also integrate information from other sensory areas, and store memories of multisensory neural inputs. Frontal eye fields control the movements of the extraocular muscles.

In the earlier discussion of the retina, the concept of two vision systems was introduced; the "what," and the "where" systems. This is a useful way to think about vision processing, and it corresponds to the roles of the parietal lobe (the "where" processing center) and the temporal lobe (the "what" processing center). Anatomically, scientists have discovered brain level neural tracts that support these two general concepts. Research indicates that there are four parallel visual processing centers in the brain; for: 1. motion processing; 2.color processing; 3. dynamic form processing; 4. form plus color processing. These brain centers are the destination points for the P and M pathways discussed above.

The P Pathway is the neural tract that contains information enabling human beings to identify people and objects; it is the neurological basis for the "What is it" vision system. The M pathway is the neurological basis for "where is it" vision.

High spatial frequency resolution (better at resolving small objects)

Low temporal frequency resolution (not good at detecting motion)

Important for color and form processing

Slower processing speed than M stream

Responses increase linearly with improved contrast

High contrast flicker resolution

Connects to the temporal lobe vision areas

When the P pathway leaves the parvicellular layer of the lateral geniculate body it travels to an area called V-1 in the visual cortex of the occipital lobe. The occipital lobe cortex (V-1) receives fibers from both the M and P pathways. P and M streams are specialized and therefore terminate in separate task centers in the brain (even though there are many feedback loops between the streams along specific areas of the pathways). P neurons go to cortical V-1 layers 4Cb and 4A, while M neurons go to V-1 layers 4Ca and 4B.

The occipital lobe cortex (V-1) contains a complete map of the contralateral visual hemifield, centered around foveal representation, and receives its main input from the LGN. V-1 is where the first binocular responses take place (signals from the two eyes are combined in a single receptive field). At this level are the first responses to movement, orientation, edges, speed, direction, binocular disparity, binocularity, changing disparity, length, spatial frequency, wavelength/hue, and luminance/ brightness. V-1 cells have two major cortical target zones as they leave layer V-1:

The M-stream pathway to the parietal lobe

The P stream pathway to the temporal lobe (middle temporal area; MTA)

After leaving V-1, P pathway fibers go to a region called V-2, formerly called the association cortex, but now called the pre-striate cortex. At the V-2 cortex, cells begin to process combinations of inputs: color plus orientation, direction plus orientation, direction plus color, sensitivity to illusory or anomalous contours (ie. the receptive fields at this level only respond to complex sets of inputs). From V-2, P pathway neurons travel to cortical area V-4, and from there to the inferotemporal cortex. The ventral/ lateral/ temporal P stream is most active for object vision. V-4 is responsive to color and to color plus form; it is in the fusiform gyrus area of the brain.

Low spatial frequency resolution (better at resolving large objects)

High temporal frequency resolution (better at detecting the time difference between the onset or offset of 2 stimuli, ie. good at motion detection)

Important for eye movement control, motion detection, visual spatial analysis (the gestalt), visual attention (a stabilizing system to allow the central system (P stream) to immobilize for focal concentration)

Flicker sensitivity is mediated by the M-pathway

Faster processing speed than P stream

Becomes saturated at high contrast

Can detect small luminance changes produced by fast moving objects

Connects to the parietal lobe vision areas

M pathway neurons travel from the magnocellular layer of the lateral geniculate nucleus, to the primary occipital cortex (V- 1), to the pre-striate cortex (V-2), to a region called V-3 (which is responsive to dynamic form -objects in motion- but not to color), to region V-5 (which is responsive to general motion- but not to color). Regions V-2, V-3, V-4, and V-5 are all in the pre-striate cortex. From areas V-3 and V-4, M pathway fibers go to the inferior parietal lobe. The dorsal/ superior/ parietal M stream is most active for spatial vision and for tasks requiring visual attention. (the M stream can be thought of as a stabilizing system that holds the image still, while central vision focuses in for analysis).

All this can be confusing, so let me summarize. Researchers have found neurological evidence for the existence of the "what is it?" and the "where is it" visual processing systems. The "What" system is about identifying and labeling the world. The "Where" system is about navigation. Furthermore, and quite importantly, research has established four brain centers, two for each pathway. The "what is it" pathway (P) passes through a brain nucleus that identifies colors, and another brain nucleus that identifies forms plus color. The "where is it" pathway (M) passes through a nuclear center that detects motion, and a second center that analyzes forms in motion. These centers are important for mobility specialists because there are impairments that can occur to the pathways and to the centers which severely impair the ability to identify landmarks or to navigate in space.

Research also suggests that some dyslexic individuals demonstrate problems in the V-5 area. MRI studies show that the V-5 area is quite active when people are reading. When dyslexic individuals read, however, the V-5 region is mostly quiet. The suggestion is that dyslexics have difficulty processing fast moving visual stimuli. To read quickly (efficiently, not in a labored way) requires rapid eye movement. It may also be that dyslexics have an overall difficultly handling fast processing speeds since auditory and tactile processing is also slow and labored.

Return to the index page Return to the top of this page.

Back to the Vision home page Ahead to vision impairments"