CHAPTER 2: VISION: Part I

www.pneuro.com

 

Eye and Brain

 

 

 “The eye...the window of the soul, is the principle means by which the central sense can most completely and abundantly appreciate the infinite works of nature; and the ear is the second, which acquires dignity by hearing of the things the eye has seen.

                                 -Leonardo da Vinci

 

     What’s most remarkable about vision is not the pure mechanics of seeing a visual image but explaining how we can see so much more than the basic equipment is designed for. The eye and brain would seem to have a limited specific capacity, yet we see so much, the gaps being filled in by mental processing, imagination and  understanding.   At the base of this process is the apparatus of visual perception, the start of a journey of the mind’s eye.

 

     The retina is something like film in a camera. The front of the eye does focus a good quality image on this screen.  There ends the analogy.  The retina is far more than the lifeless photographic film. It is sophisticated neural processing tissue, a complete neural waystation, first repository of vital visual data.  Retina is a part of the brain sent out to gaze upon the world.  The visual image transmitted from the retina has complex origins.  It does not form a single whole image like the film in a camera neatly transmitted to the brain.  Light images need to be deconstructed into electrical signals that the brain can use and process then somehow by a method poorly understood. We make mental images again from chunks of piecemeal data sent into the brain.  The retina thus transmutes light to electrical signals that the brain can use. Neurons work with electrical data, not visual images.  Vision is not different than touch, hearing, smell, and taste which all require transduction. In the brain all of these sensory modalities have a common language too and can be related to each other which means we can begin to gain a solid correlative singular conception that is part of consciousness.  

 

     For those who learn how to look at it, the retina can tell a lot about what is going on in the brain, as if the  brain sends out a part of itself to display to the world.  You have to look at it with an ophthalmoscope, first used by Helmholtz in 1850. The process is so routine, doctors aren’t  aware  they are actually examining the brain, which they are. In severe hypertension you see hemorrhages from bursted swollen arteries and a bulging optic nerve head telling of increased arterial pressure.  This reflects the same trouble in the brain.   Looking with the ophthalmoscope you may sometimes be lucky enough to see blood clots floating their way through retinal vessels. You may conclude that similar blood obstructing clots are traveling through other brain arteries. Dangerous clot material comes from the heart, great vessels or the carotid arteries in the neck and causes strokes by plugging up arteries in the brain. These tiny clots are thrown all the time in open heart surgery if you continually examine the retina, something rarely done. 

 

     When there is increased pressure in the closed cranial cavity, you see that on the retina also, by looking for swelling of the optic nerve, visible on the retina.  On a flash photograph you can see the retina in the form of a red response that is the bane of the photographer.  The pupil of the eye is dilated in a dark environment where the flash is often used and the retina reflects light quite nicely.

 

DEVELOPMENT:  

     Human embryology proves the retina is an extension of the brain.  The first sign of the retina in development is at about 22 days post-conception, when two slits or sulci appear in the front part of the developing brain.  Three or four days later an optic stalk already forms and an infolding to form an optic cup. The optic nerve carrying impulses from the retina to the brain is misnamed because it is really not a peripheral nerve but a tract, part of the central nervous system. The difference between a tract and a nerve, it will be recalled, is that a tract is a cable or bundle of axons that conduct impulses inside the central nervous system, actually part of the brain or spinal cord, while a nerve is in the peripheral nervous system. The blood supply is part of the brain’s blood, the myelin of the optic nerve is brain myelin, not nerve myelin. The retina is the only part of the brain outside the cranium. ¥

 

[i]

Figure 1: The retina starts as an outpouching of the brain. There is a space between layers as it merges with skin and other tissues that make up the rest of the eye.

 

     The next thing that happens in the embryo is that the front of the optic stalk folds in. (Figure 1).  This structure will eventually accept a lens whose origin is the ectoderm that forms skin. The cornea of the eye does most of the light refraction or bending focusing a beam of light on the retina, not the lens.  That is why you can correct your vision with a contact lens or by radial keratotomy, an operation on the cornea.  The lens in humans is there mostly for adjustment of the focus to accommodate various distances of objects.[ii]  The optical refracting equipment inside the eye has to be at least as good as the finest optical instruments developed by man.  Our cameras, microscopes and telescopes utilized advanced materials technology and the most advanced engineering and planning available.  But the eye’s refracting equipment is comprised only of biological materials, water, proteins and other chemicals. 

 

 

Figure 2: In a nearsighted or myopic eye, the eyeball is elongated and a distant image focuses in front of the retina.

 

 

 

     One of the most basic questions about the development of the eye is what determines whether or not a person will need corrective lenses.  From the biological perspective our eyes are mostly suited for distant outside use, scanning the horizon for prey or dangers and indeed in most of us a distant object will focus directly on the retina without any need for adjustment of the eye’s lens or accomodation.  Only close objects will require the lens to focus.  But a quarter of us are near sighted which means we are more suited for near vision, distant photographically infinitely distant objects, focusing in front of the retina instead of right on it.  This is because of a pathological elongation of our eyeball that seems to be a developmental defect. This difficulty, taken care of by corrective lenses, seems largely to be inherited.  Near-sightedness runs in families.  Certain experiments have shown that at least in experimental rhesus monkeys, near sightedness can be induced by environmental intervention, especially any action that interferes with adequate formation of an image[iii]. When images are blocked out or otherwise interfered with the eyeball develops in such a way as to be too long to focus the image properly.  A similar process might occur in children raised in a close environment looking mostly at televisions and books rather than the great outdoors. Control of elongation of the eyeball, near or far sightedness resides in the brain.  Early in development, brain and eye interact, to determine the length of the eyeball.   There is even some speculation that the secretion of a peptide called vasoactive intestinal polypeptide (VIP) may play a role in this abnormal development. Some day a chemical antagonist may be able to prevent nearsightedness.

 

     Soon, an inner and an outer layer of cells forms on the nascent retina, the inner layer closer to the lens of the eye, coming from brain neural tissue will house many kinds of neurons and become a visual processing center. The outer layer is the pigment epithelium.  Between these in the embryo is an intraretinal space.  As the eye develops this space will close as pigment epithelium is apposed to the inner retinal layer and outer and inner retinas fuse. But it is as an adult that this embryonic space becomes a problem. There is always a tendency for retinal disease to cause detachment of the retina that occurs just along the border between the inner and the outer retina.  (Diagrams) . A certain quirk in the retina also comes from the embryo.  The light receiving cells the rods and cones, as they are called because of their shape, lie at the hindmost part of the retina.  Cells that the rods and cones transmit to and which connect to the rest of the brain actually lie in front of the rods and cones. This means that light has to go through a lot of other cells and tissue on the retina in order to get to the light-sensitive cells. Because of how it is originally formed the retina is layered in such a way that light has to go through cells and processes in order just to reach the light gathering rods and cones.  This presents an interesting dilemma.  Either turn the entire structure around so that light reaches the sensitive cells first, or build the structure to be exquisitely thin and transparent.  The retina is designed according to the second option.

 

 

 

 [iv]

Figure 3: Microscopic picture of a retina. The pigment layer is on top. Just below are rods and cones, then other cell bodies and processes. Light must go through the other layers to get to light receptors.  Bottom is toward lens or front of eye.

RETINA:

     As seen in the microscope, the retina has 10 layers of cells and processes. And light has to go through almost to the hindmost layer to be detected. The outermost layer is made of melanin pigment, the same thing that makes our skin dark and makes the eye a dark light processing center.  In some animals who depend on night vision, this layer reflects light back into the retina instead of just absorbing it.  Albinos lack pigment and see poorly. Or, the pigment can proliferate too much and be out of control, as happens in forms of retinitis pigmentosa.  Tumors of melanocytes, the pigment making cells can also start in the eye.  The  light detecting cells, the rods and cones, are found just in front or internal i.e. towards the lens to the pigment layer. Rods detect the presence or absence of light and as such are merely black and white sensitive but they will react to trace quantities of light and as such are useful for low light and night vision. They lie primarily in the periphery of the retina.  An electrode in the rod cell can show the rod responds to the tiniest quantum of light, a single photon. Rhodopsin is the key to rod light sensitivity.  Rhodopsin is really a cooperative relationship between retinal that comes from vitamin A, and a long chain protein that sits on the cell membrane called opsin.  Retinal + Opsin = Rhodopsin. The trick is that  retinal deforms with a twist in a chemical bond, it becomes a cis isomer only  when light hits it, then reforms again when the light is taken off. The close association of retinal and opsin weakens when light hits it and reforms again in the dark.  Opsin is a large molecule a protein that essentially cradles the small retinal.  This sets off a chain of events that involves second messenger, other substances that change the charge of the inside of the rod or cone cell.  The light shines on the cell an the cell changes its electric charge.  This is the course of events that begins the odyssey of light sensitivity in the nervous system. Virtually the same molecules and the same scheme holds in all of the very different light sensitive organs in other animals that are very different eyes such as in the insects or crabs with compound eyes, other mammals, even primitive animals that have simple “eyes” of only a few cells.  These molecules and chemicals are highly conserved.  That means that they developed early on in evolution and then were inherited as more advanced animals came onto the scene.  In fact, some bacteria share a similar molecule designated bacteriorhodopsin raising the possibility that higher animals inherited their ability to make similar chemicals via bacterial infection (see similar discussion as relates to mitochondria and chloroplasts “Inside the Neuron”).  This light response may have evolved in parallel, or alternately, according to more conventional thinking, the bacteria and us may have descended from the same progenitor which carried its own type of rhodopsin molecule.

 

     By a strange twist, computer scientists are harnessing the bacteriorhodosin molecule which is much more stable than human rhodopsin and has interesting conformational properties as a wonderful memory storage device.  Simply the cis and trans isomeric forms of the molecule which change on exposure to light sources are exactly analogous to computer 0’s and 1’s or “on” and “off” states.  This means that light beams can be used as computer writing and reading devices over a biological computer chip impregnated with biorhodopsin molecules which change conformational properties in response to light. Today’s computers use nonorganic molecules which are spread over a surface two dimensionally, limiting storage and data access.  One may conceive of a three dimensional read and write optical memory device that can not only store much vaster amounts of data but which may access that data at unheard of speeds. Robert R. Birge of Syracuse University one of the pioneers estimates that while perhaps 100 million bits of information (0’s and 1’s) per square centimeter can be stored on conventional memory devices, one trillion bits per cubic centimeter can be placed on a biomolecular device[v].  When you think of it such machines are the union of the material and the biological, organic with inorganic matter which may at some stage, produce superior thinking (or at the very least data storage) machines, never dreamed of by science fiction writers.

 

     In each eye about 125 million Rods serve as light dark detectors and they don’t distinguish colors. We have only about 900,000 retinal ganglion cells that go to the brain proving that a lot of rods converge on each ganglion cell, about 140 and about 6 cones go to each ganglion cell.  The 5.5 million cones are concentrated in the center of the visual field, the fovea centralis which is a pit in the center of the macula (see below) in which blood vessels, cells, and their processes are pushed out of the way so that light can get to the part of the eye responsible for our most acute vision.   Our color sensitivity resides in the fovea. They do so because of slight changes in their pigment protein opsin the protein that holds the retinal molecule.  Minimal changes in the structure of this protein make it preferentially absorb light of red, green or blue that makes it absorb light of different colors. This has only occurred recently in our evolution as primates who forage for foods mostly during the day. Other mammals with a nocturnal existence have little need for color vision.  Because these pigment proteins are encoded on the X chromosome, the different forms of color blindness are x-linked and occur almost exclusively in men.  The cones are in the area of highest acuity, reading and what is tested with a wall eye chart.    The foveae perceive light differently than the rest of the retina. Though light has still to travel to the back of the retina the fovea is kept as clear as possible from all obstructions.  Axons and other structures are kept out of the way so that light falls on the cones directly.  Even so the foveae where the cones lie comprise a very small part of the retinal surface. The retina is as screen and what is projected on it is our visual world. Our whole field of view if we had receptors all the way around our head would  be a sphere through 360 degrees of vertical and horizontal angle. At any one time with our eyes stationary, we can only see perhaps 120-150 degrees of visual angle.  Our foveas,  where we see the clearest, subtends less than three  degrees of visual angle going according to its size.  You can calculate that one millimeter of retina subtends a visual angle of about 4 degrees.  We have an excellent internal picture of our whole 360 degree spherical visual world. We know this because we can negotiate around it, pursue run or be pursued almost without a thought despite the obvious liability of design built into our eyes and retina.  All this is done automatically with the constant motion and scanning by our eyes but it involves working with visual memory to form a working picture of our environment even though the part of it that we can actually see at any one time is minimal.

 

     The rod and cone light receptors generate a dark current.  They ooze glutamate that excites the neurons they connect to.  If light shines on them, there is a decrease in this dark current and they secrete less glutamate that stimulates the next cell in the chain, the bipolar neuron.  Light causes a change in the flow of ions into the rod or cone cell, mostly a decrease in the amount of positively charged Sodium that is allowed to flow in. Light has no meaning in the nervous system.  It has to be translated into the electrical language of the brain. The first step in this process is an alteration in the flow of charged ions in the receptor rods and cones.   That change is mechanical but is brought about by a sort of Rube Goldberg system of second chemical messengers and changes in enzymes.  The flow of currents is ultimately reduced in the receptor cell. 

 

     Neither the rods nor the cones are cells that fire.  They do not have that all or none action potential but continuously graded response.  As we’ve seen the “on” or “off”, “yes” or “no”, “1” or “0” response is much like a digital or computer type response.  But in the eye and this is also true of various other type of receptor cells elsewhere in the periphery,  for example in the ear, the initial  response is analog or graded and is only later translated in the central nervous system to a digital signal.  This allows for various modulations to occur before the data are interpreted centrally.  Some of this processing takes place right in the retina.  In  “higher” animals, primates and humans, some of the work done right in the retina tends to be done more and more by the brain and even by the cortex in a process known as cephalization,  Whereas in the rabbit or the frog this work is done in the retina itself.  One could speculate that in the frog,  most of the visual responses need to be quick and automatic, the tongue needs quickly to flick at the fly picked up visually, whereas in the human visual responses need to be more considered and slow.  When we see the central hookups of the eye, we will discover that only in some animals is there really advanced cerebrocortical or higher representation of visual input.

 

[vi]

Figure 4: Rods and cones are the initial light receptors, named for their shape. Rhodopsin pigments reside in the enormous surface area here seen at the top of the cells. There are loads of mitochondria too used to provide energy. Rods and cones are derived from ciliated cells and the tops of the cells are really modified cilia.

 

Figure 5: Retinal: Closely held by the protein opsin. The key to light sensitivity is a twisting of the molecule.

 

EYE:

     The eye is always compared to a camera to which it bears a superficial resemblance.  There is an iris that changes to accommodate more or less light and helps to respond to near and far objects as well. But the eye is living tissue so much more adaptable than any camera when you consider all of the conditions in which it works.  Cameras could not survive many of the conditions to which the human eye is subject such as extremes of heat, humidity, under water work etc. The eye is adapted not only to handle images of enormously varying distances or focal lengths but also light concentrations varying over 100 thousand fold from the dark night sky to midday in the summer time or bright glistening snow in the winter. But at any one instant, we can see clearly over only about three degrees with our retina and fovea. 

 

     What is striking to me as I pursue an activity such as jogging outside, avoiding cars and dead animals along the road, is that I have a complete picture of my environment in my head.  This is an illusion. At any moment, I can see only a tiny part of my 360 degree surroundings, yet I always think I have a panoramic view. How is this accomplished?

 

     At any moment we have an excellent picture of our environment and are capable of reacting to any change. Think of a quick lunge with of your opponent’s sword in fencing or an attempted tackle in football.  These are primarily visual responses.  Yet the eye can only see clearly about 3 degrees of visual angle.  One trick is that special circuits are built right into the retina involving primarily rods that respond right away to any motion.  The rods connect to bipolar cells which then go to ganglion cells and then through the optic nerve to the brain. The intervening cells perform additional processing functions in the retina.  Part of these processing deals with movement.   As a general rule in the nervous system intervening cells are there because further processing is required.  Vision transmission is from the rod or cone, to a bipolar cell, to a ganglion cell.  The ganglion cell will fire when it gets the specific input from the other cells that connect with it.  One favorite laboratory technique is to put an electrode into a ganglion cell, then see what it responds to.  Will it fire when light is shone on one tiny area of retina or to some other stimulus?  It is easy to find out. What you discover is that different ganglion cells can be made to fire for different reasons. Some indeed will only fire if there is movement within an area on the retina.   They won’t respond to a constant light.  Some will respond if a light is turned off or turned on.  The specific stimulus type that will make a ganglion cell fire or respond is the receptive field .  Mostly we are dealing with a specific area of retina that will make a specific ganglion cell fire, but it can also be conceived as a specific movement or color, whatever has been discovered to make that cell respond.  That is its receptive field.  In vision the scheme is that certain cells are wired to respond to certain stimuli.  As a matter of fact, this principle is universal in all reception, not only with vision.  A pressure receptor on the skin will only respond when you touch the skin within a specific area and won’t respond if you stick it with a pin etc., a cell in the nose may respond only to a certain odor.  This is its receptive field.

 

     In fact most ganglion cells respond best to one of two situations.  The rods and cones are wired up in such a way that the ganglion cells will most easily fire when there are concentric circles of light or dark.  Some fire best with light in a central circle of a small size surrounded by darkness.  These are “on” center cells. Others fire best when the center circle is dark, “off” center cells.  What they respond best to is contrast between borders of light and dark.  Ganglion cells will be more excited if there is a contrasting border and inhibited partially when light is surrounded by light or dark surrounded by dark. It is a common phenomenon in sensation that two adjacent receptive cells excited by the same stimulus, a large area of light for example inhibit one another.  On the other hand two adjacent cells that have contrasting stimuli, a light area surrounded by a dark area, for example might mutually excite each other.  This scheme accentuates contrast within the field of each receiving cell, ignoring sameness is lateral inhibition. While the retina accentuates contrasts and borders so do other receiving cells such as skin pressure receptors and even hearing functions by accentuating auditory stimuli of acute onset and offset. We notice a click or acute change in volume in preference to a constant droning sound. In our retina, ganglion cells aren’t apt to fire unless there is some difference or border between light and dark that can be perceived. This is true of colors as well. Long ago it was discovered that the eye will accentuate opposites which for colors are complements, red is best next to green and so forth something noticed in the nineteenth Century by Eugene Chevreul a French chemist. Our perception of colors is colored by the color that is next to it or near to it.  That is precisely the same process.  This allows for better acuity for borders and contours.

 

     We see clearly only in a small part of the retina which is structured for high acuity known as the macula lutea or yellow spot.  Something immaculate is spotless but although the macula looks like a spot it is actually the clearest part of the retina. Here’s why. In all other parts of the retina light has to go through all those other layers of cells and cell processes but not in the macula where they are cleared out of the way.  And the macula has primarily cones tuned for color vision.  Because there are so many fewer ganglion than light receptor cells Rod and cone inputs converge on ganglion cells but only about 6 on average for cones in the macular area but some 140 for rods that are not in the central area of the retina. Cones are tuned for color vision and work best in good light. When light hits the macular area we see the object clearly. Processes that affect the macula, such as Multiple Sclerosis and macular degeneration affect our ability to read, see colors and see clearly.  But we see clearly in only a small part of the retina.

 

Eyes and cameras are different in other ways.  With a camera we are trying to get a picture of a large field, to get everything into focus over a wide angle. As anyone who has used one of older (not automatic)  SLR cameras knows, you often mess up as you try to focus on many objects at different distances from the lens and with varying light and shadows. The eye doesn’t work like that.  It takes in with clarity only a very small part of the visual world, or visual angle and thus can focus clearly on only a very small area.  That’s why you have to move your eyes around so much when you read by a complex process designated as foveation, getting the more sensitive and precise fovea centralis of the eye that contains the macula, to focus on the area of interest.  The eye thereby eliminates the problem of trying to get the light and focal length right over a wide visual area and also there are fewer valuable clear vision cone cells and retinal cells used for precise vision.

[vii]

Figure 6: The eye focuses on only a tiny portion of the whole visual sphere. Yet we store a complete picture of our 360 degree visual environment.

 

 

 

     The retina’s design of central rods and peripheral cones has an analogy in the ear..  In each ear there are inner and outer hair cells sensitive to vibration.  There are only about 3500 inner hair cells and these are the most precisely tuned to specific frequencies of sound (pitch). Just the same way that the cones in the central eye are tuned to light frequencies (colors).  The inner hair cells synapse with many more secondary cells than the outer hair cells.  There is much more convergence of input of the outer hair cells of the ear just as there is a lot more convergence of rods than cones in the eye. Approximately 20,000 outer hair cells connect with perhaps 1000 secondary neuron cells.  The reason is that outer hair cells convey much less precise information.  The outer hair cells are situated in such a way as to influence the sensitivity of inner cells.  Both kinds of cells may respond to vibration so as to translate vibrations  into electrical polarities and the flow of positive currents (ions) into and out of the cell resonates with the actual frequency of sound stimulation.  Eye and ear are  quite analogous when you examine them.  The outer hair cells correspond well to the rods in the retina, the inner hair cells to the cones.  The cones are precise and a few of them influence more secondary cells. 

 

     If you see clearly only with the macula then what is the rest of the retinal surface for? Even when you are looking at something intently something else can grab your visual attention.  That something is likely to be a moving object or a flash of light, something that signifies change. Groups of rod cells in the peripheral retina are hard wired in a conspiratorial way to respond to movement and ganglion cells that only fire when an object moves within a certain area on the retina. It’s a fact that a frog will starve to death even if surrounded by tasty insects that fail to move. Steven Spielberg capitalized on this in his JURASSIC PARK where human children were prey for flesh eating reptilian dinosaurs, and though shaking with intense fear and wanting to run away were told not to move.  Perhaps the immobilizing effect of fear is adaptive.  As prey, we survive best by failing to contrast enough to draw attention away from nonliving nonmoving surroundings.  Outlining the scheme of ganglion cells for their response to movements and other changes is much of what has been done by researchers in vision over the last 15 or twenty years. 

 

     As the inner hair cells are precisely tuned for sound frequencies, cones in the eye are tuned to respond to colors.  Large groups of rod cells are grouped together to stimulate ganglion cells that rods connect to, to react or fire only when there is a certain visual stimulus.  These ganglion cells are thus also tuned to specific stimuli that can be called their receptive field.  What are some ganglion receptive fields?  One is motion in the periphery.  Something moves and this ends up by stimulating a certain ganglion cell.  Another may be a flash of light or change in light.  A light may suddenly go on or off. Another may be simply a concentration of continuous light.  These are the stimuli that will excite retinal ganglion cells.

 

     The strategy here seems to be that if a visual stimulus, say motion, can be made to excite a ganglion cell, there may be something there of enough interest to make the eyes turn to focus with the macula right on that point. The eye is constantly moving or roving pointing its macular vision into different areas of visual field as if probing constantly.  Over time, a complete visual picture accumulates so that even though we can actually see a tiny portion of our environment, still we have the illusion that we can see almost everything.  A total picture of our environment literally accumulates through constant visual probing and darting.  Since this picture must accumulate and is not present in real time, it requires not only memory but the intervention of some form of advanced cognition just in order to give the illusion of complete vision.  A camera takes a picture over a wide angle in real time, the eye and the brain work together in some kind of virtual reality.

 

     Mechanisms behind visual hallucinations hint at the mechanisms for putting visual images together.  A hallucination is the percept of something that isn’t there.  The visual sense is more exact than hearing and visual hallucinations point more strongly to a neurological or brain derangement as opposed to a psychiatric cause more the rule when something is heard.  Visual hallucinations illustrate how we ordinarily need to construct a complete image from data extracted piecemeal.  A lot of old folks and persons with Parkinson’s disease particularly, have visual hallucinations.  Some of them seem to construct an image out of an amorphous stimulus, say a spot of light in the dark.  They seem to have a problem magnified by adverse visual circumstances especially dusk and low light situations. During the day there may be no problem. In a random spot of light or looking at a chair or some other object, they may see a person or some intruder, even a long lost relative.  Another person with more faculties can easily convince them that the hallucination isn’t there.  These old folks are impaired. They may have troubles with the lens or the maculae of their eyes.  The Parkinson’s disease affects eye movements that are critical to bring specific objects into the realm of foveal clarity and what they do see is very unclear. 

 

     I sometimes go jogging in the dark even in the middle of the night.  At those times I have noticed that rarely I will interpret an object that I don’t see clearly a mailbox post or a spot of light on the pavement, as another better formed object.  I especially tend to see another person for some reason.  When I see it more clearly I realize there is nothing there. I have noticed that this process is worse under more adverse circumstances, in the middle of the night when I am sleepier and if I should jog without corrective lenses the problem really a hallucination, becomes particularly bad.  This is why ghosts and visions and religious experiences tend to occur at night.  I take it that old folks with their myriad visual and focusing impairments may have even a worse problem. This does not mean that they are necessarily demented.  The complete visual image requires close cooperation between eye and brain. 

Figure 7: In order to form a full image,  the brain needs to store a composite of many images through object scanning.[viii]

     The retina cannot give us a complete picture of an object as a film image does.  A complete image results from a cooperative effort between eye and brain.  Our vision is very unclear in the peripheral retina.  The peripheral retina is designed to help point the fovea at areas of importance often at movement or changing light. This implies that there is some kind of  smart pointing system  that tells the brain what is relevant so that eye motor systems can point the fovea at pertinent parts of our visual field. The brain will then use memory and other faculties to create an illusion of full vision. In order to get a complete picture of an object the eye will scan the whole scene since the foveal vision covers so little of it. The eye jumps or saccades at the rate of about 2 to 3 times a second to take in a whole picture spending about 10% of its viewing time in these jumps.

 

****************************************************************************************************

 

 

 

Aside: The Computation of Eye Movements

 

     This section gets rather technical and is not necessary to make the main points of this chapter. However, there is a relationship between our sense of balance provided by apparatus in the inner ear and vision.  The inner ear apparatus or labyrinth is specially designed to maintain posture and to keep your gaze fixed.  Eye movements have a rather complex and fascinating relationship with balance organs whose main purpose, it seems, is to maintain gaze in spite of any perturbations of posture.  For example, in playing basketball, your main goal is to keep your “eyes on the prize”, that basketball, even if you are fouled by another player and as you get involved with running, dribbling and other activity. 

www.pneuro.com

 

     As persons who have lost their vestibular function find out the hard way, this can devastate their vision. Think about it.  When you move, anything that you happen to be looking at will seem to move also. Unless you have a mechanism built in which makes adjustment for your movement and keeps the visual image still.  That is what the vestibular system does.  Certain antibiotics (aminoglycosides- streptomycin, Kanamycin, gentamycin) can literally wipe out the inner ear apparatus. When that happens the poor victim’s visual world is a blur, every time he moves. Walking down the street, he will not be able to see people’s faces and recognize them- their faces will be moving as he moves-a problem called oscillopsia.  If you are reading a book and you shake your head back and forth you will still be able to read the printed page, not so a person with oscillopsia.  The reason why the image remains still, despite your own movement, is that the vestibular system sends precise data to the visual brain about your head movement which helps adjust visual perception.  Under ordinary circumstances your brain always assumes your environment is still and that you’re moving which is true most of the time.  That is why if you happen to be sitting in a rail car and the train next to you starts to move, it always feels as if your own rail car is moving. In any event, nine tenths of the time the world remains still, and we move in it, we walk or shake our head and when we do, everything has to remain visually as if it were still, which it is. 

 

     We live in the computer age. Many of our most distinguished scientists accept nowadays, the notion that the brain is nothing more than a complex computer. Nowhere is the picture of brain as computer more applicable than in the vestibular system.

Vestibular function is a fitting model for brain computation for 2 reasons:

1. The inner ear brings about eye movements and positional change automatically,  without conscious intervention. We speak of vestibulo-ocular and vestibulo-spinal reflex mechanisms. We can predict that certain movements within the labyrinth cause specific eye movements and postural adjustments.

·  Figure 8: Three semicircular canals are orthogonal and deal with angular turns. The saccule with vertical linear movement, the utricle with horizontal movement.

     2. Eye movements and postural linear and turning movements may be described mathematically. So are events within the labyrinth, and over a similar Cartesian coordinate system. Position and movement are translatable to the three familiar Cartesian orthogonal (90 degrees from each other) axes: x,y and z, with x being horizontal, y vertical,  and z depth. As it happens the three semicircular canals, designated the superior, lateral and posterior,  are also practically orthogonal, and it is thus tempting to postulate that the canals communicate with the brain using some type of Cartesian code. The two ears are mirror images of each other, of course, and the semicircles of the canals of one ear complete full circles in the same planes as the other. Between the two ears the 6 semicircular canals make up 3 full circles in three orthogonal planes.

     The canals are round because they are designed to detect turning movements, angular acceleration. A complex of data integrated over the three planes will describe completely any angular acceleration.

 

     The canals, the membranous labyrinth, named for the twisting structure from Greek mythology, sit in an outer casing,  the bony labyrinth. These tiny (they are less than the size of a quarter of a dime) delicate transducers of movement, are surrounded by liquid perilymph which dampens random vibratory noise that will contaminate signal transduction. The semicircular canals are filled with a liquid of their own, the endolymph. Inertia holds the endolymph fluid in one place as the head starts to turn (accelerates). The fluid, moving in relation to the membrane, thus exerts a shear force on the tiny hair cellsm . These cells are sensitive to the movement of the fluid. Two kinds of hair cell cilia, stereocilia and kinocilia, are exposed to fluid shear forces.

·  Figure 9: A picture of the position of one labyrinth, showing the semicircular canals in three planes.[ix]

     Hair cells sensitive to pressure of endolymph are arrayed on a sensitive part of each semicircular canal, the crista or crest on the cupula. When the endolymph applies a shear force on the cilia or hairs of the cell in a specific direction (toward the kinocilium) the hair cell depolarizes, (becomes excited). The cells oriented in such a way as to be sensitive to the direction of these forces. Movement in one direction will depolarize or excite the cell, changing its membrane potential. It communicates this excitation to the next neuron whose firing rate is altered. This second neuron has a slow constant basal firing rate. Stimulate the hair cell and the second neuron will fire faster, inhibit it and it will slow down. That is how direction of the head turning in the labyrinth is converted to the language the brain can understand, firing rates. The position of each cell and the signal derived from it are codes for specific movements which are integrated in the vestibular nuclei and other brainstem structures.

·  Figure 10: The level of excitation of the hair cell is directionally determined as seen here. The state of excitation (depolarization) of the hair cell is communicated to a second cell whose firing rate is determined by movement direction and orientation.[x]

     Each sensitive region or cupula houses cells that fire all the time at a quantifiable level. The orientation of the semicircular canal of that cupula determines the direction of stimulation. Thus we have a magnitude and direction of force which in physics defines a vector. This vector can be graphically represented on Cartesian coordinate axes whose orientation at any time is determined by the head position at that instant. Under ideal circumstances each of 6 cupulas send the brain a information about a vector, direction and magnitude describing angular acceleration. The six vectors may be added to determine a single vector that influences eye movements. That there are eight vestibular nuclei, four on each side of the brainstem, attests to the complexity of computing movements on the basis of 6 angular acceleration vectors over time. By comparison, visual imaging is taken care of by just one pair of lateral geniculate bodies, though a great deal of complex visual processing is done both right in the retina and also at higher cortical levels.

 

     Suppose the head turns suddenly to the left. Then endolymph fluid will stay still and move against the cupulas, sensitive structures in the right ear. The exact angle of torsion is written into the array of three orthogonal semicircular canals and is computed in the brain. Soon the head turn to the left will stop and the endolymph will then move in the opposite direction, that is, away from the cupulas of the right ear and toward those of left.

When the left ear is stimulated the eyes will be made to move to the right. A stimulated canal generally pushes the eyes in the opposite direction. Thus there is constant push-pull relationship between the labyrinth and the eyes. The left ear canal pushes the eyes to the right, the right canal to the left. There is a balance between these two countervailing forces in our eyes. One canal, the lateral canal, lies practically horizontal, and is most important. The lateral canal is actually horizontal as you lay supine with your head up 30 degrees. Stimulate it in this position and the eyes will move to away in a straight line. You can do this by moving the head or putting cold or warm water in the ear. You can observe the resulting eye movements and quantitate them with an electronystagmograph. Tilt the head up or down from this specific 30 degree position and the horizontal movements will become rotatory due to the combination of vector forces determining the movement of the eyes.

Observe the direction of the arrows on the Cartesian axes represented in figure 6. That axis will be serve as the same axis of nystagmoid eye movement for the eyeball.  Stimulate one semicircular canal and the eyeball will rotate directly through that semicircular canal’s axis line.  The problem in clinical assessment is that head positions which determine the orientation of these axes, constantly change and so do the positions of the eyes.  However, it is possible to assess the semicircular canals individually by keeping the head at a certain tilt, for example 30 degrees forward from the supine position to isolate the lateral semicircular canal and at the same time, control for eye positions.

 

     A second component of eye movement comes from the upper brain and this is corrective. The cerebral cortex, noting deviation of the eyes, will make a rapid corrective movement. Looking at nystagmus there is a slow component induced by vestibular forces, and a rapid component in the opposite direction courtesy of the cortex.

Disease processes usually reduce the influence of one or another canal or labyrinth. (Some diseases cause stimulation. In Ménière's disease the diseased ear is stimulated  by the accumulation of endolymph fluid.)  As mentioned, the major purpose of all of this is to maintain visual fixation (“eyes on the prize”). One outcome is that absent the influence of the upper cortical brain, the eyes will move much like a doll whose head is turned, a so-called “doll's head response”.   The eyes will appear to be fixated on a spot despite head movement and will move in a direction opposite to head movement. This is the vestibulo-ocular reflex.

 

     The Utricle and saccule, in the meantime, are primarily concerned with maintenance of posture. Recall that these structures have calciferous otoliths, tiny rocks,  inside the ear that are very heavy (specific gravity 2.5) helping to detect the direction of gravity and fall on hair cells in the maculae (sensitive regions) of these small organs. These organs perhaps are descended from ancient statocysts that also function with the aid of heavy particles in much lower animals.  Hair cells are the sensitive end organs for the detection of gravity.   Hairs (cilia) are deformed and signals sent to the brain. The utricle  affects the level of the eyes.  In certain patients whose surgery has damaged one utricle,  the eyes are on two different levels. Semicircular canal impulses then go more rostrally (higher) to control eye movements and their nerve is mostly the superior vestibular division which goes to the more rostral vestibular nuclei in the brainstem (there are four on each side.) while the utricle and saccule send signals primarily via the inferior vestibular nerve division and to the spinal cord.

 

     One influence of computer technology is how it has influenced our conception of physiology. Brain systems including balance mechanisms have come to be viewed as individual automatous but interacting modules, each computing some function and affecting neighboring processes to perfect sensory and motor function. In the brain, modules are typically bundles of neurons, either nuclei or anatomical structures. The cerebellum is a primary example of an automatous computational device that perfects movements. Cerebellar inputs are multiple simultaneous and complex deriving from such diverse sensory sources as proprioceptors, vestibular apparatus, visual and motor systems. The cerebellum connects through cerebellar nuclei to other structures primarily over white matter tracts, cables, within the central nervous system. The cerebellum has to process diverse data from many different systems all at once, in other words, in parallel. In most modern computers information flows through a single microprocessor sequentially though with many operations per second. The cerebellum and other neuronal systems process large numbers of inputs in parallel, and so there is a fundamental difference in style of processing. The brain is considered a massively parallel device, the computer as sequentially processing device. In recent years, computer scientists understanding this basic difference in data handling and having great admiration for the accomplishments of biological systems (biocomputation of gait and balance is great example), have sought to emulate processing in the brain utilizing parallel arrays of microprocessors in their instruments. We in biology on the other hand, have benefited from computer science by acquiring a modular view of brain processes.

 

     The cerebellum receives information on posture and movement form the vestibular apparatus, and processes this in context with other data it receives from the spinal cord and visual system. Information from the utricle and saccule affect mostly the midline cerebellum (flocculus) and the cerebellum, in turn, uses this to compute instructions on balance and stance. This data is transmitted to the motor output areas in spinal cord via cerebello-spinal connections and there is a reciprocal feedback relation with the cerebellum via large spino-cerebellar tracts that relay information from stretch receptors in muscle. The vestibular apparatus also connects widely with higher brain centers affecting gaze mechanisms through the medial longitudinal fasciculus which also connects with the cord and brainstem to affect balance directly. Vestibulospinal tracts make a direct connection to the spinal cord. Motor systems also  modulate the basic function of the vestibular apparatus itself, through up and down regulation of sensory cells, making these cells more or less excitable or sensitive to stimulation. Data from the vestibular and all other sensory systems help determine general arousal levels though the reticular system responsible for general arousal in higher animals. Sensory information finally reaches the conscious cortical levels after processing in the thalamus. The cortex also helps initiate movements directly and with the help of the cerebellum and basal ganglia which perfect the rather crude motor plans of the motor cortex. Hence we have autonomous but intensely interactive modules that comprise motor functions. Underlying all of this, as we have seen are crude hard-wired reflex mechanisms built right into the spinal cord.

 

     We may fully expect an intense cross-fertilization of neurophysiology and computer science in the coming decades. Silicon (computer) devices and Carbon based (biological) modules will certainly interact more closely. Not only will communication between brain and computer be easier, but biological and silicon modules will likely mix. There is nothing to prevent Silicon devices being implanted in the brain and biological devices from being embedded in computers.

 

·  Figure 11: A Modular view of vestibular influence.

To Part II

www.pneuro.com